Knowledge	B-KEY
management	I-KEY
The	O
article	O
defines	O
knowledge	B-KEY
management	I-KEY
,	O
discusses	O
its	O
role	O
,	O
and	O
describes	O
its	O
functions	O
.	O
It	O
also	O
explains	O
the	O
principles	O
of	O
knowledge	B-KEY
management	I-KEY
,	O
enumerates	O
the	O
strategies	O
involved	O
in	O
knowledge	B-KEY
management	I-KEY
,	O
and	O
traces	O
its	O
history	O
in	O
brief	O
.	O
The	O
focus	O
is	O
on	O
its	O
interdisciplinary	O
nature	O
.	O
The	O
steps	O
involved	O
in	O
knowledge	B-KEY
management	I-KEY
i.e.	O
identifying	O
,	O
collecting	O
and	O
capturing	O
,	O
selecting	O
,	O
organizing	O
and	O
storing	O
,	O
sharing	O
,	O
applying	O
,	O
and	O
creating	O
,	O
are	O
explained	O
.	O
The	O
pattern	O
of	O
knowledge	B-KEY
management	I-KEY
initiatives	O
is	O
also	O
considered	O

High-level	O
language	O
support	O
for	O
user-defined	O
reductions	B-KEY
The	O
optimized	O
handling	O
of	O
reductions	B-KEY
on	O
parallel	B-KEY
supercomputers	I-KEY
or	O
clusters	B-KEY
of	I-KEY
workstations	I-KEY
is	O
critical	O
to	O
high	O
performance	O
because	O
reductions	B-KEY
are	O
common	O
in	O
scientific	O
codes	O
and	O
a	O
potential	O
source	O
of	O
bottlenecks	O
.	O
Yet	O
in	O
many	O
high-level	O
languages	O
,	O
a	O
mechanism	O
for	O
writing	O
efficient	O
reductions	B-KEY
remains	O
surprisingly	O
absent	O
.	O
Further	O
,	O
when	O
such	O
mechanisms	O
do	O
exist	O
,	O
they	O
often	O
do	O
not	O
provide	O
the	O
flexibility	O
a	O
programmer	O
needs	O
to	O
achieve	O
a	O
desirable	O
level	O
of	O
performance	O
.	O
In	O
this	O
paper	O
,	O
we	O
present	O
a	O
new	O
language	B-KEY
construct	I-KEY
for	O
arbitrary	O
reductions	B-KEY
that	O
lets	O
a	O
programmer	O
achieve	O
a	O
level	O
of	O
performance	O
equal	O
to	O
that	O
achievable	O
with	O
the	O
highly	O
flexible	O
,	O
but	O
low-level	O
combination	O
of	O
Fortran	O
and	O
MPI	O
.	O
We	O
have	O
implemented	O
this	O
construct	O
in	O
the	O
ZPL	O
language	O
and	O
evaluate	O
it	O
in	O
the	O
context	O
of	O
the	O
initialization	O
of	O
the	O
NAS	O
MG	O
benchmark	O
.	O
We	O
show	O
a	O
45	O
times	O
speedup	O
over	O
the	O
same	O
code	O
written	O
in	O
ZPL	O
without	O
this	O
construct	O
.	O
In	O
addition	O
,	O
performance	O
on	O
a	O
large	O
number	O
of	O
processors	O
surpasses	O
that	O
achieved	O
in	O
the	O
NAS	O
implementation	O
showing	O
that	O
our	O
mechanism	O
provides	O
programmers	O
with	O
the	O
needed	O
flexibility	O

Accelerating	O
filtering	B-KEY
techniques	I-KEY
for	O
numeric	B-KEY
CSPs	I-KEY
Search	B-KEY
algorithms	I-KEY
for	O
solving	O
Numeric	B-KEY
CSPs	I-KEY
-LRB-	O
Constraint	B-KEY
Satisfaction	I-KEY
Problems	I-KEY
-RRB-	O
make	O
an	O
extensive	O
use	O
of	O
filtering	B-KEY
techniques	I-KEY
.	O
In	O
this	O
paper	O
we	O
show	O
how	O
those	O
filtering	B-KEY
techniques	I-KEY
can	O
be	O
accelerated	O
by	O
discovering	O
and	O
exploiting	O
some	O
regularities	O
during	O
the	O
filtering	O
process	O
.	O
Two	O
kinds	O
of	O
regularities	O
are	O
discussed	O
,	O
cyclic	O
phenomena	O
in	O
the	O
propagation	B-KEY
queue	O
and	O
numeric	O
regularities	O
of	O
the	O
domains	O
of	O
the	O
variables	O
.	O
We	O
also	O
present	O
in	O
this	O
paper	O
an	O
attempt	O
to	O
unify	O
numeric	B-KEY
CSPs	I-KEY
solving	O
methods	O
from	O
two	O
distinct	O
communities	O
,	O
that	O
of	O
CSP	O
in	O
artificial	B-KEY
intelligence	I-KEY
,	O
and	O
that	O
of	O
interval	B-KEY
analysis	I-KEY

Non-optimal	O
universal	O
quantum	O
deleting	O
machine	O
We	O
verify	O
the	O
non-existence	O
of	O
some	O
standard	O
universal	O
quantum	O
deleting	O
machine	O
.	O
Then	O
a	O
non-optimal	O
universal	O
quantum	O
deleting	O
machine	O
is	O
constructed	O
and	O
we	O
emphasize	O
the	O
difficulty	O
for	O
improving	O
its	O
fidelity	B-KEY
.	O
In	O
a	O
way	O
,	O
our	O
results	O
complement	O
the	O
universal	B-KEY
quantum	I-KEY
cloning	I-KEY
machine	I-KEY
established	O
by	O
Buzek	O
and	O
Hillery	O
-LRB-	O
1996	O
-RRB-	O
,	O
and	O
manifest	O
some	O
of	O
their	O
distinctions	O

An	O
effective	O
feedback	B-KEY
control	I-KEY
mechanism	O
for	O
DiffServ	B-KEY
architecture	O
As	O
a	O
scalable	O
QoS	B-KEY
-LRB-	O
Quality	O
of	O
Service	O
-RRB-	O
architecture	O
,	O
Diffserv	B-KEY
-LRB-	O
Differentiated	O
Service	O
-RRB-	O
mainly	O
consists	O
of	O
two	O
components	O
:	O
traffic	B-KEY
conditioning	I-KEY
at	O
the	O
edge	O
of	O
the	O
Diffserv	B-KEY
domain	O
and	O
simple	O
packet	B-KEY
forwarding	I-KEY
inside	O
the	O
DiffServ	B-KEY
domain	O
.	O
DiffServ	B-KEY
has	O
many	O
advantages	O
such	O
as	O
flexibility	O
,	O
scalability	O
and	O
simplicity	O
.	O
But	O
when	O
providing	O
AF	B-KEY
-LRB-	O
Assured	O
Forwarding	O
-RRB-	O
services	O
,	O
DiffServ	B-KEY
has	O
some	O
problems	O
such	O
as	O
unfairness	O
among	O
aggregated	O
flows	O
or	O
among	O
micro-flows	O
belonging	O
to	O
an	O
aggregated	O
flow	O
.	O
In	O
this	O
paper	O
,	O
a	O
feedback	B-KEY
mechanism	I-KEY
for	O
AF	B-KEY
aggregated	O
flows	O
is	O
proposed	O
to	O
solve	O
this	O
problem	O
.	O
Simulation	O
results	O
show	O
that	O
this	O
mechanism	O
does	O
improve	O
the	O
performance	O
of	O
DiffServ	B-KEY
.	O
First	O
,	O
it	O
can	O
improve	O
the	O
fairness	B-KEY
among	O
aggregated	O
flows	O
and	O
make	O
DiffServ	B-KEY
more	O
friendly	O
toward	O
TCP	B-KEY
-LRB-	O
Transmission	O
Control	O
Protocol	O
-RRB-	O
flows	O
.	O
Second	O
,	O
it	O
can	O
decrease	O
the	O
buffer	O
requirements	O
at	O
the	O
congested	O
router	O
and	O
thus	O
obtain	O
lower	O
delay	O
and	O
packet	O
loss	O
rate	O
.	O
Third	O
,	O
it	O
also	O
keeps	O
almost	O
the	O
same	O
link	O
utility	O
as	O
in	O
normal	O
DiffServ	B-KEY
.	O
Finally	O
,	O
it	O
is	O
simple	O
and	O
easy	O
to	O
be	O
implemented	O

Interaction	O
and	O
presence	B-KEY
in	O
the	O
clinical	O
relationship	O
:	O
virtual	B-KEY
reality	I-KEY
-LRB-	O
VR	O
-RRB-	O
as	O
communicative	O
medium	O
between	O
patient	O
and	O
therapist	O
The	O
great	O
potential	O
offered	O
by	O
virtual	B-KEY
reality	I-KEY
-LRB-	O
VR	O
-RRB-	O
to	O
clinical	O
psychologists	O
derives	O
prevalently	O
from	O
the	O
central	O
role	O
,	O
in	O
psychotherapy	B-KEY
,	O
occupied	O
by	O
the	O
imagination	B-KEY
and	O
by	O
memory	B-KEY
.	O
These	O
two	O
elements	O
,	O
which	O
are	O
fundamental	O
in	O
our	O
life	O
,	O
present	O
absolute	O
and	O
relative	O
limits	O
to	O
the	O
individual	O
potential	O
.	O
Using	O
VR	O
as	O
an	O
advanced	O
imaginal	B-KEY
system	O
,	O
an	O
experience	O
that	O
is	O
able	O
to	O
reduce	O
the	O
gap	O
existing	O
between	O
imagination	B-KEY
and	O
reality	O
,	O
it	O
is	O
possible	O
to	O
transcend	O
these	O
limits	O
.	O
In	O
this	O
sense	O
,	O
VR	O
can	O
improve	O
the	O
efficacy	O
of	O
a	O
psychological	B-KEY
therapy	I-KEY
for	O
its	O
capability	O
of	O
reducing	O
the	O
distinction	O
between	O
the	O
computer	O
's	O
reality	O
and	O
the	O
conventional	O
reality	O
.	O
Two	O
are	O
the	O
core	O
characteristics	O
of	O
this	O
synthetic	O
imaginal	B-KEY
experience	O
:	O
the	O
perceptual	O
illusion	O
of	O
nonmediation	O
and	O
the	O
possibility	O
of	O
building	O
and	O
sharing	O
a	O
common	O
ground	O
.	O
In	O
this	O
sense	O
,	O
experiencing	O
presence	B-KEY
in	O
a	O
clinical	B-KEY
virtual	I-KEY
environment	I-KEY
-LRB-	O
VE	O
-RRB-	O
,	O
such	O
as	O
a	O
shared	B-KEY
virtual	I-KEY
hospital	I-KEY
,	O
requires	O
more	O
than	O
reproduction	O
of	O
the	O
physical	O
features	O
of	O
external	O
reality	O
.	O
It	O
requires	O
the	O
creation	O
and	O
sharing	O
of	O
the	O
cultural	O
web	O
that	O
makes	O
meaningful	O
,	O
and	O
therefore	O
visible	O
,	O
both	O
people	O
and	O
objects	O
populating	O
the	O
environment	O
.	O
The	O
paper	O
outlines	O
a	O
framework	O
for	O
supporting	O
the	O
development	O
and	O
tuning	O
of	O
clinically	O
oriented	O
VR	O
systems	O

Quantum	B-KEY
universal	I-KEY
variable-length	I-KEY
source	I-KEY
coding	I-KEY
We	O
construct	O
an	O
optimal	B-KEY
quantum	I-KEY
universal	I-KEY
variable-length	I-KEY
code	I-KEY
that	O
achieves	O
the	O
admissible	B-KEY
minimum	I-KEY
rate	I-KEY
,	O
i.e.	O
,	O
our	O
code	O
is	O
used	O
for	O
any	O
probability	B-KEY
distribution	I-KEY
of	O
quantum	B-KEY
states	I-KEY
.	O
Its	O
probability	O
of	O
exceeding	O
the	O
admissible	B-KEY
minimum	I-KEY
rate	I-KEY
exponentially	O
goes	O
to	O
0	O
.	O
Our	O
code	O
is	O
optimal	O
in	O
the	O
sense	O
of	O
its	O
exponent	B-KEY
.	O
In	O
addition	O
,	O
its	O
average	B-KEY
error	I-KEY
asymptotically	O
tends	O
to	O
0	O

Experimental	O
investigations	O
on	O
monitoring	O
and	O
control	O
of	O
induction	B-KEY
heating	I-KEY
process	I-KEY
for	O
semi-solid	O
alloys	O
using	O
the	O
heating	O
coil	O
as	O
sensor	O
A	O
method	O
of	O
monitoring	O
the	O
state	O
of	O
metal	O
alloys	O
during	O
induction	O
heating	O
and	O
control	O
of	O
the	O
heating	O
process	O
utilizing	O
the	O
heating	O
coil	O
itself	O
as	O
a	O
sensor	O
is	O
proposed	O
,	O
and	O
its	O
usefulness	O
and	O
effectiveness	O
were	O
experimentally	O
investigated	O
using	O
aluminium	O
A357	O
billets	O
for	O
the	O
semi-solid	O
metal	O
-LRB-	O
SSM	O
-RRB-	O
casting	O
processes	O
.	O
The	O
impedance	O
of	O
the	O
coil	O
containing	O
the	O
billet	O
was	O
continuously	O
measured	O
by	O
the	O
proposed	O
method	O
in	O
the	O
temperature	O
range	O
between	O
room	O
temperature	O
and	O
700	O
degrees	O
C	O
.	O
It	O
was	O
found	O
that	O
the	O
reactance	B-KEY
component	I-KEY
of	O
the	O
impedance	O
varied	O
distinctively	O
according	O
to	O
the	O
billet	B-KEY
state	I-KEY
and	O
could	O
clearly	O
monitor	O
the	O
deformation	O
of	O
the	O
billet	O
,	O
while	O
the	O
resistance	B-KEY
component	I-KEY
increased	O
with	O
temperature	O
,	O
reflecting	O
the	O
variation	O
of	O
the	O
resistivity	O
of	O
the	O
billet	O
which	O
has	O
strong	O
correlation	O
to	O
the	O
solid/liquid	B-KEY
fraction	I-KEY
of	O
the	O
billets	O
.	O
The	O
measured	O
impedance	O
is	O
very	O
sensitive	O
to	O
the	O
billet	B-KEY
states	I-KEY
such	O
as	O
temperature	O
,	O
deformation	O
and	O
solid/liquid	B-KEY
fraction	I-KEY
and	O
could	O
be	O
used	O
as	O
a	O
parameter	O
to	O
monitor	O
and	O
control	O
the	O
heating	O
process	O
for	O
SSMs	O

Take	O
it	O
to	O
the	O
next	O
level	O
-LSB-	O
law	B-KEY
firm	I-KEY
innovation	B-KEY
-RSB-	O
It	O
's	O
called	O
innovating	B-KEY
.	O
Our	O
clients	O
do	O
it	O
.	O
Our	O
culture	O
worships	O
it	O
.	O
Our	O
future	O
hinges	O
on	O
it	O
.	O
Why	O
is	O
it	O
so	O
difficult	O
in	O
law	B-KEY
firms	I-KEY
?	O
How	O
can	O
we	O
make	O
it	O
easier	O
?	O
Viva	O
la	O
difference	O
!	O

The	O
rise	O
and	O
fall	O
and	O
rise	O
again	O
of	O
customer	O
care	O
Taking	O
care	O
of	O
customers	O
has	O
never	O
gone	O
out	O
of	O
style	O
,	O
but	O
as	O
the	O
recession	O
fades	O
,	O
interest	O
is	O
picking	O
up	O
in	O
a	O
significant	O
retooling	O
of	O
the	O
CRM	O
solutions	O
banks	B-KEY
have	O
been	O
using	O
.	O
The	O
goal	O
:	O
usable	B-KEY
knowledge	I-KEY
to	O
help	O
improve	O
service	O

Comparative	B-KEY
statistical	I-KEY
analysis	I-KEY
of	O
hole	B-KEY
taper	I-KEY
and	O
circularity	B-KEY
in	O
laser	B-KEY
percussion	I-KEY
drilling	I-KEY
Investigates	O
the	O
relationships	O
and	O
parameter	O
interactions	O
between	O
six	O
controllable	O
variables	O
on	O
the	O
hole	B-KEY
taper	I-KEY
and	O
circularity	B-KEY
in	O
laser	B-KEY
percussion	I-KEY
drilling	I-KEY
.	O
Experiments	O
have	O
been	O
conducted	O
on	O
stainless	B-KEY
steel	I-KEY
workpieces	I-KEY
and	O
a	O
comparison	O
was	O
made	O
between	O
stainless	O
steel	O
and	O
mild	B-KEY
steel	I-KEY
.	O
The	O
central	B-KEY
composite	I-KEY
design	I-KEY
was	O
employed	O
to	O
plan	O
the	O
experiments	O
in	O
order	O
to	O
achieve	O
required	O
information	O
with	O
reduced	O
number	O
of	O
experiments	O
.	O
The	O
process	B-KEY
performance	I-KEY
was	O
evaluated	O
.	O
The	O
ratio	O
of	O
minimum	O
to	O
maximum	O
Feret	O
's	O
diameter	O
was	O
considered	O
as	O
circularity	B-KEY
characteristic	O
of	O
the	O
hole	O
.	O
The	O
models	O
of	O
these	O
three	O
process	O
characteristics	O
were	O
developed	O
by	O
linear	B-KEY
multiple	I-KEY
regression	I-KEY
technique	I-KEY
.	O
The	O
significant	O
coefficients	O
were	O
obtained	O
by	O
performing	O
analysis	B-KEY
of	I-KEY
variance	I-KEY
-LRB-	O
ANOVA	B-KEY
-RRB-	O
at	O
1	O
,	O
5	O
and	O
7	O
%	O
levels	O
of	O
significance	O
.	O
The	O
final	O
models	O
were	O
checked	O
by	O
complete	B-KEY
residual	I-KEY
analysis	I-KEY
and	O
finally	O
were	O
experimentally	O
verified	O
.	O
It	O
was	O
found	O
that	O
the	O
pulse	B-KEY
frequency	I-KEY
had	O
a	O
significant	O
effect	O
on	O
the	O
hole	O
entrance	O
diameter	O
and	O
hole	O
circularity	B-KEY
in	O
drilling	O
stainless	O
steel	O
unlike	O
the	O
drilling	O
of	O
mild	B-KEY
steel	I-KEY
where	O
the	O
pulse	B-KEY
frequency	I-KEY
had	O
no	O
significant	O
effect	O
on	O
the	O
hole	O
characteristics	O

Surface	O
micromachined	O
paraffin-actuated	O
microvalve	O
Normally-open	B-KEY
microvalves	I-KEY
have	O
been	O
fabricated	O
and	O
tested	O
which	O
use	O
a	O
paraffin	B-KEY
microactuator	I-KEY
as	O
the	O
active	B-KEY
element	I-KEY
.	O
The	O
entire	O
structure	O
with	O
nominal	O
dimension	O
of	O
phi	O
600	O
mu	O
m	O
*	O
30	O
mu	O
m	O
is	O
batch-fabricated	O
by	O
surface	O
micromachining	O
the	O
actuator	O
and	O
channel	B-KEY
materials	I-KEY
on	O
top	O
of	O
a	O
single	O
substrate	O
.	O
Gas	B-KEY
flow	I-KEY
rates	I-KEY
in	O
the	O
0.01-0	O
.1	O
sccm	O
range	O
have	O
been	O
measured	O
for	O
several	O
devices	O
with	O
actuation	O
powers	O
ranging	O
from	O
50	O
to	O
150	O
mW	O
on	O
glass	O
substrates	O
.	O
Leak	B-KEY
rates	I-KEY
as	O
low	O
as	O
500	O
mu	O
sccm	O
have	O
been	O
measured	O
.	O
The	O
normally-open	O
blocking	O
microvalve	O
structure	O
has	O
been	O
used	O
to	O
fabricate	O
a	O
precision	O
flow	O
control	O
system	O
of	O
microvalves	O
consisting	O
of	O
four	O
blocking	B-KEY
valve	I-KEY
structures	I-KEY
.	O
The	O
control	O
valve	O
is	O
designed	O
to	O
operate	O
over	O
a	O
0.01-5	O
.0	O
sccm	O
flow	O
range	O
at	O
a	O
differential	B-KEY
pressure	I-KEY
of	O
800	B-KEY
torr	I-KEY
.	O
Flow	B-KEY
rates	I-KEY
ranging	O
from	O
0.02	O
to	O
4.996	O
sccm	O
have	O
been	O
measured	O
.	O
Leak	B-KEY
rates	I-KEY
as	O
low	O
as	O
3.2	O
msccm	O
for	O
the	O
four	O
valve	O
system	O
have	O
been	O
measured	O

Becoming	O
a	O
computer	O
scientist	O
The	O
focus	O
of	O
this	O
report	O
is	O
pipeline	B-KEY
shrinkage	I-KEY
for	O
women	B-KEY
in	O
computer	B-KEY
science	I-KEY
.	O
We	O
describe	O
the	O
situation	O
for	O
women	B-KEY
at	O
all	O
stages	O
of	O
training	O
in	O
computer	B-KEY
science	I-KEY
,	O
from	O
the	O
precollege	O
level	O
through	O
graduate	O
school	O
.	O
Because	O
many	O
of	O
the	O
problems	O
discussed	O
are	O
related	O
to	O
the	O
lack	O
of	O
role	B-KEY
models	I-KEY
for	O
women	B-KEY
who	O
are	O
in	O
the	O
process	O
of	O
becoming	O
computer	O
scientists	O
,	O
we	O
also	O
concern	O
ourselves	O
with	O
the	O
status	O
of	O
women	B-KEY
faculty	O
members	O
.	O
We	O
not	O
only	O
describe	O
the	O
problems	O
,	O
but	O
also	O
make	O
specific	O
recommendations	O
for	O
change	O
and	O
encourage	O
further	O
study	O
of	O
those	O
problems	O
whose	O
solutions	O
are	O
not	O
yet	O
well	O
understood	O

Modeling	B-KEY
and	O
simulation	B-KEY
of	O
adaptive	B-KEY
available	I-KEY
bit	I-KEY
rate	I-KEY
voice	I-KEY
over	O
asynchronous	O
transfer	O
mode	O
networks	O
This	O
article	O
presents	O
a	O
modeling	B-KEY
and	O
simulation	B-KEY
methodology	O
to	O
analyze	O
the	O
performance	O
of	O
voice	B-KEY
quality	I-KEY
when	O
sent	O
over	O
the	O
available	O
bit	O
rate	O
service	O
in	O
asynchronous	O
transfer	O
mode	O
networks	O
.	O
Sources	O
can	O
modify	O
the	O
rate	O
at	O
which	O
they	O
send	O
traffic	B-KEY
to	O
the	O
network	O
based	O
on	O
the	O
feedback	B-KEY
carried	O
in	O
the	O
resource	B-KEY
management	I-KEY
cells	I-KEY
.	O
This	O
is	O
achieved	O
by	O
changing	O
the	O
encoding	B-KEY
level	I-KEY
.	O
As	O
the	O
contention	O
increases	O
to	O
network	O
resources-bandwidth	O
in	O
this	O
case-sources	O
start	O
reducing	O
the	O
rate	O
at	O
which	O
they	O
generate	O
and	O
send	O
traffic	B-KEY
.	O
The	O
efficiency	O
of	O
the	O
scheme	O
under	O
different	O
scheduling/drop	B-KEY
policies	I-KEY
and	O
other	O
operating	O
conditions	O
and	O
environments	O
is	O
evaluated	O
using	O
simulation	B-KEY
modeling	B-KEY
.	O
Furthermore	O
,	O
sensitivity	O
analysis	O
is	O
applied	O
to	O
different	O
parameters	O
,	O
such	O
as	O
queue	B-KEY
size	I-KEY
and	O
averaging	B-KEY
interval	I-KEY
length	I-KEY
,	O
to	O
investigate	O
their	O
impact	O
on	O
the	O
performance	B-KEY
metrics	I-KEY
.	O
Results	O
show	O
that	O
limiting	O
the	O
load	O
to	O
41	O
%	O
of	O
the	O
link	B-KEY
capacity	I-KEY
results	O
in	O
an	O
acceptable	O
quality	O

Post-projected	B-KEY
Runge-Kutta	I-KEY
methods	I-KEY
for	O
index-2	B-KEY
differential-algebraic	I-KEY
equations	I-KEY
A	O
new	O
projection	O
technique	O
for	O
Runge-Kutta	O
methods	O
applied	O
to	O
index-2	B-KEY
differential-algebraic	I-KEY
equations	I-KEY
is	O
presented	O
in	O
which	O
the	O
numerical	B-KEY
approximation	I-KEY
is	O
projected	O
only	O
as	O
part	O
of	O
the	O
output	O
process	O
.	O
It	O
is	O
shown	O
that	O
for	O
methods	O
that	O
are	O
strictly	O
stable	O
at	O
infinity	O
,	O
the	O
order	B-KEY
of	I-KEY
convergence	I-KEY
is	O
unaffected	O
compared	O
to	O
standard	O
projected	B-KEY
methods	I-KEY
.	O
Gauss	O
methods	O
,	O
for	O
which	O
this	O
technique	O
is	O
of	O
special	O
interest	O
when	O
some	O
symmetry	O
is	O
to	O
be	O
preserved	O
,	O
are	O
studied	O
in	O
more	O
detail	O

Satellite	B-KEY
image	I-KEY
collection	I-KEY
optimization	I-KEY
Imaging	B-KEY
satellite	I-KEY
systems	I-KEY
represent	O
a	O
high	O
capital	O
cost	O
.	O
Optimizing	O
the	O
collection	O
of	O
images	O
is	O
critical	O
for	O
both	O
satisfying	O
customer	O
orders	O
and	O
building	O
a	O
sustainable	O
satellite	O
operations	O
business	O
.	O
We	O
describe	O
the	O
functions	O
of	O
an	O
operational	O
,	O
multivariable	O
,	O
time	O
dynamic	O
optimization	O
system	O
that	O
maximizes	O
the	O
daily	O
collection	O
of	O
satellite	O
images	O
.	O
A	O
graphical	B-KEY
user	I-KEY
interface	I-KEY
allows	O
the	O
operator	O
to	O
quickly	O
see	O
the	O
results	O
of	O
what	O
if	O
adjustments	O
to	O
an	O
image	B-KEY
collection	I-KEY
plan	I-KEY
.	O
Used	O
for	O
both	O
long	B-KEY
range	I-KEY
planning	I-KEY
and	O
daily	B-KEY
collection	I-KEY
scheduling	I-KEY
of	O
Space	O
Imaging	O
's	O
IKONOS	O
satellite	O
,	O
the	O
satellite	O
control	O
and	O
tasking	O
-LRB-	O
SCT	O
-RRB-	O
software	O
allows	O
collection	B-KEY
commands	I-KEY
to	O
be	O
altered	O
up	O
to	O
10	O
min	O
before	O
upload	O
to	O
the	O
satellite	O

Critical	O
lines	O
identification	O
on	O
voltage	O
collapse	O
analysis	O
This	O
paper	O
deals	O
with	O
critical	O
lines	O
identification	O
on	O
voltage	O
collapse	O
analysis	O
.	O
It	O
is	O
known	O
,	O
from	O
the	O
literature	O
,	O
that	O
voltage	O
collapse	O
is	O
a	O
local	B-KEY
phenomenon	I-KEY
that	O
spreads	O
around	O
an	O
initial	O
neighborhood	O
Therefore	O
,	O
identifying	O
the	O
system	O
critical	O
bus	O
plays	O
an	O
important	O
role	O
on	O
voltage	O
collapse	O
prevention	O
.	O
For	O
this	O
purpose	O
,	O
the	O
system	O
critical	O
transmission	O
lines	O
should	O
also	O
be	O
identified	O
In	O
this	O
paper	O
,	O
these	O
issues	O
are	O
addressed	O
,	O
yielding	O
reliable	O
results	O
in	O
a	O
short	O
computational	O
time	O
.	O
Tests	O
are	O
done	O
with	O
the	O
help	O
of	O
the	O
IEEE-118	B-KEY
bus	I-KEY
and	O
the	O
Southeastern	O
Brazilian	O
systems	O

New	O
kit	O
on	O
the	O
block	O
-LSB-	O
IT	B-KEY
upgrades	I-KEY
-RSB-	O
As	O
time	O
passes	O
,	O
new	O
hardware	O
and	O
software	O
replace	O
the	O
old	O
.	O
The	O
hows	O
are	O
straightforward	O
:	B-KEY
IT	I-KEY
resellers	O
and	B-KEY
consultants	O
can	O
help	O
with	O
upgrade	O
practicalities	O
.	O
Will	O
Dalrymple	O
examines	O
the	O
business	B-KEY
issues	I-KEY
and	O
costs	B-KEY
involved	O
in	O
IT	B-KEY
upgrades	I-KEY

Public	B-KEY
business	I-KEY
libraries	I-KEY
:	O
the	O
next	O
chapter	O
Traces	O
the	O
history	B-KEY
of	O
the	O
provision	O
of	O
business	O
information	O
by	O
Leeds	B-KEY
Public	I-KEY
Libraries	I-KEY
,	O
UK	O
,	O
from	O
the	O
opening	O
of	O
the	O
Public	B-KEY
Commercial	I-KEY
and	I-KEY
Technical	I-KEY
Library	I-KEY
in	O
1918	O
to	O
the	O
revolutionary	O
impact	O
of	O
the	O
Internet	B-KEY
in	O
the	O
1990s	O
.	O
Describes	O
how	O
the	O
Library	O
came	O
to	O
terms	O
with	O
the	O
need	O
to	O
integrate	O
the	O
Internet	B-KEY
into	O
its	O
mainstream	O
business	B-KEY
information	I-KEY
services	I-KEY
,	O
with	O
particular	O
reference	O
to	O
its	O
limitations	O
and	O
to	O
the	O
provision	O
of	O
company	B-KEY
information	I-KEY
,	O
market	B-KEY
research	O
,	O
British	O
Standards	O
information	O
,	O
press	O
cuttings	O
and	O
articles	O
from	O
specialized	O
trade	O
and	O
scientific	O
journals	O
,	O
and	O
patents	O
information	O
.	O
Focuses	O
on	O
some	O
of	O
the	O
reasons	O
why	O
the	O
public	B-KEY
business	I-KEY
library	I-KEY
is	O
still	O
needed	O
as	O
a	O
service	O
to	O
businesses	O
,	O
even	O
after	O
the	O
introduction	O
of	O
the	O
Internet	B-KEY
and	O
considers	O
the	O
Library	O
's	O
changing	O
role	O
and	O
the	O
need	O
to	O
impress	O
on	O
all	O
concerned	O
,	O
especially	O
government	B-KEY
,	O
the	O
continuing	O
value	O
of	O
these	O
services	O
.	O
Looks	O
to	O
the	O
partnerships	O
formed	O
by	O
the	O
Library	O
over	O
the	O
years	O
and	O
the	O
ways	O
in	O
which	O
these	O
are	O
expected	O
to	O
assist	O
in	O
realizing	O
future	O
opportunities	O
,	O
in	O
particular	O
,	O
the	O
fact	O
that	O
all	O
public	O
libraries	O
in	O
England	O
gained	O
free	O
Internet	B-KEY
access	O
at	O
the	O
end	O
of	O
2001	O
.	O
Offers	O
some	O
useful	O
ideas	O
about	O
how	O
the	O
Library	O
could	O
develop	O
,	O
noting	O
that	O
SINTO	B-KEY
,	O
a	O
Sheffield	O
based	O
information	B-KEY
network	I-KEY
formed	O
in	O
1938	O
and	O
originally	O
a	O
partnership	O
between	O
the	O
public	O
library	O
,	O
the	O
two	O
Sheffield	B-KEY
universities	I-KEY
and	O
various	O
leading	O
steel	B-KEY
companies	I-KEY
of	O
the	O
time	O
,	O
is	O
being	O
examined	O
as	O
a	O
model	O
for	O
future	O
services	O
in	O
Leeds	O
.	O
Concludes	O
that	O
the	O
way	O
forward	O
can	O
be	O
defined	O
in	O
terms	O
of	O
five	O
actions	O
:	O
redefinition	O
of	O
priorities	O
;	O
marketing	B-KEY
;	O
budgets	B-KEY
;	O
resources	B-KEY
;	O
and	O
the	O
use	O
of	O
information	O
technology	O
-LRB-	O
IT	O
-RRB-	O

Leveraging	O
an	O
alternative	O
source	O
of	O
computer	O
scientists	O
:	O
reentry	B-KEY
programs	I-KEY
Much	O
has	O
been	O
written	O
about	O
the	O
leaky	O
pipeline	O
of	O
women	B-KEY
in	O
computer	B-KEY
science	I-KEY
-LRB-	O
CS	O
-RRB-	O
,	O
with	O
the	O
percentage	O
of	O
women	B-KEY
decreasing	O
as	O
one	O
moves	O
from	O
lower	O
levels	O
,	O
such	O
as	O
college	O
,	O
to	O
higher	O
levels	O
,	O
culminating	O
in	O
full	O
professorship	O
.	O
While	O
significant	O
attention	O
focused	O
on	O
keeping	O
women	B-KEY
from	O
leaving	O
the	O
pipeline	O
,	O
there	O
is	O
also	O
an	O
opportunity	O
to	O
bring	O
women	B-KEY
into	O
the	O
pipeline	O
through	O
non-traditional	O
programs	O
,	O
instead	O
of	O
requiring	O
that	O
everyone	O
enter	O
at	O
the	O
undergraduate	O
level	O
.	O
Both	O
Mills	B-KEY
College	I-KEY
,	O
a	O
small	O
liberal	O
arts	O
institution	O
for	O
women	B-KEY
,	O
and	O
UC	B-KEY
Berkeley	I-KEY
,	O
a	O
large	O
research	O
university	O
,	O
established	O
programs	O
in	O
the	O
80	O
's	O
to	O
increase	O
the	O
number	O
of	O
women	B-KEY
in	O
computer	B-KEY
science	I-KEY
by	O
tapping	O
non-traditional	O
students	B-KEY
.	O
Both	O
programs	O
share	O
the	O
core	O
value	O
of	O
accommodating	O
older	O
students	B-KEY
lacking	O
technical	O
backgrounds	O
.	O
The	O
two	O
programs	O
have	O
produced	O
similar	O
results	O
:	O
graduate	B-KEY
degrees	I-KEY
earned	O
in	O
computer	B-KEY
science	I-KEY
by	O
students	B-KEY
who	O
would	O
not	O
have	O
qualified	O
without	O
these	O
programs	O
,	O
professional	B-KEY
employment	I-KEY
in	O
the	O
computer	B-KEY
science	I-KEY
field	O
by	O
women	B-KEY
and	O
minorities	B-KEY
,	O
and	O
a	O
recognition	O
that	O
this	O
population	O
represents	O
a	O
rich	O
source	O
of	O
talent	O
for	O
our	O
nation	O

Model	B-KEY
theory	I-KEY
for	O
hereditarily	O
finite	O
superstructures	O
We	O
study	O
model-theoretic	B-KEY
properties	I-KEY
of	O
hereditarily	O
finite	O
superstructures	O
over	O
models	O
of	O
not	O
more	O
than	O
countable	B-KEY
signatures	I-KEY
.	O
A	O
question	O
is	O
answered	O
in	O
the	O
negative	O
inquiring	O
whether	O
theories	O
of	O
hereditarily	O
finite	O
superstructures	O
which	O
have	O
a	O
unique	O
-LRB-	O
up	O
to	O
isomorphism	O
-RRB-	O
hereditarily	O
finite	O
superstructure	O
can	O
be	O
described	O
via	O
definable	O
functions	O
.	O
Yet	O
theories	O
for	O
such	O
superstructures	O
admit	O
a	O
description	O
in	O
terms	O
of	O
iterated	B-KEY
families	I-KEY
TF	O
and	O
SF	O
.	O
These	O
are	O
constructed	O
using	O
a	O
definable	B-KEY
union	I-KEY
taken	O
over	O
countable	O
ordinals	O
in	O
the	O
subsets	O
which	O
are	O
unions	O
of	O
finitely	B-KEY
many	I-KEY
complete	I-KEY
subsets	I-KEY
and	O
of	O
finite	O
subsets	O
,	O
respectively	O
.	O
Simultaneously	O
,	O
we	O
describe	O
theories	O
that	O
share	O
a	O
unique	O
-LRB-	O
up	O
to	O
isomorphism	O
-RRB-	O
countable	B-KEY
hereditarily	I-KEY
finite	I-KEY
superstructure	I-KEY

Rational	B-KEY
systems	I-KEY
exhibit	O
moderate	B-KEY
risk	I-KEY
aversion	I-KEY
with	O
respect	O
to	O
``	O
gambles	B-KEY
''	O
on	O
variable-resolution	B-KEY
compression	I-KEY
In	O
an	O
embedded	B-KEY
wavelet	I-KEY
scheme	I-KEY
for	O
progressive	B-KEY
transmission	I-KEY
,	O
a	O
tree	B-KEY
structure	I-KEY
naturally	O
defines	O
the	O
spatial	O
relationship	O
on	O
the	O
hierarchical	O
pyramid	O
.	O
Transform	B-KEY
coefficients	I-KEY
over	O
each	O
tree	O
correspond	O
to	O
a	O
unique	O
local	B-KEY
spatial	I-KEY
region	I-KEY
of	O
the	O
original	O
image	O
,	O
and	O
they	O
can	O
be	O
coded	O
bit-plane	O
by	O
bit-plane	O
through	O
successive-approximation	B-KEY
quantization	I-KEY
.	O
After	O
receiving	O
the	O
approximate	O
value	O
of	O
some	O
coefficients	O
,	O
the	O
decoder	O
can	O
obtain	O
a	O
reconstructed	B-KEY
image	I-KEY
.	O
We	O
show	O
a	O
rational	B-KEY
system	I-KEY
for	O
progressive	B-KEY
transmission	I-KEY
that	O
,	O
in	O
absence	O
of	O
a	O
priori	O
knowledge	O
about	O
regions	O
of	O
interest	O
,	O
chooses	O
at	O
any	O
truncation	B-KEY
time	I-KEY
among	O
alternative	O
trees	O
for	O
further	O
transmission	O
in	O
such	O
a	O
way	O
as	O
to	O
avoid	O
certain	O
forms	O
of	O
behavioral	O
inconsistency	O
.	O
We	O
prove	O
that	O
some	O
rational	O
transmission	O
systems	O
might	O
exhibit	O
aversion	O
to	O
risk	O
involving	O
``	O
gambles	B-KEY
''	O
on	O
tree-dependent	O
quality	O
of	O
encoding	O
while	O
others	O
favor	O
taking	O
such	O
risks	O
.	O
Based	O
on	O
an	O
acceptable	B-KEY
predictor	I-KEY
for	O
visual	B-KEY
distinctness	I-KEY
from	O
digital	B-KEY
imagery	I-KEY
,	O
we	O
demonstrate	O
that	O
,	O
without	O
any	O
outside	O
knowledge	O
,	O
risk-prone	O
systems	O
as	O
well	O
as	O
those	O
with	O
strong	O
risk	O
aversion	O
appear	O
in	O
capable	O
of	O
attaining	O
the	O
quality	O
of	O
reconstructions	O
that	O
can	O
be	O
achieved	O
with	O
moderate	O
risk-averse	O
behavior	O

Grey-box	B-KEY
model	I-KEY
identification	O
via	O
evolutionary	O
computing	O
This	O
paper	O
presents	O
an	O
evolutionary	O
grey-box	B-KEY
model	I-KEY
identification	O
methodology	O
that	O
makes	O
the	O
best	O
use	O
of	O
a	O
priori	O
knowledge	O
on	O
a	O
clear-box	O
model	O
with	O
a	O
global	O
structural	O
representation	O
of	O
the	O
physical	O
system	O
under	O
study	O
,	O
whilst	O
incorporating	O
accurate	O
blackbox	O
models	O
for	O
immeasurable	O
and	O
local	O
nonlinearities	O
of	O
a	O
practical	O
system	O
.	O
The	O
evolutionary	O
technique	O
is	O
applied	O
to	O
building	O
dominant	O
structural	O
identification	O
with	O
local	O
parametric	O
tuning	O
without	O
the	O
need	O
of	O
a	O
differentiable	O
performance	O
index	O
in	O
the	O
presence	O
of	O
noisy	O
data	O
.	O
It	O
is	O
shown	O
that	O
the	O
evolutionary	O
technique	O
provides	O
an	O
excellent	O
fitting	O
performance	O
and	O
is	O
capable	O
of	O
accommodating	O
multiple	O
objectives	O
such	O
as	O
to	O
examine	O
the	O
relationships	O
between	O
model	O
complexity	O
and	O
fitting	O
accuracy	O
during	O
the	O
model	O
building	O
process	O
.	O
Validation	O
results	O
show	O
that	O
the	O
proposed	O
method	O
offers	O
robust	O
,	O
uncluttered	O
and	O
accurate	O
models	O
for	O
two	O
practical	O
systems	O
.	O
It	O
is	O
expected	O
that	O
this	O
type	O
of	O
grey-box	B-KEY
models	I-KEY
will	O
accommodate	O
many	O
practical	O
engineering	O
systems	O
for	O
a	O
better	O
modelling	O
accuracy	O

Design	O
of	O
high-performance	B-KEY
wavelets	I-KEY
for	O
image	B-KEY
coding	I-KEY
using	O
a	O
perceptual	B-KEY
time	I-KEY
domain	I-KEY
criterion	I-KEY
This	O
paper	O
presents	O
a	O
new	O
biorthogonal	B-KEY
linear-phase	I-KEY
wavelet	I-KEY
design	I-KEY
for	O
image	B-KEY
compression	I-KEY
.	O
Instead	O
of	O
calculating	O
the	O
prototype	B-KEY
filters	I-KEY
as	O
spectral	O
factors	O
of	O
a	O
half-band	B-KEY
filter	I-KEY
,	O
the	O
design	O
is	O
based	O
on	O
the	O
direct	O
optimization	O
of	O
the	O
low	O
pass	O
analysis	B-KEY
filter	I-KEY
using	O
an	O
objective	B-KEY
function	I-KEY
directly	O
related	O
to	O
a	O
perceptual	O
criterion	O
for	O
image	B-KEY
compression	I-KEY
.	O
This	O
function	O
is	O
defined	O
as	O
the	O
product	O
of	O
the	O
theoretical	O
coding	B-KEY
gain	I-KEY
and	O
an	O
index	O
called	O
the	O
peak-to-peak	B-KEY
ratio	I-KEY
,	O
which	O
was	O
shown	O
to	O
have	O
high	O
correlation	O
with	O
perceptual	O
quality	O
.	O
A	O
distinctive	O
feature	O
of	O
the	O
proposed	O
technique	O
is	O
a	O
procedure	O
by	O
which	O
,	O
given	O
a	O
``	O
good	O
''	O
starting	O
filter	O
,	O
``	O
good	O
''	O
filters	O
of	O
longer	O
lengths	O
are	O
generated	O
.	O
The	O
results	O
are	O
excellent	O
,	O
showing	O
a	O
clear	O
improvement	O
in	O
perceptual	B-KEY
image	I-KEY
quality	I-KEY
.	O
Also	O
,	O
we	O
devised	O
a	O
criterion	O
for	O
constraining	O
the	O
coefficients	O
of	O
the	O
filters	O
in	O
order	O
to	O
design	O
wavelets	O
with	O
minimum	O
ringing	O

Application	O
of	O
traditional	O
system	B-KEY
design	I-KEY
techniques	I-KEY
to	O
Web	O
site	O
design	O
After	O
several	O
decades	O
of	O
computer	O
program	O
construction	O
there	O
emerged	O
a	O
set	O
of	O
principles	O
that	O
provided	O
guidance	O
to	O
produce	O
more	O
manageable	O
programs	O
.	O
With	O
the	O
emergence	O
of	O
the	O
plethora	O
of	O
Internet	O
web	O
sites	O
one	O
wonders	O
if	O
similar	O
guidelines	O
are	O
followed	O
in	O
their	O
construction	O
.	O
Since	O
this	O
is	O
a	O
new	O
technology	O
no	O
apparent	O
universally	O
accepted	O
methods	O
have	O
emerged	O
to	O
guide	O
the	O
designer	O
in	O
Web	O
site	O
construction	O
.	O
This	O
paper	O
reviews	O
the	O
traditional	O
principles	O
of	O
structured	B-KEY
programming	I-KEY
and	O
the	O
preferred	O
characteristics	O
of	O
Web	O
sites	O
.	O
Finally	O
a	O
mapping	O
of	O
how	O
the	O
traditional	O
guidelines	O
may	O
be	O
applied	O
to	O
Web	O
site	O
construction	O
is	O
presented	O
.	O
The	O
application	O
of	O
the	O
traditional	O
principles	O
of	O
structured	B-KEY
programming	I-KEY
to	O
the	O
design	O
of	O
a	O
Web	O
site	O
can	O
provide	O
a	O
more	O
usable	O
site	O
for	O
the	O
visitors	O
to	O
the	O
site	O
.	O
The	O
additional	O
benefit	O
of	O
using	O
these	O
time-honored	O
techniques	O
is	O
the	O
creation	O
of	O
a	O
Web	O
site	O
that	O
will	O
be	O
easier	O
to	O
maintain	O
by	O
the	O
development	O
staff	O

Q-learning	O
for	O
risk-sensitive	B-KEY
control	I-KEY
We	O
propose	O
for	O
risk-sensitive	B-KEY
control	I-KEY
of	O
finite	B-KEY
Markov	I-KEY
chains	I-KEY
a	O
counterpart	O
of	O
the	O
popular	O
Q-learning	B-KEY
algorithm	I-KEY
for	O
classical	B-KEY
Markov	I-KEY
decision	I-KEY
processes	I-KEY
.	O
The	O
algorithm	O
is	O
shown	O
to	O
converge	O
with	O
probability	O
one	O
to	O
the	O
desired	O
solution	O
.	O
The	O
proof	B-KEY
technique	I-KEY
is	O
an	O
adaptation	O
of	O
the	O
o.d.e.	O
approach	O
for	O
the	O
analysis	O
of	O
stochastic	B-KEY
approximation	I-KEY
algorithms	I-KEY
,	O
with	O
most	O
of	O
the	O
work	O
involved	O
used	O
for	O
the	O
analysis	O
of	O
the	O
specific	O
o.d.e.s	O
that	O
arise	O

A	O
model	O
for	O
choosing	O
an	O
electronic	B-KEY
reserves	I-KEY
system	I-KEY
:	O
a	O
pre-implementation	O
study	O
at	O
the	O
library	O
of	O
Long	O
Island	O
University	O
's	O
Brooklyn	O
campus	O
This	O
study	O
explores	O
the	O
nature	O
of	O
electronic	O
reserves	O
-LRB-	O
e-reserves	O
-RRB-	O
and	O
investigates	O
the	O
possibilities	O
of	O
implementing	O
the	O
e-reserves	O
at	O
the	O
Long	O
Island	O
University/Brooklyn	O
Campus	O
Library	O
-LRB-	O
LIU/BCL	O
-RRB-	O

Testing	O
statistical	B-KEY
bounds	I-KEY
on	O
entanglement	B-KEY
using	O
quantum	B-KEY
chaos	I-KEY
Previous	O
results	O
indicate	O
that	O
while	O
chaos	O
can	O
lead	O
to	O
substantial	O
entropy	B-KEY
production	I-KEY
,	O
thereby	O
maximizing	B-KEY
dynamical	O
entanglement	B-KEY
,	O
this	O
still	O
falls	O
short	O
of	O
maximality	B-KEY
.	O
Random	B-KEY
matrix	I-KEY
theory	I-KEY
modeling	O
of	O
composite	B-KEY
quantum	I-KEY
systems	I-KEY
,	O
investigated	O
recently	O
,	O
entails	O
a	O
universal	B-KEY
distribution	I-KEY
of	O
the	O
eigenvalues	O
of	O
the	O
reduced	B-KEY
density	I-KEY
matrices	I-KEY
.	O
We	O
demonstrate	O
that	O
these	O
distributions	O
are	O
realized	O
in	O
quantized	B-KEY
chaotic	I-KEY
systems	I-KEY
by	O
using	O
a	O
model	O
of	O
two	O
coupled	O
and	O
kicked	B-KEY
tops	I-KEY
.	O
We	O
derive	O
an	O
explicit	O
statistical	O
universal	O
bound	O
on	O
entanglement	B-KEY
,	O
which	O
is	O
also	O
valid	O
for	O
the	O
case	O
of	O
unequal	O
dimensionality	O
of	O
the	O
Hilbert	B-KEY
spaces	I-KEY
involved	O
,	O
and	O
show	O
that	O
this	O
describes	O
well	O
the	O
bounds	O
observed	O
using	O
composite	O
quantized	B-KEY
chaotic	I-KEY
systems	I-KEY
such	O
as	O
coupled	O
tops	O

P2P	O
is	O
dead	O
,	O
long	O
live	O
P2P	O
Picture	O
the	O
problem	O
:	O
a	O
sprawling	O
multinational	O
has	O
hundreds	O
of	O
offices	O
,	O
thousands	O
of	O
workers	O
,	O
and	O
countless	O
amounts	O
of	O
intellectual	O
property	O
scattered	O
here	O
,	O
there	O
,	O
everywhere	O
.	O
In	O
Kuala	O
Lumpur	O
an	O
executive	O
needs	O
to	O
see	O
an	O
internally-generated	O
report	O
on	O
oil	O
futures	O
in	O
central	O
Asia-but	O
where	O
is	O
it	O
?	O
London	O
?	O
New	O
York	O
?	O
Moscow	O
?	O
With	O
a	O
few	O
clicks	O
of	O
the	O
mouse-and	O
the	O
right	O
P2P	B-KEY
technology	I-KEY
deployed	O
in-house-that	O
executive	O
will	O
find	O
and	O
retrieve	O
the	O
report	O
.	O
Without	O
P2P	O
that	O
might	O
be	O
impossible-certainly	O
it	O
would	O
be	O
time-consuming-and	O
,	O
right	O
there	O
,	O
the	O
argument	O
for	O
P2P	O
implementations	O
inside	O
enterprises	O
becomes	O
clear	O
.	O
Who	O
are	O
the	O
players	O
?	O
No	O
companies	O
have	O
managed	O
to	O
stake	O
out	O
clear	O
leads	O
and	O
the	O
fact	O
is	O
that	O
the	O
P2P	O
marketplace	O
now	O
is	O
up	O
for	O
grabs-but	O
the	O
exciting	O
news	O
is	O
that	O
a	O
range	O
of	O
small	O
and	O
startup	O
businesses	B-KEY
are	O
trying	O
to	O
grab	O
turf	O
and	O
quite	O
probably	O
,	O
if	O
the	O
analysts	O
are	O
right	O
,	O
a	O
few	O
of	O
these	O
now	O
little-known	O
companies	O
will	O
emerge	O
as	O
digital	B-KEY
content	I-KEY
stars	O
within	O
the	O
next	O
few	O
years	O
.	O
Cases	O
in	O
point	O
:	O
Groove	B-KEY
Networks	I-KEY
,	O
Avaki	B-KEY
,	O
WorldStreet	B-KEY
,	O
Yaga	B-KEY
,	O
NextPage	B-KEY
,	O
and	O
Kontiki	B-KEY
.	O
Very	O
different	O
companies-their	O
approach	O
to	O
the	O
markets	O
radically	O
differ-but	O
,	O
say	O
the	O
analysts	O
,	O
each	O
is	O
worth	O
a	O
close	O
look	O
because	O
among	O
them	O
they	O
are	O
defining	O
the	O
future	O
of	O
P2P	O

Quantum	B-KEY
retrodiction	I-KEY
in	O
open	B-KEY
systems	I-KEY
Quantum	B-KEY
retrodiction	I-KEY
involves	O
finding	O
the	O
probabilities	B-KEY
for	O
various	O
preparation	B-KEY
events	I-KEY
given	O
a	O
measurement	B-KEY
event	I-KEY
.	O
This	O
theory	O
has	O
been	O
studied	O
for	O
some	O
time	O
but	O
mainly	O
as	O
an	O
interesting	O
concept	O
associated	O
with	O
time	B-KEY
asymmetry	I-KEY
in	O
quantum	B-KEY
mechanics	I-KEY
.	O
Recent	O
interest	O
in	O
quantum	B-KEY
communications	I-KEY
and	O
cryptography	B-KEY
,	O
however	O
,	O
has	O
provided	O
retrodiction	O
with	O
a	O
potential	O
practical	O
application	O
.	O
For	O
this	O
purpose	O
quantum	B-KEY
retrodiction	I-KEY
in	O
open	B-KEY
systems	I-KEY
should	O
be	O
more	O
relevant	O
than	O
in	O
closed	O
systems	O
isolated	O
from	O
the	O
environment	O
.	O
In	O
this	O
paper	O
we	O
study	O
retrodiction	O
in	O
open	B-KEY
systems	I-KEY
and	O
develop	O
a	O
general	O
master	O
equation	O
for	O
the	O
backward	B-KEY
time	I-KEY
evolution	I-KEY
of	O
the	O
measured	O
state	O
,	O
which	O
can	O
be	O
used	O
for	O
calculating	O
preparation	B-KEY
probabilities	I-KEY
.	O
We	O
solve	O
the	O
master	O
equation	O
,	O
by	O
way	O
of	O
example	O
,	O
for	O
the	O
driven	O
two-level	O
atom	O
coupled	O
to	O
the	O
electromagnetic	O
field	O

Estimation	O
of	O
the	O
vanishing	O
point	O
for	O
automatic	B-KEY
driving	I-KEY
system	I-KEY
using	O
a	O
cross	B-KEY
ratio	I-KEY
This	O
paper	O
proposes	O
a	O
new	O
method	O
to	O
estimate	O
the	O
vanishing	O
point	O
used	O
as	O
the	O
vehicle	O
heading	O
,	O
which	O
is	O
essential	O
in	O
automatic	B-KEY
driving	I-KEY
systems	I-KEY
.	O
The	O
proposed	O
method	O
uses	O
a	O
cross	B-KEY
ratio	I-KEY
comprised	O
of	O
a	O
ratio	O
of	O
lengths	O
from	O
four	O
collinear	B-KEY
points	I-KEY
for	O
extracting	O
the	O
edges	O
that	O
shape	O
the	O
vanishing	O
point	O
.	O
Then	O
,	O
lines	O
that	O
intersect	O
at	O
one	O
point	O
are	O
fitted	O
to	O
the	O
edges	O
in	O
a	O
Hough	B-KEY
space	I-KEY
.	O
Consequently	O
,	O
the	O
vanishing	O
point	O
is	O
estimated	O
robustly	O
even	O
when	O
the	O
lane	B-KEY
markings	I-KEY
are	O
occluded	O
by	O
other	O
vehicles	O
.	O
In	O
the	O
presence	O
of	O
lane	B-KEY
markings	I-KEY
,	O
the	O
road	O
boundaries	O
are	O
also	O
estimated	O
at	O
the	O
same	O
time	O
.	O
Experimental	O
results	O
from	O
images	O
of	O
a	O
real	B-KEY
road	I-KEY
scene	I-KEY
show	O
the	O
effectiveness	O
of	O
the	O
proposed	O
method	O

Recent	O
researches	O
of	O
human	B-KEY
science	I-KEY
on	O
railway	B-KEY
systems	I-KEY
This	O
paper	O
presents	O
research	O
of	O
human	B-KEY
science	I-KEY
on	O
railway	B-KEY
systems	I-KEY
at	O
RTRI	B-KEY
.	O
They	O
are	O
roughly	O
divided	O
into	O
two	O
categories	O
:	O
research	O
to	O
improve	O
safety	O
and	O
those	O
to	O
improve	O
comfort	O
.	O
On	O
the	O
former	O
subject	O
,	O
for	O
the	O
safeguard	O
against	O
accidents	B-KEY
caused	O
by	O
human	B-KEY
errors	I-KEY
,	O
we	O
have	O
promoted	O
studies	O
of	O
psychological	B-KEY
aptitude	I-KEY
test	I-KEY
,	O
various	O
research	O
to	O
evaluate	O
train	O
drivers	O
'	O
working	O
conditions	O
and	O
environments	O
,	O
and	O
new	O
investigations	O
to	O
minimize	O
the	O
risk	O
of	O
passenger	O
casualties	O
at	O
train	B-KEY
accidents	I-KEY
.	O
On	O
the	O
latter	O
subject	O
,	O
we	O
have	O
developed	O
new	O
methods	O
to	O
evaluate	O
the	O
riding	B-KEY
comfort	I-KEY
including	O
that	O
of	O
tilt	B-KEY
train	I-KEY
,	O
and	O
started	O
research	O
on	O
the	O
improvement	O
of	O
railway	B-KEY
facilities	I-KEY
for	O
the	O
aged	O
and	O
the	O
disabled	O
from	O
the	O
viewpoint	O
of	O
universal	O
design	O

A	O
virtual	O
victory	O
-LSB-	O
virtual	O
networks	O
-RSB-	O
Newly	O
fashionable	O
virtual	B-KEY
network	I-KEY
operators	I-KEY
look	O
all	O
set	O
to	O
clean	O
up	O
in	O
the	O
corporate	B-KEY
sector	I-KEY

Improved	O
detection	O
of	O
lung	O
nodules	O
by	O
using	O
a	O
temporal	B-KEY
subtraction	I-KEY
technique	I-KEY
The	O
authors	O
evaluated	O
the	O
effect	O
of	O
a	O
temporal	B-KEY
subtraction	I-KEY
technique	I-KEY
for	O
digital	B-KEY
chest	I-KEY
radiography	I-KEY
with	O
regard	O
to	O
the	O
accuracy	O
of	O
detection	O
of	O
lung	O
nodules	O
.	O
Twenty	O
solitary	O
lung	O
nodules	O
smaller	O
than	O
30	B-KEY
mm	I-KEY
in	O
diameter	O
,	O
including	O
10	O
lung	O
cancers	O
and	O
10	O
benign	O
nodules	O
,	O
were	O
used	O
.	O
The	O
nodules	O
were	O
grouped	O
subjectively	O
according	O
to	O
their	O
subtlety	O
.	O
For	O
nonnodular	O
cases	O
,	O
20	O
nodules	O
without	O
perceptible	B-KEY
interval	I-KEY
changes	I-KEY
were	O
selected	O
.	O
All	O
chest	O
radiographs	O
were	O
obtained	O
by	O
using	O
a	O
computed	B-KEY
radiographic	I-KEY
system	I-KEY
,	O
and	O
temporal	O
subtraction	O
images	O
were	O
produced	O
by	O
using	O
a	O
program	O
developed	O
at	O
the	O
University	B-KEY
of	I-KEY
Chicago	I-KEY
.	O
The	O
effect	O
of	O
the	O
temporal	O
subtraction	O
image	O
was	O
evaluated	O
by	O
using	O
an	O
observer	B-KEY
performance	I-KEY
study	O
,	O
with	O
use	O
of	O
receiver	O
operating	O
characteristic	O
analysis	O
.	O
Observer	B-KEY
performance	I-KEY
with	O
temporal	O
subtraction	O
images	O
was	O
substantially	O
improved	O
-LRB-	O
A/sub	O
z	O
/	O
=	O
0.980	O
and	O
0.958	O
-RRB-	O
,	O
as	O
compared	O
with	O
that	O
without	O
temporal	O
subtraction	O
images	O
-LRB-	O
A/sub	O
z	O
/	O
=	O
0.920	O
and	O
0.825	O
-RRB-	O
for	O
the	O
certified	B-KEY
radiologists	I-KEY
and	O
radiology	B-KEY
residents	I-KEY
,	O
respectively	O
.	O
The	O
temporal	B-KEY
subtraction	I-KEY
technique	I-KEY
clearly	O
improved	O
diagnostic	O
accuracy	O
for	O
detecting	O
lung	O
nodules	O
,	O
especially	O
subtle	B-KEY
cases	I-KEY
.	O
In	O
conclusion	O
,	O
the	O
temporal	B-KEY
subtraction	I-KEY
technique	I-KEY
is	O
useful	O
for	O
improving	O
detection	O
accuracy	O
for	O
peripheral	B-KEY
lung	I-KEY
nodules	I-KEY
on	O
digital	O
chest	O
radiographs	O

Direct	B-KEY
gear	I-KEY
tooth	I-KEY
contact	I-KEY
analysis	I-KEY
for	O
hypoid	B-KEY
bevel	I-KEY
gears	I-KEY
A	O
new	O
methodology	O
for	O
tooth	O
contact	O
analysis	O
based	O
on	O
a	O
very	O
general	O
mathematical	B-KEY
model	I-KEY
of	O
the	O
generating	B-KEY
process	I-KEY
is	O
proposed	O
.	O
Considering	O
the	O
line	O
of	O
action	O
as	O
a	O
first	B-KEY
order	I-KEY
singularity	I-KEY
of	O
a	O
certain	O
operator	B-KEY
equation	I-KEY
we	O
develop	O
first	O
and	O
second	B-KEY
order	I-KEY
conditions	I-KEY
for	O
a	O
pair	O
of	O
generated	B-KEY
gear	I-KEY
tooth	I-KEY
flanks	I-KEY
to	O
be	O
in	O
contact	O
.	O
The	O
constructive	O
approach	O
allows	O
the	O
direct	O
computation	B-KEY
of	O
the	O
paths	O
of	O
contact	O
as	O
the	O
solution	O
of	O
a	O
nonlinear	B-KEY
equation	I-KEY
system	I-KEY
including	O
the	O
exact	O
determination	O
of	O
the	O
bounds	O
of	O
the	O
paths	O
of	O
contact	O
.	O
The	O
transmission	B-KEY
error	I-KEY
as	O
well	O
as	O
curvature	B-KEY
properties	I-KEY
in	O
the	O
contact	O
points	O
are	O
obtained	O
in	O
a	O
convenient	O
way	O
.	O
The	O
resulting	O
contact	B-KEY
ellipses	I-KEY
approximate	O
the	O
bearing	B-KEY
area	I-KEY
.	O
Through	O
the	O
use	O
of	O
automatic	B-KEY
differentiation	I-KEY
all	O
the	O
geometric	B-KEY
quantities	I-KEY
are	O
calculable	O
within	O
the	O
machine	B-KEY
accuracy	I-KEY
of	O
the	O
computer	B-KEY

Quadratic	B-KEY
programming	I-KEY
algorithms	I-KEY
for	O
large-scale	B-KEY
model	I-KEY
predictive	I-KEY
control	I-KEY
Quadratic	O
programming	O
-LRB-	O
QP	O
-RRB-	O
methods	O
are	O
an	O
important	O
element	O
in	O
the	O
application	O
of	O
model	O
predictive	O
control	O
-LRB-	O
MPC	O
-RRB-	O
.	O
As	O
larger	O
and	O
more	O
challenging	O
MPC	O
applications	O
are	O
considered	O
,	O
more	O
attention	O
needs	O
to	O
be	O
focused	O
on	O
the	O
construction	O
and	O
tailoring	O
of	O
efficient	O
QP	O
algorithms	O
.	O
In	O
this	O
study	O
,	O
we	O
tailor	O
and	O
apply	O
a	O
new	O
QP	O
method	O
,	O
called	O
QPSchur	B-KEY
,	O
to	O
large	O
MPC	O
applications	O
,	O
such	O
as	O
cross	B-KEY
directional	I-KEY
control	I-KEY
problems	I-KEY
in	O
paper	B-KEY
machines	I-KEY
.	O
Written	O
in	O
C++	O
,	O
QPSchur	B-KEY
is	O
an	O
object	B-KEY
oriented	I-KEY
implementation	I-KEY
of	O
a	O
novel	O
dual	O
space	O
,	O
Schur	O
complement	O
algorithm	O
.	O
We	O
compare	O
this	O
approach	O
to	O
three	O
widely	O
applied	O
QP	O
algorithms	O
and	O
show	O
that	O
QPSchur	B-KEY
is	O
significantly	O
more	O
efficient	O
-LRB-	O
up	O
to	O
two	O
orders	O
of	O
magnitude	O
-RRB-	O
than	O
the	O
other	O
algorithms	O
.	O
In	O
addition	O
,	O
detailed	O
simulations	B-KEY
are	O
considered	O
that	O
demonstrate	O
the	O
importance	O
of	O
the	O
flexible	O
,	O
object	O
oriented	O
construction	O
of	O
QPSchur	B-KEY
,	O
along	O
with	O
additional	O
features	O
for	O
constraint	B-KEY
handling	I-KEY
,	O
warm	B-KEY
starts	I-KEY
and	O
partial	B-KEY
solution	I-KEY

The	O
cataloger	O
's	O
workstation	O
revisited	O
:	O
utilizing	O
Cataloger	O
's	O
Desktop	O
A	O
few	O
years	O
into	O
the	O
development	O
of	O
Cataloger	O
's	O
Desktop	O
,	O
an	O
electronic	B-KEY
cataloging	I-KEY
tool	I-KEY
aggregator	O
available	O
through	O
the	O
Library	O
of	O
Congress	O
,	O
is	O
an	O
opportune	O
time	O
to	O
assess	O
its	O
impact	O
on	O
cataloging	O
operations	O
.	O
A	O
search	O
for	O
online	B-KEY
cataloging	I-KEY
tools	I-KEY
on	O
the	O
Internet	B-KEY
indicates	O
a	O
proliferation	O
of	O
cataloging	B-KEY
tool	I-KEY
aggregators	I-KEY
;	O
which	O
provide	O
access	O
to	O
online	B-KEY
documentation	I-KEY
related	O
to	O
cataloging	O
practices	O
and	O
procedures	O
.	O
Cataloger	O
's	O
Desktop	O
stands	O
out	O
as	O
a	O
leader	O
among	O
these	O
aggregators	O
.	O
Results	O
of	O
a	O
survey	O
to	O
assess	O
159	O
academic	B-KEY
ARL	I-KEY
and	O
large	B-KEY
public	I-KEY
libraries	I-KEY
'	O
reasons	O
for	O
use	O
or	O
non-use	O
of	O
Cataloger	O
's	O
Desktop	O
highlight	O
the	O
necessity	O
of	O
developing	O
strategies	O
for	O
its	O
successful	O
implementation	O
including	O
training	O
staff	O
,	O
providing	O
documentation	B-KEY
,	O
and	O
managing	B-KEY
technical	I-KEY
issues	I-KEY

Advanced	O
optimization	O
strategies	O
in	O
the	O
Rice	B-KEY
dHPF	I-KEY
compiler	I-KEY
High-Performance	O
Fortran	O
-LRB-	O
HPF	O
-RRB-	O
was	O
envisioned	O
as	O
a	O
vehicle	O
for	O
modernizing	O
legacy	B-KEY
Fortran	I-KEY
codes	I-KEY
to	O
achieve	O
scalable	O
parallel	B-KEY
performance	I-KEY
.	O
To	O
a	O
large	O
extent	O
,	O
today	O
's	O
commercially	O
available	O
HPF	B-KEY
compilers	I-KEY
have	O
failed	O
to	O
deliver	O
scalable	O
parallel	B-KEY
performance	I-KEY
for	O
a	O
broad	O
spectrum	O
of	O
applications	O
because	O
of	O
insufficiently	O
powerful	O
compiler	B-KEY
analysis	I-KEY
and	O
optimization	O
.	O
Substantial	O
restructuring	O
and	O
hand-optimization	O
can	O
be	O
required	O
to	O
achieve	O
acceptable	O
performance	O
with	O
an	O
HPF	O
port	O
of	O
an	O
existing	O
Fortran	O
application	O
,	O
even	O
for	O
regular	O
data-parallel	O
applications	O
.	O
A	O
key	O
goal	O
of	O
the	O
Rice	B-KEY
dHPF	I-KEY
compiler	I-KEY
project	O
has	O
been	O
to	O
develop	O
optimization	O
techniques	O
that	O
enable	O
a	O
wide	O
range	O
of	O
existing	O
scientific	O
applications	O
to	O
be	O
ported	O
easily	O
to	O
efficient	O
HPF	O
with	O
minimal	O
restructuring	O
.	O
This	O
paper	O
describes	O
the	O
challenges	O
to	O
effective	O
parallelization	O
presented	O
by	O
complex	O
-LRB-	O
but	O
regular	O
-RRB-	O
data-parallel	O
applications	O
,	O
and	O
then	O
describes	O
how	O
the	O
novel	O
analysis	O
and	O
optimization	O
technologies	O
in	O
the	O
dHPF	O
compiler	O
address	O
these	O
challenges	O
effectively	O
,	O
without	O
major	O
rewriting	O
of	O
the	O
applications	O
.	O
We	O
illustrate	O
the	O
techniques	O
by	O
describing	O
their	O
use	O
for	O
parallelizing	O
the	O
NAS	O
SP	O
and	O
BT	O
benchmarks	O
.	O
The	O
dHPF	O
compiler	O
generates	O
multipartitioned	B-KEY
parallelizations	O
of	O
these	O
codes	O
that	O
are	O
approaching	O
the	O
scalability	O
and	O
efficiency	O
of	O
sophisticated	O
hand-coded	O
parallelizations	O

Efficient	O
feasibility	B-KEY
testing	I-KEY
for	O
dial-a-ride	B-KEY
problems	I-KEY
Dial-a-ride	O
systems	O
involve	O
dispatching	B-KEY
a	O
vehicle	O
to	O
satisfy	O
demands	O
from	O
a	O
set	O
of	O
customers	O
who	O
call	O
a	O
vehicle-operating	B-KEY
agency	I-KEY
requesting	O
that	O
an	O
item	O
tie	O
picked	O
up	O
from	O
a	O
specific	O
location	O
and	O
delivered	O
to	O
a	O
specific	O
destination	O
.	O
Dial-a-ride	B-KEY
problems	I-KEY
differ	O
from	O
other	O
routing	B-KEY
and	O
scheduling	B-KEY
problems	O
,	O
in	O
that	O
they	O
typically	O
involve	O
service-related	B-KEY
constraints	I-KEY
.	O
It	O
is	O
common	O
to	O
have	O
maximum	B-KEY
wait	I-KEY
time	I-KEY
constraints	I-KEY
and	O
maximum	B-KEY
ride	I-KEY
time	I-KEY
constraints	I-KEY
.	O
In	O
the	O
presence	O
of	O
maximum	O
wait	O
time	O
and	O
maximum	O
ride	O
time	O
restrictions	O
,	O
it	O
is	O
not	O
clear	O
how	O
to	O
efficiently	O
determine	O
,	O
given	O
a	O
sequence	O
of	O
pickups	O
and	O
deliveries	O
,	O
whether	O
a	O
feasible	O
schedule	B-KEY
exists	O
.	O
We	O
demonstrate	O
that	O
this	O
,	O
in	O
fact	O
,	O
can	O
be	O
done	O
in	O
linear	O
time	O

Enhanced	O
product	O
support	O
through	O
intelligent	B-KEY
product	I-KEY
manuals	I-KEY
The	O
scope	O
of	O
this	O
paper	O
is	O
the	O
provision	O
of	O
intelligent	B-KEY
product	I-KEY
support	I-KEY
within	O
the	O
distributed	O
Intranet/Internet	O
environment	O
.	O
From	O
the	O
point	O
of	O
view	O
of	O
user	O
requirements	O
,	O
the	O
limitations	O
of	O
conventional	O
product	B-KEY
manuals	I-KEY
and	O
methods	O
of	O
authoring	O
them	O
are	O
first	O
outlined	O
.	O
It	O
is	O
argued	O
that	O
enhanced	O
product	O
support	O
requires	O
new	O
technology	O
solutions	O
both	O
for	O
product	B-KEY
manuals	I-KEY
and	O
for	O
their	O
authoring	O
and	O
presentation	O
.	O
The	O
concept	O
and	O
the	O
architecture	O
of	O
intelligent	B-KEY
product	I-KEY
manuals	I-KEY
are	O
then	O
discussed	O
.	O
A	O
prototype	O
system	O
called	O
ProARTWeb	B-KEY
is	O
presented	O
to	O
demonstrate	O
advanced	O
features	O
of	O
intelligent	B-KEY
product	I-KEY
manuals	I-KEY
.	O
Next	O
,	O
the	O
problem	O
of	O
producing	O
such	O
manuals	O
in	O
a	O
cost-effective	O
way	O
is	O
addressed	O
and	O
a	O
concurrent	B-KEY
engineering	I-KEY
approach	O
to	O
their	O
authoring	O
is	O
proposed	O
.	O
An	O
integrated	O
environment	O
for	O
collaborative	O
authoring	O
called	O
ProAuthor	O
is	O
described	O
to	O
illustrate	O
the	O
approach	O
suggested	O
and	O
to	O
show	O
how	O
consistent	O
,	O
up-to-date	O
and	O
user-oriented-product	O
manuals	O
can	O
be	O
designed	O
.	O
The	O
solutions	O
presented	O
here	O
enable	O
product	B-KEY
knowledge	I-KEY
to	O
be	O
captured	O
and	O
delivered	O
to	O
users	O
and	O
developers	O
of	O
product	B-KEY
manuals	I-KEY
when	O
,	O
where	O
and	O
in	O
the	O
form	O
they	O
need	O
it	O

Universal	B-KEY
approximation	I-KEY
by	O
hierarchical	B-KEY
fuzzy	I-KEY
system	I-KEY
with	O
constraints	O
on	O
the	O
fuzzy	B-KEY
rule	I-KEY
This	O
paper	O
presents	O
a	O
special	O
hierarchical	B-KEY
fuzzy	I-KEY
system	I-KEY
where	O
the	O
outputs	O
of	O
the	O
previous	O
layer	O
are	O
not	O
used	O
in	O
the	O
IF-parts	O
,	O
but	O
used	O
only	O
in	O
the	O
THEN-parts	O
of	O
the	O
fuzzy	B-KEY
rules	I-KEY
of	O
the	O
current	O
layer	O
.	O
The	O
proposed	O
scheme	O
can	O
be	O
shown	O
to	O
be	O
a	O
universal	B-KEY
approximator	I-KEY
to	O
any	O
continuous	B-KEY
function	I-KEY
on	O
a	O
compact	O
set	O
if	O
complete	O
fuzzy	O
sets	O
are	O
used	O
in	O
the	O
IF-parts	O
of	O
the	O
fuzzy	B-KEY
rules	I-KEY
with	O
singleton	O
fuzzifier	O
and	O
center	O
average	O
defuzzifier	O
.	O
From	O
the	O
simulation	O
of	O
ball	B-KEY
and	I-KEY
beam	I-KEY
control	I-KEY
system	I-KEY
,	O
it	O
is	O
demonstrated	O
that	O
the	O
proposed	O
scheme	O
approximates	O
with	O
good	O
accuracy	O
the	O
model	O
nonlinear	O
controller	O
with	O
fewer	O
fuzzy	B-KEY
rules	I-KEY
than	O
the	O
centralized	O
fuzzy	O
system	O
and	O
its	O
control	O
performance	O
is	O
comparable	O
to	O
that	O
of	O
the	O
nonlinear	O
controller	O

Automating	O
the	O
compliance	B-KEY
and	O
supervision	B-KEY
process	O
New	O
technology	O
enables	O
large	O
broker/dealers	O
to	O
supervise	O
and	O
ensure	O
compliance	O
across	O
multiple	O
branches	O
and	O
managers	O

Controlled	B-KEY
projective	B-KEY
synchronization	I-KEY
in	O
nonpartially-linear	B-KEY
chaotic	I-KEY
systems	I-KEY
Projective	B-KEY
synchronization	I-KEY
-LRB-	O
PS	O
-RRB-	O
,	O
in	O
which	O
the	O
state	O
vectors	O
synchronize	O
up	O
to	O
a	O
scaling	B-KEY
factor	I-KEY
,	O
is	O
usually	O
observable	O
only	O
in	O
partially	O
linear	O
systems	O
.	O
We	O
show	O
that	O
PS	O
could	O
,	O
by	O
means	O
of	O
control	B-KEY
,	O
be	O
extended	O
to	O
general	O
classes	O
of	O
chaotic	O
systems	O
with	O
nonpartial	O
linearity	O
.	O
Performance	O
of	O
PS	O
may	O
also	O
be	O
manipulated	O
by	O
controlling	B-KEY
the	O
scaling	B-KEY
factor	I-KEY
to	O
any	O
desired	O
value	O
.	O
In	O
numerical	O
experiments	O
,	O
we	O
illustrate	O
the	O
applications	O
to	O
a	O
Rossler	B-KEY
system	I-KEY
and	O
a	O
Chua	O
's	O
circuit	O
.	O
The	O
feasibility	O
of	O
the	O
control	B-KEY
for	O
high	O
dimensional	O
systems	O
is	O
demonstrated	O
in	O
a	O
hyperchaotic	B-KEY
system	I-KEY

Wavelet-based	B-KEY
level-of-detail	I-KEY
representation	I-KEY
of	O
3D	O
objects	O
In	O
this	O
paper	O
,	O
we	O
propose	O
a	O
3D	O
object	O
LOD	O
-LRB-	O
Level	O
of	O
Detail	O
-RRB-	O
modeling	O
system	O
that	O
constructs	O
a	O
mesh	O
from	O
range	B-KEY
images	I-KEY
and	O
generates	O
the	O
mesh	O
of	O
various	O
LOD	O
using	O
the	O
wavelet	B-KEY
transform	I-KEY
.	O
In	O
the	O
initial	O
mesh	O
generation	O
,	O
we	O
use	O
the	O
marching	B-KEY
cube	I-KEY
algorithm	I-KEY
.	O
We	O
modify	O
the	O
original	O
algorithm	O
to	O
apply	O
it	O
to	O
construct	O
the	O
mesh	O
from	O
multiple	O
range	B-KEY
images	I-KEY
efficiently	O
.	O
To	O
get	O
the	O
base	B-KEY
mesh	I-KEY
we	O
use	O
the	O
decimation	B-KEY
algorithm	I-KEY
which	O
simplifies	O
a	O
mesh	O
with	O
preserving	O
the	O
topology	O
.	O
Finally	O
,	O
when	O
reconstructing	O
new	O
mesh	O
which	O
is	O
similar	O
to	O
initial	O
mesh	O
we	O
calculate	O
the	O
wavelet	B-KEY
coefficients	I-KEY
by	O
using	O
the	O
wavelet	B-KEY
transform	I-KEY
.	O
We	O
solve	O
the	O
critical	B-KEY
problem	I-KEY
of	O
wavelet-based	O
methods	O
-	O
the	O
surface	B-KEY
crease	I-KEY
problem	I-KEY
-	O
by	O
using	O
the	O
mesh	B-KEY
simplification	I-KEY
as	O
the	O
base	B-KEY
mesh	I-KEY
generation	O
method	O

Modeling	O
frequently	O
accessed	O
wireless	O
data	O
with	O
weak	B-KEY
consistency	I-KEY
To	O
reduce	O
the	O
response	O
times	O
of	O
wireless	B-KEY
data	I-KEY
access	I-KEY
in	O
a	O
mobile	B-KEY
network	I-KEY
,	O
caches	B-KEY
are	O
utilized	O
in	O
wireless	B-KEY
handheld	I-KEY
devices	I-KEY
.	O
If	O
the	O
original	O
data	B-KEY
entry	I-KEY
has	O
been	O
updated	O
,	O
the	O
cached	B-KEY
data	O
in	O
the	O
handheld	O
device	O
becomes	O
stale	O
.	O
Thus	O
,	O
a	O
mechanism	O
is	O
required	O
to	O
predict	O
when	O
the	O
cached	B-KEY
copy	O
will	O
expire	O
.	O
This	O
paper	O
studies	O
a	O
weakly	O
consistent	O
data	O
access	O
mechanism	O
that	O
computes	O
the	O
time-to-live	O
-LRB-	O
TTL	O
-RRB-	O
interval	O
to	O
predict	O
the	O
expiration	O
time	O
.	O
We	O
propose	O
an	O
analytic	B-KEY
model	I-KEY
to	O
investigate	O
this	O
TTL-based	O
algorithm	O
for	O
frequently	O
accessed	O
data	O
.	O
The	O
analytic	B-KEY
model	I-KEY
is	O
validated	O
against	O
simulation	B-KEY
experiments	I-KEY
.	O
Our	O
study	O
quantitatively	O
indicates	O
how	O
the	O
TTL-based	O
algorithm	O
reduces	O
the	O
wireless	B-KEY
communication	I-KEY
cost	I-KEY
by	O
increasing	O
the	O
probability	O
of	O
stale	O
accesses	O
.	O
Depending	O
on	O
the	O
requirements	O
of	O
the	O
application	O
,	O
appropriate	O
parameter	O
values	O
can	O
be	O
selected	O
based	O
on	O
the	O
guidelines	O
provided	O

BT	O
voices	O
its	O
support	O
for	O
IP	O
BTexact	B-KEY
's	O
chief	O
technology	O
officer	O
,	O
Mick	O
Reeve	O
,	O
gives	O
his	O
views	O
on	O
the	O
future	O
for	O
voice	B-KEY
over	I-KEY
DSL	I-KEY
services	O
and	O
virtual	B-KEY
private	I-KEY
networks	I-KEY
,	O
and	O
defends	O
the	O
slow	O
rollout	O
of	O
public	B-KEY
access	I-KEY
WLANs	I-KEY

Scribe	B-KEY
:	O
a	O
large-scale	O
and	O
decentralized	B-KEY
application-level	I-KEY
multicast	I-KEY
infrastructure	I-KEY
This	O
paper	O
presents	O
Scribe	B-KEY
,	O
a	O
scalable	B-KEY
application-level	I-KEY
multicast	I-KEY
infrastructure	I-KEY
.	O
Scribe	B-KEY
supports	O
large	O
numbers	O
of	O
groups	O
,	O
with	O
a	O
potentially	O
large	O
number	O
of	O
members	O
per	O
group	O
.	O
Scribe	B-KEY
is	O
built	O
on	O
top	O
of	O
Pastry	B-KEY
,	O
a	O
generic	B-KEY
peer-to-peer	I-KEY
object	I-KEY
location	I-KEY
and	O
routing	O
substrate	O
overlayed	O
on	O
the	O
Internet	B-KEY
,	O
and	O
leverages	O
Pastry	B-KEY
's	O
reliability	O
,	O
self-organization	B-KEY
,	O
and	O
locality	B-KEY
properties	I-KEY
.	O
Pastry	B-KEY
is	O
used	O
to	O
create	O
and	O
manage	O
groups	O
and	O
to	O
build	O
efficient	O
multicast	O
trees	O
for	O
the	O
dissemination	O
of	O
messages	O
to	O
each	O
group	O
.	O
Scribe	B-KEY
provides	O
best-effort	B-KEY
reliability	I-KEY
guarantees	I-KEY
,	O
and	O
we	O
outline	O
how	O
an	O
application	O
can	O
extend	O
Scribe	B-KEY
to	O
provide	O
stronger	O
reliability	O
.	O
Simulation	B-KEY
results	I-KEY
,	O
based	O
on	O
a	O
realistic	O
network	B-KEY
topology	I-KEY
model	I-KEY
,	O
show	O
that	O
Scribe	B-KEY
scales	O
across	O
a	O
wide	O
range	O
of	O
groups	O
and	O
group	B-KEY
sizes	I-KEY
.	O
Also	O
,	O
it	O
balances	O
the	O
load	O
on	O
the	O
nodes	O
while	O
achieving	O
acceptable	O
delay	B-KEY
and	O
link	B-KEY
stress	I-KEY
when	O
compared	O
with	O
Internet	B-KEY
protocol	O
multicast	O

Automated	B-KEY
post	I-KEY
bonding	I-KEY
inspection	I-KEY
by	O
using	O
machine	B-KEY
vision	I-KEY
techniques	O
Inspection	O
plays	O
an	O
important	O
role	O
in	O
the	O
semiconductor	B-KEY
industry	I-KEY
.	O
In	O
this	O
paper	O
,	O
we	O
focus	O
on	O
the	O
inspection	O
task	O
after	O
wire	B-KEY
bonding	I-KEY
in	O
packaging	B-KEY
.	O
The	O
purpose	O
of	O
wire	B-KEY
bonding	I-KEY
-LRB-	O
W/B	O
-RRB-	O
is	O
to	O
connect	O
the	O
bond	O
pads	O
with	O
the	O
lead	B-KEY
fingers	I-KEY
.	O
Two	O
major	O
types	O
of	O
defects	O
are	O
-LRB-	O
1	O
-RRB-	O
bonding	B-KEY
line	I-KEY
missing	I-KEY
and	O
-LRB-	O
2	O
-RRB-	O
bonding	B-KEY
line	I-KEY
breakage	I-KEY
.	O
The	O
numbers	O
of	O
bonding	O
lines	O
and	O
bonding	B-KEY
balls	I-KEY
are	O
used	O
as	O
the	O
features	O
for	O
defect	B-KEY
classification	I-KEY
.	O
The	O
proposed	O
method	O
consists	O
of	O
image	B-KEY
preprocessing	I-KEY
,	O
orientation	B-KEY
determination	I-KEY
,	O
connection	B-KEY
detection	I-KEY
,	O
bonding	B-KEY
line	I-KEY
detection	I-KEY
,	O
bonding	B-KEY
ball	I-KEY
detection	O
,	O
and	O
defect	O
classification	O
.	O
The	O
proposed	O
method	O
is	O
simple	O
and	O
fast	O
.	O
The	O
experimental	O
results	O
show	O
that	O
the	O
proposed	O
method	O
can	O
detect	O
the	O
defects	O
effectively	O

Use	O
of	O
extra	O
degrees	B-KEY
of	I-KEY
freedom	I-KEY
in	O
multilevel	B-KEY
drives	I-KEY
Multilevel	O
converters	O
with	O
series	B-KEY
connection	I-KEY
of	O
semiconductors	B-KEY
allow	O
power	B-KEY
electronics	I-KEY
to	O
reach	O
medium	B-KEY
voltages	I-KEY
-LRB-	O
1-10	O
kV	O
-RRB-	O
with	O
relatively	O
standard	O
components	O
.	O
The	O
increase	O
of	O
the	O
number	O
of	O
semiconductors	B-KEY
provides	O
extra	O
degrees	B-KEY
of	I-KEY
freedom	I-KEY
,	O
which	O
can	O
be	O
used	O
to	O
improve	O
different	O
characteristics	O
.	O
This	O
paper	O
is	O
focused	O
on	O
variable-speed	B-KEY
drives	I-KEY
and	O
it	O
is	O
shown	O
that	O
with	O
the	O
proposed	O
multilevel	B-KEY
direct	I-KEY
torque	I-KEY
control	I-KEY
strategy	I-KEY
-LRB-	O
DiCoIF	O
-RRB-	O
the	O
tradeoff	O
between	O
the	O
performances	O
of	O
the	O
drive	O
-LRB-	O
harmonic	B-KEY
distortions	I-KEY
,	O
torque	B-KEY
dynamics	I-KEY
,	O
voltage	B-KEY
step	I-KEY
gradients	I-KEY
,	O
etc.	O
-RRB-	O
and	O
the	O
switching	B-KEY
frequency	I-KEY
of	O
the	O
semiconductors	B-KEY
is	O
improved	O
.	O
Then	O
,	O
a	O
slightly	O
modified	O
strategy	O
reducing	O
common-mode	O
voltage	O
and	O
bearing	B-KEY
currents	I-KEY
is	O
presented	O

PacketVideo	B-KEY
.	O
One	O
step	O
ahead	O
of	O
the	O
streaming	O
wireless	O
market	O
Go	O
beyond	O
the	O
hype	O
,	O
however	O
,	O
and	O
it	O
's	O
clear	O
that	O
PacketVideo	B-KEY
is	O
making	O
strides	O
in	O
delivering	O
streaming	O
multimedia	O
content	O
to	O
wireless	B-KEY
devices	I-KEY
.	O
For	O
one	O
thing	O
,	O
its	O
technology	O
,	O
based	O
on	O
the	O
industry-standard	O
Motion	O
Pictures	O
Expert	O
Group	O
4	O
-LRB-	O
MPEG-4	B-KEY
-RRB-	O
video	O
encoder/decoder	O
,	O
actually	O
works	O
as	O
promised	O
.	O
Secondly	O
,	O
the	O
company	O
has	O
forged	O
a	O
broad-based	O
band	O
of	O
alliances	O
that	O
not	O
only	O
will	O
eventually	O
help	O
it	O
reach	O
potential	O
customers	O
down	O
the	O
road	O
,	O
but	O
provides	O
it	O
financial	O
support	O
until	O
the	O
company	O
can	O
ramp	O
up	O
sales	O
.	O
The	O
list	O
of	O
PacketVideo	B-KEY
's	O
technology	O
partners	O
who	O
are	O
also	O
investors-and	O
who	O
have	O
pumped	O
more	O
than	O
$	O
121	O
million	O
into	O
the	O
company-includes	O
not	O
just	O
wireless	B-KEY
device	I-KEY
manufacturers	O
,	O
but	O
content	O
providers	O
and	O
semiconductor	O
vendors	O
,	O
all	O
of	O
whom	O
stand	O
to	O
benefit	O
by	O
increased	O
sales	O
of	O
handheld	O
wireless	O
terminals	O

A	O
typed	B-KEY
representation	I-KEY
for	O
HTML	O
and	O
XML	B-KEY
documents	I-KEY
in	O
Haskell	B-KEY
We	O
define	O
a	O
family	O
of	O
embedded	B-KEY
domain	I-KEY
specific	I-KEY
languages	I-KEY
for	O
generating	O
HTML	O
and	O
XML	B-KEY
documents	I-KEY
.	O
Each	O
language	O
is	O
implemented	O
as	O
a	O
combinator	B-KEY
library	I-KEY
in	O
Haskell	B-KEY
.	O
The	O
generated	O
HTML/XML	O
documents	O
are	O
guaranteed	O
to	O
be	O
well-formed	O
.	O
In	O
addition	O
,	O
each	O
library	O
can	O
guarantee	O
that	O
the	O
generated	O
documents	O
are	O
valid	O
XML	B-KEY
documents	I-KEY
to	O
a	O
certain	O
extent	O
-LRB-	O
for	O
HTML	O
only	O
a	O
weaker	O
guarantee	O
is	O
possible	O
-RRB-	O
.	O
On	O
top	O
of	O
the	O
libraries	O
,	O
Haskell	B-KEY
serves	O
as	O
a	O
meta	B-KEY
language	I-KEY
to	O
define	O
parameterized	B-KEY
documents	I-KEY
,	O
to	O
map	O
structured	O
documents	O
to	O
HTML/XML	O
,	O
to	O
define	O
conditional	B-KEY
content	I-KEY
,	O
or	O
to	O
define	O
entire	O
Web	B-KEY
sites	I-KEY
.	O
The	O
combinator	B-KEY
libraries	I-KEY
support	O
element-transforming	B-KEY
style	I-KEY
,	O
a	O
programming	O
style	O
that	O
allows	O
programs	O
to	O
have	O
a	O
visual	O
appearance	O
similar	O
to	O
HTML/XML	O
documents	O
,	O
without	O
modifying	O
the	O
syntax	O
of	O
Haskell	O

A	O
strategy	O
for	O
a	O
payoff-switching	O
differential	B-KEY
game	I-KEY
based	O
on	O
fuzzy	O
reasoning	O
In	O
this	O
paper	O
,	O
a	O
new	O
concept	O
of	O
a	O
payoff-switching	O
differential	B-KEY
game	I-KEY
is	O
introduced	O
.	O
In	O
this	O
new	O
game	O
,	O
any	O
one	O
player	O
at	O
any	O
time	O
may	O
have	O
several	O
choices	O
of	O
payoffs	O
for	O
the	O
future	O
.	O
Moreover	O
,	O
the	O
payoff-switching	O
process	O
,	O
including	O
the	O
time	O
of	O
payoff	B-KEY
switching	I-KEY
and	O
the	O
outcome	B-KEY
payoff	I-KEY
,	O
of	O
any	O
one	O
player	O
is	O
unknown	O
to	O
the	O
other	O
.	O
Indeed	O
,	O
the	O
overall	O
payoff	O
,	O
which	O
is	O
a	O
sequence	O
of	O
several	O
payoffs	O
,	O
is	O
unknown	O
until	O
the	O
game	O
ends	O
.	O
An	O
algorithm	O
for	O
determining	O
a	O
reasoning	B-KEY
strategy	I-KEY
based	O
on	O
fuzzy	B-KEY
reasoning	I-KEY
is	O
proposed	O
.	O
In	O
this	O
algorithm	O
,	O
the	O
fuzzy	O
theory	O
is	O
used	O
to	O
estimate	O
the	O
behavior	O
of	O
one	O
player	O
during	O
a	O
past	O
time	O
interval	O
.	O
By	O
deriving	O
two	O
fuzzy	B-KEY
matrices	I-KEY
GSM	O
,	O
game	B-KEY
similarity	I-KEY
matrix	I-KEY
,	O
and	O
VGSM	O
,	O
variation	O
of	O
GSM	O
,	O
the	O
behavior	O
of	O
the	O
player	O
can	O
be	O
quantified	O
.	O
Two	O
weighting	B-KEY
vectors	I-KEY
are	O
selected	O
to	O
weight	O
the	O
relative	O
importance	O
of	O
the	O
player	O
's	O
behavior	O
at	O
each	O
past	O
time	O
instant	O
.	O
Finally	O
a	O
simple	O
fuzzy	B-KEY
inference	I-KEY
rule	O
is	O
adopted	O
to	O
generate	O
a	O
linear	O
reasoning	B-KEY
strategy	I-KEY
.	O
The	O
advantage	O
of	O
this	O
algorithm	O
is	O
that	O
it	O
provides	O
a	O
flexible	O
way	O
for	O
differential	B-KEY
game	I-KEY
specialists	O
to	O
convert	O
their	O
knowledge	O
into	O
a	O
``	O
reasonable	O
''	O
strategy	O
.	O
A	O
practical	O
example	O
of	O
guarding	O
three	O
territories	O
is	O
given	O
to	O
illustrate	O
our	O
main	O
ideas	O

Application	O
of	O
an	O
internally	O
consistent	O
material	O
model	O
to	O
determine	O
the	O
effect	O
of	O
tool	B-KEY
edge	I-KEY
geometry	I-KEY
in	O
orthogonal	O
machining	O
It	O
is	O
well	O
known	O
that	O
the	O
edge	B-KEY
geometry	I-KEY
of	O
a	O
cutting	B-KEY
tool	I-KEY
affects	O
the	O
forces	O
measured	O
in	O
metal	B-KEY
cutting	I-KEY
.	O
Two	O
experimental	O
methods	O
have	O
been	O
suggested	O
in	O
the	O
past	O
to	O
extract	O
the	O
ploughing	O
-LRB-	O
non-cutting	O
-RRB-	O
component	O
from	O
the	O
total	O
measured	O
force	O
:	O
-LRB-	O
1	O
-RRB-	O
the	O
extrapolation	B-KEY
approach	O
,	O
and	O
-LRB-	O
2	O
-RRB-	O
the	O
dwell	B-KEY
force	I-KEY
technique	O
.	O
This	O
study	O
reports	O
the	O
behavior	O
of	O
zinc	B-KEY
during	O
orthogonal	B-KEY
machining	I-KEY
using	O
tools	O
of	O
controlled	O
edge	O
radius	O
.	O
Applications	O
of	O
both	O
the	O
extrapolation	B-KEY
and	O
dwell	O
approaches	O
show	O
that	O
neither	O
produces	O
an	O
analysis	O
that	O
yields	O
a	O
material	O
response	O
consistent	O
with	O
the	O
known	O
behavior	O
of	O
zinc	B-KEY
.	O
Further	O
analysis	O
shows	O
that	O
the	O
edge	B-KEY
geometry	I-KEY
modifies	O
the	O
shear	O
zone	O
of	O
the	O
material	O
and	O
thereby	O
modifies	O
the	O
forces	O
.	O
When	O
analyzed	O
this	O
way	O
,	O
the	O
measured	O
force	O
data	O
yield	O
the	O
expected	O
material	O
response	O
without	O
requiring	O
recourse	O
to	O
an	O
additional	O
ploughing	B-KEY
component	I-KEY

Equilibrium	O
swelling	O
and	O
kinetics	O
of	O
pH-responsive	B-KEY
hydrogels	I-KEY
:	O
models	O
,	O
experiments	O
,	O
and	O
simulations	O
The	O
widespread	O
application	O
of	O
ionic	B-KEY
hydrogels	I-KEY
in	O
a	O
number	O
of	O
applications	O
like	O
control	O
of	O
microfluidic	B-KEY
flow	I-KEY
,	O
development	O
of	O
muscle-like	B-KEY
actuators	I-KEY
,	O
filtration/separation	B-KEY
and	O
drug	B-KEY
delivery	I-KEY
makes	O
it	O
important	O
to	O
properly	O
understand	O
these	O
materials	O
.	O
Understanding	O
hydrogel	O
properties	O
is	O
also	O
important	O
from	O
the	O
standpoint	O
of	O
their	O
similarity	O
to	O
many	O
biological	O
tissues	O
.	O
Typically	O
,	O
gel	B-KEY
size	I-KEY
is	O
sensitive	O
to	O
outer	O
solution	O
pH	O
and	O
salt	O
concentration	O
.	O
In	O
this	O
paper	O
,	O
we	O
develop	O
models	O
to	O
predict	O
the	O
swelling/deswelling	B-KEY
of	O
hydrogels	O
in	O
buffered	B-KEY
pH	I-KEY
solutions	I-KEY
.	O
An	O
equilibrium	B-KEY
model	I-KEY
has	O
been	O
developed	O
to	O
predict	O
the	O
degree	O
of	O
swelling	O
of	O
the	O
hydrogel	O
at	O
a	O
given	O
pH	O
and	O
salt	O
concentration	O
in	O
the	O
solution	O
.	O
A	O
kinetic	O
model	O
has	O
been	O
developed	O
to	O
predict	O
the	O
rate	O
of	O
swelling	O
of	O
the	O
hydrogel	O
when	O
the	O
solution	O
pH	O
is	O
changed	O
.	O
Experiments	O
are	O
performed	O
to	O
characterize	O
the	O
mechanical	B-KEY
properties	I-KEY
of	O
the	O
hydrogel	O
in	O
different	O
pH	O
solutions	O
.	O
The	O
degree	O
of	O
swelling	O
as	O
well	O
as	O
the	O
rate	O
of	O
swelling	O
of	O
the	O
hydrogel	O
are	O
also	O
studied	O
through	O
experiments	O
.	O
The	O
simulations	O
are	O
compared	O
with	O
experimental	O
results	O
and	O
the	O
models	O
are	O
found	O
to	O
predict	O
the	O
swelling/deswelling	B-KEY
processes	O
accurately	O

Application	O
of	O
time-frequency	B-KEY
principal	I-KEY
component	I-KEY
analysis	I-KEY
to	O
text-independent	B-KEY
speaker	I-KEY
identification	I-KEY
We	O
propose	O
a	O
formalism	O
,	O
called	O
vector	B-KEY
filtering	I-KEY
of	O
spectral	B-KEY
trajectories	I-KEY
,	O
that	O
allows	O
the	O
integration	O
of	O
a	O
number	O
of	O
speech	B-KEY
parameterization	I-KEY
approaches	O
-LRB-	O
cepstral	B-KEY
analysis	I-KEY
,	O
Delta	O
and	O
Delta	B-KEY
Delta	I-KEY
parameterizations	I-KEY
,	O
auto-regressive	O
vector	O
modeling	O
,	O
...	O
-RRB-	O
under	O
a	O
common	O
formalism	O
.	O
We	O
then	O
propose	O
a	O
new	O
filtering	O
,	O
called	O
contextual	B-KEY
principal	I-KEY
components	I-KEY
-LRB-	O
CPC	O
-RRB-	O
or	O
time-frequency	O
principal	O
components	O
-LRB-	O
TFPC	O
-RRB-	O
.	O
This	O
filtering	O
consists	O
in	O
extracting	O
the	O
principal	O
components	O
of	O
the	O
contextual	B-KEY
covariance	I-KEY
matrix	I-KEY
,	O
which	O
is	O
the	O
covariance	O
matrix	O
of	O
a	O
sequence	O
of	O
vectors	O
expanded	O
by	O
their	O
context	O
.	O
We	O
apply	O
this	O
new	O
filtering	O
in	O
the	O
framework	O
of	O
closed-set	B-KEY
speaker	I-KEY
identification	I-KEY
,	O
using	O
a	O
subset	O
of	O
the	O
POLYCOST	B-KEY
database	I-KEY
.	O
When	O
using	O
speaker-dependent	O
TFPC	O
filters	O
,	O
our	O
results	O
show	O
a	O
relative	O
improvement	O
of	O
approximately	O
20	O
%	O
compared	O
to	O
the	O
use	O
of	O
the	O
classical	O
cepstral	B-KEY
coefficients	I-KEY
augmented	O
by	O
their	O
Delta	O
-	O
coefficients	O
,	O
which	O
is	O
significantly	O
better	O
with	O
a	O
90	O
%	O
confidence	B-KEY
level	I-KEY

Pipelined	B-KEY
broadcast	I-KEY
with	O
enhanced	B-KEY
wormhole	I-KEY
routers	I-KEY
This	O
paper	O
proposes	O
a	O
pipelined	B-KEY
broadcast	I-KEY
that	O
broadcasts	O
a	O
message	O
of	O
size	O
m	O
in	O
O	O
-LRB-	O
m	O
+	O
n-1	O
-RRB-	O
time	O
in	O
an	O
n-dimensional	B-KEY
hypercube	I-KEY
.	O
It	O
is	O
based	O
on	O
the	O
replication	B-KEY
tree	I-KEY
,	O
which	O
is	O
derived	O
from	O
reachable	B-KEY
sets	I-KEY
.	O
It	O
has	O
greatly	O
improved	O
performance	B-KEY
compared	O
to	O
Ho-Kao	O
's	O
-LRB-	O
1995	O
-RRB-	O
algorithm	O
with	O
the	O
time	O
of	O
O	O
-LRB-	O
m	O
-LSB-	O
n/log	O
-LRB-	O
n	O
+1	O
-RRB-	O
-RSB-	O
-RRB-	O
.	O
The	O
communication	O
in	O
the	O
broadcast	O
uses	O
an	O
all-port	B-KEY
wormhole	I-KEY
router	I-KEY
with	O
message	B-KEY
replication	I-KEY
capability	I-KEY
.	O
This	O
paper	O
includes	O
the	O
algorithm	O
together	O
with	O
performance	B-KEY
comparisons	O
to	O
previous	O
schemes	O
in	O
a	O
practical	O
implementation	O

Estimation	O
of	O
trifocal	O
tensor	O
using	O
GMM	B-KEY
A	O
novel	O
estimation	O
of	O
a	O
trifocal	O
tensor	O
based	O
on	O
the	O
Gaussian	B-KEY
mixture	I-KEY
model	I-KEY
-LRB-	O
GMM	B-KEY
-RRB-	O
is	O
presented	O
.	O
The	O
mixture	O
model	O
is	O
built	O
assuming	O
that	O
the	O
residuals	O
of	O
inliers	B-KEY
and	O
outliers	B-KEY
belong	O
to	O
different	O
Gaussian	B-KEY
distributions	I-KEY
.	O
The	O
Bayesian	B-KEY
rule	I-KEY
is	O
then	O
employed	O
to	O
detect	O
the	O
inliers	B-KEY
for	O
re-estimation	O
.	O
Experiments	O
show	O
that	O
the	O
presented	O
method	O
is	O
more	O
precise	O
and	O
relatively	O
unaffected	O
by	O
outliers	B-KEY

Global	B-KEY
action	I-KEY
rules	I-KEY
in	O
distributed	O
knowledge	O
systems	O
Previously	O
Z.	O
Ras	O
and	O
J.M.	O
Zytkow	O
-LRB-	O
2000	O
-RRB-	O
introduced	O
and	O
investigated	O
query	O
answering	B-KEY
system	I-KEY
based	I-KEY
on	O
distributed	O
knowledge	B-KEY
mining	I-KEY
.	I-KEY
The	O
notion	O
of	O
an	O
action	B-KEY
rule	I-KEY
was	O
introduced	O
by	O
Z.	O
Ras	O
and	O
A.	O
Wieczorkowska	O
-LRB-	O
2000	O
-RRB-	O
and	O
its	O
application	O
domain	O
e-business	O
was	O
taken	O
.	O
In	O
this	O
paper	O
,	O
we	O
generalize	O
the	O
notion	O
of	O
action	B-KEY
rules	I-KEY
in	O
a	O
similar	O
way	O
to	O
handling	O
global	O
queries	O
.	O
Mainly	O
,	O
when	O
values	O
of	O
attributes	B-KEY
for	O
a	O
given	O
customer	O
,	O
used	O
in	O
action	B-KEY
rules	I-KEY
,	O
can	O
not	O
be	O
easily	O
changed	O
by	O
business	O
user	O
,	O
definitions	O
of	O
these	O
attributes	B-KEY
are	O
extracted	O
from	O
other	O
sites	O
of	O
a	O
distributed	O
knowledge	O
system	O
.	O
To	O
be	O
more	O
precise	O
,	O
attributes	B-KEY
at	O
every	O
site	O
of	O
a	O
distributed	O
knowledge	O
system	O
are	O
divided	O
into	O
two	O
sets	O
:	O
stable	O
and	O
flexible	O
.	O
Values	O
of	O
flexible	O
attributes	B-KEY
,	O
for	O
a	O
given	O
consumer	O
,	O
sometime	O
can	O
be	O
changed	O
and	O
this	O
change	O
can	O
be	O
influenced	O
and	O
controlled	O
by	O
a	O
business	O
user	O
.	O
However	O
,	O
some	O
of	O
these	O
changes	O
-LRB-	O
for	O
instance	O
to	O
the	O
attribute	B-KEY
``	O
profit	O
'	O
-RRB-	O
can	O
not	O
be	O
done	O
directly	O
to	O
a	O
chosen	O
attribute	B-KEY
.	O
In	O
this	O
case	O
,	O
definitions	O
of	O
such	O
an	O
attribute	B-KEY
in	O
terms	O
of	O
other	O
attributes	B-KEY
have	O
to	O
be	O
learned	O
.	O
These	O
new	O
definitions	O
are	O
used	O
to	O
construct	O
action	B-KEY
rules	I-KEY
showing	O
what	O
changes	O
in	O
values	O
of	O
flexible	O
attributes	B-KEY
,	O
for	O
a	O
given	O
consumer	O
,	O
are	O
needed	O
in	O
order	O
to	O
re-classify	O
this	O
consumer	O
the	O
way	O
business	O
user	O
wants	O
.	O
But	O
,	O
business	O
user	O
may	O
be	O
either	O
unable	O
or	O
unwilling	O
to	O
proceed	O
with	O
actions	O
leading	O
to	O
such	O
changes	O
.	O
In	O
all	O
such	O
cases	O
we	O
may	O
search	O
for	O
definitions	O
of	O
these	O
flexible	O
attributes	B-KEY
looking	O
at	O
either	O
local	O
or	O
remote	O
sites	O
for	O
help	O

Presenting-a	O
better	O
mousetrap	O
-LSB-	O
Leeza	B-KEY
outboard	B-KEY
video	I-KEY
signal	I-KEY
processor	I-KEY
-RSB-	O
Scaling	O
interlaced	O
video	O
to	O
match	O
high-resolution	O
plasma	O
,	O
LCD	O
,	O
and	O
DLP	B-KEY
displays	I-KEY
is	O
a	O
tough	O
job	O
,	O
but	O
Key	O
Digital	O
's	O
Leeza	B-KEY
is	O
zip	O
to	O
the	O
tack	O
.	O
And	O
it	O
's	O
digitally	O
bilingual	O
,	O
too	O
.	O
There	O
's	O
no	O
question	O
that	O
outboard	B-KEY
video	I-KEY
signal	I-KEY
processors	I-KEY
like	O
Leeza	B-KEY
help	O
overcome	O
the	O
inherent	O
limitations	O
of	O
fixed-pixel	B-KEY
displays	I-KEY
.	O
Being	O
able	O
to	O
match	O
a	O
native	O
display	O
rate	O
with	O
heavily	B-KEY
processed	I-KEY
video	I-KEY
makes	O
the	O
viewing	O
experience	O
much	O
more	O
enjoyable	O
.	O
But	O
it	O
seemed	O
that	O
70	O
%	O
of	O
the	O
improvement	O
in	O
image	O
quality	O
came	O
from	O
using	O
a	O
digital	O
interface	O
to	O
the	O
DVD	O
player	O
,	O
as	O
most	O
noise	O
and	O
picture	O
artifacts	O
are	O
introduced	O
in	O
the	O
analog	O
video	O
encoding	O
process	O

Some	O
properties	O
of	O
Hadamard	B-KEY
matrices	I-KEY
coming	O
from	O
dihedral	B-KEY
groups	I-KEY
H.	O
Kimura	O
-LRB-	O
1996	O
-RRB-	O
introduced	O
a	O
method	O
to	O
construct	O
Hadamard	B-KEY
matrices	I-KEY
of	O
degree	O
8n	O
+	O
4	O
from	O
the	O
dihedral	B-KEY
group	I-KEY
of	O
order	O
2n	O
.	O
In	O
this	O
paper	O
we	O
study	O
some	O
properties	O
of	O
this	O
construction	O

Weighted	O
energy	O
linear	O
quadratic	O
regulator	O
vibration	B-KEY
control	I-KEY
of	O
piezoelectric	O
composite	O
plates	O
In	O
this	O
paper	O
on	O
finite	B-KEY
element	I-KEY
linear	I-KEY
quadratic	I-KEY
regulator	I-KEY
-LRB-	O
LQR	O
-RRB-	O
vibration	B-KEY
control	I-KEY
of	O
smart	B-KEY
piezoelectric	I-KEY
composite	I-KEY
plates	I-KEY
,	O
we	O
propose	O
the	O
use	O
of	O
the	O
total	B-KEY
weighted	I-KEY
energy	I-KEY
method	O
to	O
select	O
the	O
weighting	B-KEY
matrices	I-KEY
.	O
By	O
constructing	O
the	O
optimal	B-KEY
performance	I-KEY
function	I-KEY
as	O
a	O
relative	O
measure	O
of	O
the	O
total	B-KEY
kinetic	I-KEY
energy	I-KEY
,	O
strain	B-KEY
energy	I-KEY
and	O
input	O
energy	O
of	O
the	O
system	O
,	O
only	O
three	O
design	O
variables	O
need	O
to	O
be	O
considered	O
to	O
achieve	O
a	O
balance	O
between	O
the	O
desired	O
higher	O
damping	B-KEY
effect	I-KEY
and	O
lower	O
input	O
cost	O
.	O
Modal	B-KEY
control	I-KEY
analysis	I-KEY
is	O
used	O
to	O
interpret	O
the	O
effects	O
of	O
three	O
energy	O
weight	O
factors	O
on	O
the	O
damping	B-KEY
ratios	I-KEY
and	O
modal	O
voltages	O
and	O
it	O
is	O
shown	O
that	O
the	O
modal	O
damping	B-KEY
effect	I-KEY
will	O
increase	O
with	O
the	O
kinetic	O
energy	O
weight	O
factor	O
,	O
approaching	O
square	O
root	O
-LRB-	O
2/2	O
-RRB-	O
as	O
the	O
strain	B-KEY
energy	I-KEY
weight	O
factor	O
increases	O
and	O
decrease	O
with	O
the	O
input	O
energy	O
weight	O
factor	O
.	O
Numerical	B-KEY
results	I-KEY
agree	O
well	O
with	O
those	O
from	O
the	O
modal	B-KEY
control	I-KEY
analysis	I-KEY
.	O
Since	O
the	O
control	O
problem	O
is	O
simplified	O
to	O
three	O
design	O
variables	O
only	O
,	O
the	O
computational	B-KEY
cost	I-KEY
will	O
be	O
greatly	O
reduced	O
and	O
a	O
more	O
accurate	O
structural	B-KEY
control	I-KEY
analysis	I-KEY
becomes	O
more	O
attractive	O
for	O
large	O
systems	O

On	O
M/D/1	B-KEY
queue	I-KEY
with	O
deterministic	B-KEY
server	I-KEY
vacations	I-KEY
We	O
study	O
a	O
single	O
server	O
vacation	O
queue	O
with	O
Poisson	B-KEY
arrivals	I-KEY
,	O
deterministic	B-KEY
service	I-KEY
of	O
constant	O
duration	O
b	O
-LRB-	O
>	O
0	O
-RRB-	O
and	O
deterministic	B-KEY
vacations	I-KEY
of	O
constant	O
duration	O
d	O
-LRB-	O
>	O
0	O
-RRB-	O
and	O
designate	O
this	O
model	O
as	O
M/D/D	O
/	O
1	O
.	O
After	O
completion	O
of	O
each	O
service	O
,	O
the	O
server	O
may	O
take	O
a	O
vacation	O
with	O
probability	O
p	O
or	O
may	O
continue	O
working	O
in	O
the	O
system	O
with	O
probability	O
1	O
-	O
p	O
.	O
We	O
obtain	O
time-dependent	O
as	O
well	O
as	O
steady	B-KEY
state	I-KEY
probability	I-KEY
generation	I-KEY
functions	I-KEY
for	O
the	O
number	O
in	O
the	O
system	O
.	O
For	O
the	O
steady	O
state	O
we	O
obtain	O
explicitly	O
the	O
mean	B-KEY
number	I-KEY
and	O
the	O
mean	B-KEY
waiting	I-KEY
time	I-KEY
for	O
the	O
system	O
and	O
for	O
the	O
queue	O
.	O
All	O
known	O
results	O
of	O
the	O
M/D/1	B-KEY
queue	I-KEY
are	O
derived	O
as	O
a	O
special	O
case	O
.	O
Finally	O
,	O
a	O
numerical	O
illustration	O
is	O
discussed	O

Taking	O
it	O
to	O
the	O
max	O
-LSB-	O
ventilation	O
systems	O
-RSB-	O
Raising	O
the	O
volumetric	B-KEY
air	I-KEY
supply	I-KEY
rate	I-KEY
is	O
one	O
way	O
of	O
increasing	O
the	O
cooling	B-KEY
capacity	I-KEY
of	O
displacement	B-KEY
ventilation	I-KEY
systems	I-KEY
.	O
David	O
Butler	O
and	O
Michael	O
Swainson	O
explore	O
how	O
different	O
types	O
of	O
diffusers	B-KEY
can	O
help	O
make	O
this	O
work	O

Some	O
recent	O
advances	O
in	O
validated	B-KEY
methods	I-KEY
for	O
IVPs	O
for	O
ODEs	O
Compared	O
to	O
standard	O
numerical	O
methods	O
for	O
initial	B-KEY
value	I-KEY
problems	I-KEY
-LRB-	O
IVPs	O
-RRB-	O
for	O
ordinary	B-KEY
differential	I-KEY
equations	I-KEY
-LRB-	O
ODEs	O
-RRB-	O
,	O
validated	B-KEY
methods	I-KEY
-LRB-	O
often	O
called	O
interval	B-KEY
methods	I-KEY
-RRB-	O
for	O
IVPs	O
for	O
ODEs	O
have	O
two	O
important	O
advantages	O
:	O
if	O
they	O
return	O
a	O
solution	O
to	O
a	O
problem	O
,	O
then	O
-LRB-	O
1	O
-RRB-	O
the	O
problem	O
is	O
guaranteed	O
to	O
have	O
a	O
unique	O
solution	O
,	O
and	O
-LRB-	O
2	O
-RRB-	O
an	O
enclosure	O
of	O
the	O
true	O
solution	O
is	O
produced	O
.	O
We	O
present	O
a	O
brief	O
overview	O
of	O
interval	B-KEY
Taylor	I-KEY
series	I-KEY
-LRB-	O
ITS	O
-RRB-	O
methods	O
for	O
IVPs	O
for	O
ODEs	O
and	O
discuss	O
some	O
recent	O
advances	O
in	O
the	O
theory	O
of	O
validated	B-KEY
methods	I-KEY
for	O
IVPs	O
for	O
ODEs	O
.	O
In	O
particular	O
,	O
we	O
discuss	O
an	O
interval	O
Hermite-Obreschkoff	O
-LRB-	O
IHO	O
-RRB-	O
scheme	O
for	O
computing	O
rigorous	O
bounds	O
on	O
the	O
solution	O
of	O
an	O
IVP	O
for	O
an	O
ODE	O
,	O
the	O
stability	O
of	O
ITS	O
and	O
IHO	O
methods	O
,	O
and	O
a	O
new	O
perspective	O
on	O
the	O
wrapping	B-KEY
effect	I-KEY
,	O
where	O
we	O
interpret	O
the	O
problem	O
of	O
reducing	O
the	O
wrapping	B-KEY
effect	I-KEY
as	O
one	O
of	O
finding	O
a	O
more	O
stable	O
scheme	O
for	O
advancing	O
the	O
solution	O

JPEG2000	B-KEY
:	O
standard	O
for	O
interactive	B-KEY
imaging	I-KEY
JPEG2000	B-KEY
is	O
the	O
latest	O
image	B-KEY
compression	I-KEY
standard	O
to	O
emerge	O
from	O
the	O
Joint	B-KEY
Photographic	I-KEY
Experts	I-KEY
Group	I-KEY
-LRB-	O
JPEG	O
-RRB-	O
working	O
under	O
the	O
auspices	O
of	O
the	O
International	B-KEY
Standards	I-KEY
Organization	I-KEY
.	O
Although	O
the	O
new	O
standard	O
does	O
offer	O
superior	O
compression	O
performance	O
to	O
JPEG	O
,	O
JPEG2000	B-KEY
provides	O
a	O
whole	O
new	O
way	O
of	O
interacting	O
with	O
compressed	O
imagery	O
in	O
a	O
scalable	O
and	O
interoperable	O
fashion	O
.	O
This	O
paper	O
provides	O
a	O
tutorial-style	O
review	B-KEY
of	O
the	O
new	O
standard	O
,	O
explaining	O
the	O
technology	O
on	O
which	O
it	O
is	O
based	O
and	O
drawing	O
comparisons	O
with	O
JPEG	O
and	O
other	O
compression	O
standards	O
.	O
The	O
paper	O
also	O
describes	O
new	O
work	O
,	O
exploiting	O
the	O
capabilities	O
of	O
JPEG2000	B-KEY
in	O
client-server	B-KEY
systems	I-KEY
for	O
efficient	O
interactive	O
browsing	O
of	O
images	O
over	O
the	O
Internet	O

On-line	B-KEY
robust	I-KEY
processing	I-KEY
techniques	I-KEY
for	O
elimination	O
of	O
measurement	O
drop-out	O
When	O
processing	O
measurement	O
data	O
,	O
it	O
is	O
usually	O
assumed	O
that	O
some	O
amount	O
of	O
normally	B-KEY
distributed	I-KEY
measurement	I-KEY
noise	I-KEY
is	O
present	O
.	O
In	O
some	O
situations	O
,	O
outliers	O
are	O
present	O
in	O
the	O
measurements	O
and	O
consequently	O
the	O
noise	O
is	O
far	O
from	O
normally	O
distributed	O
.	O
In	O
this	O
case	O
classical	B-KEY
least-squares	I-KEY
procedures	I-KEY
for	O
estimating	O
Fourier	B-KEY
spectra	I-KEY
-LRB-	O
or	O
derived	O
quantities	O
like	O
the	O
frequency	B-KEY
response	I-KEY
function	I-KEY
-RRB-	O
can	O
give	O
results	O
which	O
are	O
inaccurate	O
or	O
even	O
useless	O
.	O
In	O
this	O
paper	O
,	O
a	O
novel	O
technique	O
for	O
the	O
on-line	O
processing	O
of	O
measurement	B-KEY
outliers	I-KEY
will	O
be	O
proposed	O
.	O
Both	O
the	O
computation	B-KEY
speed	I-KEY
and	O
the	O
accuracy	O
of	O
the	O
technique	O
presented	O
will	O
be	O
compared	O
with	O
different	O
classical	O
approaches	O
for	O
handling	O
outliers	O
in	O
measurement	O
data	O
-LRB-	O
i.e.	O
filtering	O
techniques	O
,	O
outlier	O
rejection	O
techniques	O
and	O
robust	B-KEY
regression	I-KEY
techniques	O
-RRB-	O
.	O
In	O
particular	O
,	O
all	O
processing	O
techniques	O
will	O
be	O
validated	O
by	O
applying	O
them	O
to	O
the	O
problem	O
of	O
speckle	O
drop-out	O
in	O
optical	B-KEY
vibration	I-KEY
measurements	I-KEY
-LRB-	O
performed	O
with	O
a	O
laser	B-KEY
Doppler	I-KEY
vibrometer	I-KEY
-RRB-	O
,	O
which	O
typically	O
causes	O
outliers	O
in	O
the	O
measurements	O

Entangling	O
atoms	O
in	O
bad	B-KEY
cavities	I-KEY
We	O
propose	O
a	O
method	O
to	O
produce	O
entangled	B-KEY
spin	I-KEY
squeezed	I-KEY
states	I-KEY
of	O
a	O
large	O
number	O
of	O
atoms	O
inside	O
an	O
optical	B-KEY
cavity	I-KEY
.	O
By	O
illuminating	O
the	O
atoms	O
with	O
bichromatic	O
light	O
,	O
the	O
coupling	B-KEY
to	O
the	O
cavity	O
induces	O
pairwise	B-KEY
exchange	I-KEY
of	O
excitations	B-KEY
which	O
entangles	O
the	O
atoms	O
.	O
Unlike	O
most	O
proposals	O
for	O
entangling	O
atoms	O
by	O
cavity	B-KEY
QED	I-KEY
,	O
our	O
proposal	O
does	O
not	O
require	O
the	O
strong	O
coupling	B-KEY
regime	O
g/sup	O
2	O
/	O
/	O
kappa	O
Gamma	O
>>	O
1	O
,	O
where	O
g	O
is	O
the	O
atom	O
cavity	O
coupling	O
strength	O
,	O
kappa	O
is	O
the	O
cavity	O
decay	O
rate	O
,	O
and	O
Gamma	O
is	O
the	O
decay	O
rate	O
of	O
the	O
atoms	O
.	O
In	O
this	O
work	O
the	O
important	O
parameter	O
is	O
Ng/sup	O
2	O
/	O
/	O
kappa	O
Gamma	O
,	O
where	O
N	O
is	O
the	O
number	O
of	O
atoms	O
,	O
and	O
our	O
proposal	O
permits	O
the	O
production	O
of	O
entanglement	O
in	O
bad	B-KEY
cavities	I-KEY
as	O
long	O
as	O
they	O
contain	O
a	O
large	O
number	O
of	O
atoms	O

Orthogonal	B-KEY
decompositions	I-KEY
of	O
complete	B-KEY
digraphs	I-KEY
A	O
family	O
G	O
of	O
isomorphic	B-KEY
copies	I-KEY
of	O
a	O
given	O
digraph	O
G	O
is	O
said	O
to	O
be	O
an	O
orthogonal	B-KEY
decomposition	I-KEY
of	O
the	O
complete	B-KEY
digraph	I-KEY
D/sub	O
n	O
/	O
by	O
G	O
,	O
if	O
every	O
arc	O
of	O
D/sub	O
n	O
/	O
belongs	O
to	O
exactly	O
one	O
member	O
of	O
G	O
and	O
the	O
union	O
of	O
any	O
two	O
different	O
elements	O
from	O
G	O
contains	O
precisely	O
one	O
pair	O
of	O
reverse	O
arcs	O
.	O
Given	O
a	O
digraph	O
h	O
,	O
an	O
h	O
family	O
mh	O
is	O
the	O
vertex-disjoint	B-KEY
union	I-KEY
of	O
m	O
copies	O
of	O
h	O
.	O
In	O
this	O
paper	O
,	O
we	O
consider	O
orthogonal	B-KEY
decompositions	I-KEY
by	O
h-families	O
.	O
Our	O
objective	O
is	O
to	O
prove	O
the	O
existence	O
of	O
such	O
an	O
orthogonal	B-KEY
decomposition	I-KEY
whenever	O
certain	O
necessary	B-KEY
conditions	I-KEY
hold	O
and	O
m	O
is	O
sufficiently	O
large	O

Banks	O
pin	O
their	O
back-office	O
hopes	O
on	O
successors	O
to	O
screen	O
scrapers	O
The	O
big	O
name	O
in	O
account	B-KEY
aggregation	I-KEY
has	O
been	O
Yodlee	B-KEY
,	O
based	O
in	O
Redwood	O
Shores	O
,	O
CA	O
.	O
It	O
pioneered	O
the	O
art	O
of	O
screen	B-KEY
scraping	I-KEY
,	O
or	O
pulling	O
data	O
off	O
Web	B-KEY
sites	I-KEY
and	O
aggregating	O
it	O
into	O
a	O
single	O
statement	O
.	O
That	O
data	O
,	O
however	O
,	O
is	O
a	O
snapshot	O
and	O
does	O
not	O
include	O
a	O
customer	O
's	O
investment	B-KEY
history	O
.	O
Also	O
,	O
because	O
Web	B-KEY
sites	I-KEY
update	O
data	O
at	O
different	O
times	O
,	O
scraping	O
them	O
can	O
provide	O
an	O
inaccurate	O
picture	O
of	O
a	O
customer	O
's	O
financial	O
situation	O
,	O
making	O
it	O
difficult	O
for	O
reps	O
seeking	O
to	O
provide	O
timely	O
and	O
accurate	O
advice	O
.	O
The	O
objective	O
is	O
to	O
access	O
both	O
fresh	O
and	O
historical	O
data	O
across	O
a	O
client	O
's	O
financial	O
spectrum	O
,	O
from	O
investments	B-KEY
to	O
checking	B-KEY
accounts	O
and	O
loans	B-KEY
to	O
insurance	B-KEY
policies	O
,	O
a	O
Complete	O
Customer	O
balance	O
sheet	O
.	O
At	O
least	O
two	O
technology	O
vendors	O
are	O
progressing	O
in	O
that	O
direction	O
,	O
each	O
coming	O
from	O
different	O
directions	O
.	O
One	O
is	O
Advent	B-KEY
,	O
based	O
in	O
San	O
Francisco	O
,	O
another	O
is	O
Fincentric	B-KEY
,	O
out	O
of	O
Vancouver	O

HPCVIEW	B-KEY
:	O
a	O
tool	O
for	O
top-down	B-KEY
analysis	I-KEY
of	O
node	B-KEY
performance	I-KEY
It	O
is	O
increasingly	O
difficult	O
for	O
complex	B-KEY
scientific	I-KEY
programs	I-KEY
to	O
attain	O
a	O
significant	O
fraction	O
of	O
peak	B-KEY
performance	I-KEY
on	O
systems	O
that	O
are	O
based	O
on	O
microprocessors	O
with	O
substantial	O
instruction-level	B-KEY
parallelism	I-KEY
and	O
deep	B-KEY
memory	I-KEY
hierarchies	I-KEY
.	O
Despite	O
this	O
trend	O
,	O
performance	B-KEY
analysis	I-KEY
and	O
tuning	O
tools	O
are	O
still	O
not	O
used	O
regularly	O
by	O
algorithm	O
and	O
application	O
designers	O
.	O
To	O
a	O
large	O
extent	O
,	O
existing	O
performance	O
tools	O
fail	O
to	O
meet	O
many	O
user	O
needs	O
and	O
are	O
cumbersome	O
to	O
use	O
.	O
To	O
address	O
these	O
issues	O
,	O
we	O
developed	O
HPCVIEW	B-KEY
-	O
a	O
toolkit	O
for	O
combining	O
multiple	O
sets	O
of	O
program	O
profile	O
data	O
,	O
correlating	O
the	O
data	O
with	O
source	B-KEY
code	I-KEY
,	O
and	O
generating	O
a	O
database	O
that	O
can	O
be	O
analyzed	O
anywhere	O
with	O
a	O
commodity	B-KEY
Web	I-KEY
browser	I-KEY
.	O
We	O
argue	O
that	O
HPCVIEW	B-KEY
addresses	O
many	O
of	O
the	O
issues	O
that	O
have	O
limited	O
the	O
usability	O
and	O
the	O
utility	O
of	O
most	O
existing	O
tools	O
.	O
We	O
originally	O
built	O
HPCVIEW	B-KEY
to	O
facilitate	O
our	O
own	O
work	O
on	O
data	B-KEY
layout	I-KEY
and	O
optimizing	B-KEY
compilers	I-KEY
.	O
Now	O
,	O
in	O
addition	O
to	O
daily	O
use	O
within	O
our	O
group	O
,	O
HPCVIEW	B-KEY
is	O
being	O
used	O
by	O
several	O
code	O
development	O
teams	O
in	O
DoD	O
and	O
DoE	O
laboratories	O
as	O
well	O
as	O
at	O
NCSA	O

At	O
your	O
service	O
-LSB-	O
agile	B-KEY
businesses	I-KEY
-RSB-	O
Senior	O
software	O
executives	O
from	O
three	O
of	O
the	O
world	O
's	O
leading	O
software	B-KEY
companies	I-KEY
,	O
and	O
one	O
smaller	O
,	O
entrepreneurial	O
software	O
developer	O
,	O
explain	O
the	O
impact	O
that	O
web	B-KEY
services	I-KEY
,	O
business	B-KEY
process	I-KEY
management	I-KEY
and	O
integrated	B-KEY
application	I-KEY
architectures	I-KEY
are	O
having	O
on	O
their	O
product	O
development	O
plans	O
,	O
and	O
share	O
their	O
vision	O
of	O
the	O
roles	O
these	O
products	O
will	O
play	O
in	O
creating	O
agile	B-KEY
businesses	I-KEY

New	O
methods	O
for	O
oscillatory	B-KEY
problems	I-KEY
based	O
on	O
classical	B-KEY
codes	I-KEY
The	O
numerical	B-KEY
integration	I-KEY
of	O
differential	B-KEY
equations	I-KEY
with	O
oscillatory	B-KEY
solutions	I-KEY
is	O
a	O
very	O
common	O
problem	O
in	O
many	O
fields	O
of	O
the	O
applied	O
sciences	O
.	O
Some	O
methods	O
have	O
been	O
specially	O
devised	O
for	O
this	O
kind	O
of	O
problem	O
.	O
In	O
most	O
of	O
them	O
,	O
the	O
calculation	O
of	O
the	O
coefficients	O
needs	O
more	O
computational	O
effort	O
than	O
the	O
classical	B-KEY
codes	I-KEY
because	O
such	O
coefficients	O
depend	O
on	O
the	O
step-size	O
in	O
a	O
not	O
simple	O
manner	O
.	O
On	O
the	O
contrary	O
,	O
in	O
this	O
work	O
we	O
present	O
new	O
algorithms	O
specially	O
designed	O
for	O
perturbed	B-KEY
oscillators	I-KEY
whose	O
coefficients	O
have	O
a	O
simple	O
dependence	O
on	O
the	O
step-size	O
.	O
The	O
methods	O
obtained	O
are	O
competitive	O
when	O
comparing	O
with	O
classical	O
and	O
special	O
codes	O

Multivariable	O
H/sub	O
infinity	O
/	O
/	O
mu	O
feedback	B-KEY
control	O
design	O
for	O
high-precision	O
wafer	O
stage	O
motion	O
Conventional	O
PID-like	O
SISO	O
controllers	O
are	O
still	O
the	O
most	O
common	O
in	O
industry	O
,	O
but	O
with	O
performance	O
requirements	O
becoming	O
tighter	O
there	O
is	O
a	O
growing	O
need	O
for	O
advanced	O
controllers	O
.	O
For	O
the	O
positioning	O
devices	O
in	O
IC-manufacturing	O
,	O
plant	O
interaction	O
is	O
a	O
major	O
performance-limiting	O
factor	O
.	O
MIMO	O
control	O
can	O
be	O
invoked	O
to	O
tackle	O
this	O
problem	O
.	O
A	O
practically	O
feasible	O
procedure	O
is	O
presented	O
to	O
design	O
MIMO	O
feedback	B-KEY
controllers	O
for	O
electromechanical	O
positioning	O
devices	O
,	O
using	O
H/sub	O
infinity	O
/	O
/	O
mu	O
techniques	O
.	O
Weighting	B-KEY
filters	I-KEY
are	O
proposed	O
to	O
straightforwardly	O
and	O
effectively	O
impose	O
performance	O
and	O
uncertainty	O
specifications	O
.	O
Experiments	O
show	O
that	O
MIMO	O
control	O
can	O
considerably	O
improve	O
upon	O
the	O
performance	O
with	O
multiloop	O
SISO	O
control	O
.	O
Some	O
problems	O
are	O
highlighted	O
that	O
are	O
important	O
for	O
industrial	O
practice	O
,	O
but	O
lacking	O
a	O
workable	O
solution	O

Meshed	B-KEY
atlases	I-KEY
for	O
real-time	B-KEY
procedural	I-KEY
solid	I-KEY
texturing	I-KEY
We	O
describe	O
an	O
implementation	O
of	O
procedural	O
solid	O
texturing	O
that	O
uses	O
the	O
texture	B-KEY
atlas	I-KEY
,	O
a	O
one-to-one	B-KEY
mapping	I-KEY
from	O
an	O
object	O
's	O
surface	O
into	O
its	O
texture	B-KEY
space	I-KEY
.	O
The	O
method	O
uses	O
the	O
graphics	B-KEY
hardware	I-KEY
to	O
rasterize	B-KEY
the	O
solid	B-KEY
texture	I-KEY
coordinates	I-KEY
as	O
colors	B-KEY
directly	O
into	O
the	O
atlas	O
.	O
A	O
texturing	O
procedure	O
is	O
applied	O
per-pixel	O
to	O
the	O
texture	O
map	O
,	O
replacing	O
each	O
solid	B-KEY
texture	I-KEY
coordinate	I-KEY
with	O
its	O
corresponding	O
procedural	O
solid	O
texture	O
result	O
.	O
The	O
procedural	O
solid	O
texture	O
is	O
then	O
mapped	O
back	O
onto	O
the	O
object	B-KEY
surface	I-KEY
using	O
standard	O
texture	O
mapping	O
.	O
The	O
implementation	O
renders	B-KEY
procedural	O
solid	O
textures	O
in	O
real	O
time	O
,	O
and	O
the	O
user	O
can	O
design	O
them	O
interactively	O
.	O
The	O
quality	O
of	O
this	O
technique	O
depends	O
greatly	O
on	O
the	O
layout	O
of	O
the	O
texture	B-KEY
atlas	I-KEY
.	O
A	O
broad	O
survey	O
of	O
texture	B-KEY
atlas	I-KEY
schemes	O
is	O
used	O
to	O
develop	O
a	O
set	O
of	O
general	O
purpose	O
mesh	B-KEY
atlases	I-KEY
and	O
tools	O
for	O
measuring	O
their	O
effectiveness	O
at	O
distributing	O
as	O
many	O
available	O
texture	O
samples	O
as	O
evenly	O
across	O
the	O
surface	O
as	O
possible	O
.	O
The	O
main	O
contribution	O
of	O
this	O
paper	O
is	O
a	O
new	O
multiresolution	B-KEY
texture	I-KEY
atlas	I-KEY
.	O
It	O
distributes	O
all	O
available	O
texture	O
samples	O
in	O
a	O
nearly	O
uniform	O
distribution	O
.	O
This	O
multiresolution	B-KEY
texture	I-KEY
atlas	I-KEY
also	O
supports	O
MIP-mapped	O
minification	O
antialiasing	O
and	O
linear	O
magnification	O
filtering	O

Dynamic	B-KEY
spectrum	I-KEY
management	I-KEY
for	O
next-generation	O
DSL	O
systems	O
The	O
performance	O
of	O
DSL	O
systems	O
is	O
severely	O
constrained	O
by	O
crosstalk	O
due	O
to	O
the	O
electromagnetic	B-KEY
coupling	I-KEY
among	O
the	O
multiple	O
twisted	B-KEY
pairs	I-KEY
making	O
up	O
a	O
phone	B-KEY
cable	I-KEY
.	O
In	O
order	O
to	O
reduce	O
performance	O
loss	O
arising	O
from	O
crosstalk	O
,	O
DSL	O
systems	O
are	O
currently	O
designed	O
under	O
the	O
assumption	O
of	O
worst-case	O
crosstalk	O
scenarios	O
leading	O
to	O
overly	O
conservative	O
DSL	O
deployments	O
.	O
This	O
article	O
presents	O
a	O
new	O
paradigm	O
for	O
DSL	B-KEY
system	I-KEY
design	I-KEY
,	O
which	O
takes	O
into	O
account	O
the	O
multi-user	O
aspects	O
of	O
the	O
DSL	O
transmission	O
environment	O
.	O
Dynamic	B-KEY
spectrum	I-KEY
management	I-KEY
-LRB-	O
DSM	O
-RRB-	O
departs	O
from	O
the	O
current	O
design	O
philosophy	O
by	O
enabling	O
transceivers	B-KEY
to	O
autonomously	O
and	O
dynamically	O
optimize	O
their	O
communication	O
settings	O
with	O
respect	O
to	O
both	O
the	O
channel	O
and	O
the	O
transmissions	O
of	O
neighboring	O
systems	O
.	O
Along	O
with	O
this	O
distributed	B-KEY
optimization	I-KEY
,	O
when	O
an	O
additional	O
degree	O
of	O
coordination	O
becomes	O
available	O
for	O
future	O
DSL	O
deployment	O
,	O
DSM	O
will	O
allow	O
even	O
greater	O
improvement	O
in	O
DSL	O
performance	O
.	O
Implementations	O
are	O
readily	O
applicable	O
without	O
causing	O
any	O
performance	O
degradation	O
to	O
the	O
existing	O
DSLs	O
under	O
static	B-KEY
spectrum	I-KEY
management	I-KEY
.	O
After	O
providing	O
an	O
overview	O
of	O
the	O
DSM	O
concept	O
,	O
this	O
article	O
reviews	O
two	O
practical	O
DSM	O
methods	O
:	O
iterative	B-KEY
water-filling	I-KEY
,	O
an	O
autonomous	B-KEY
distributed	I-KEY
power	I-KEY
control	I-KEY
method	I-KEY
enabling	O
great	O
improvement	O
in	O
performance	O
,	O
which	O
can	O
be	O
implemented	O
through	O
software	B-KEY
options	I-KEY
in	O
some	O
existing	O
ADSL	O
and	O
VDSL	B-KEY
systems	I-KEY
;	O
and	O
vectored-DMT	B-KEY
,	O
a	O
coordinated	B-KEY
transmission/reception	I-KEY
technique	O
achieving	O
crosstalk-free	B-KEY
communication	I-KEY
for	O
DSL	O
systems	O
,	O
which	O
brings	O
within	O
reach	O
the	O
dream	O
of	O
providing	O
universal	B-KEY
Internet	I-KEY
access	I-KEY
at	O
speeds	O
close	O
to	O
100	O
Mb/s	O
to	O
500	O
m	O
on	O
1-2	O
lines	O
and	O
beyond	O
1	O
km	O
on	O
2-4	O
lines	O
.	O
DSM-capable	O
DSL	O
thus	O
enables	O
the	O
broadband	O
age	O

A	O
digital	B-KEY
fountain	I-KEY
approach	O
to	O
asynchronous	B-KEY
reliable	I-KEY
multicast	I-KEY
The	O
proliferation	O
of	O
applications	O
that	O
must	O
reliably	O
distribute	O
large	O
,	O
rich	O
content	O
to	O
a	O
vast	O
number	O
of	O
autonomous	B-KEY
receivers	I-KEY
motivates	O
the	O
design	O
of	O
new	O
multicast	O
and	O
broadcast	B-KEY
protocols	I-KEY
.	O
We	O
describe	O
an	O
ideal	O
,	O
fully	O
scalable	B-KEY
protocol	I-KEY
for	O
these	O
applications	O
that	O
we	O
call	O
a	O
digital	B-KEY
fountain	I-KEY
.	O
A	O
digital	B-KEY
fountain	I-KEY
allows	O
any	O
number	O
of	O
heterogeneous	B-KEY
receivers	I-KEY
to	O
acquire	O
content	O
with	O
optimal	B-KEY
efficiency	I-KEY
at	O
times	O
of	O
their	O
choosing	O
.	O
Moreover	O
,	O
no	O
feedback	O
channels	O
are	O
needed	O
to	O
ensure	O
reliable	O
delivery	O
,	O
even	O
in	O
the	O
face	O
of	O
high	B-KEY
loss	I-KEY
rates	I-KEY
.	O
We	O
develop	O
a	O
protocol	O
that	O
closely	O
approximates	O
a	O
digital	B-KEY
fountain	I-KEY
using	O
two	O
new	O
classes	O
of	O
erasure	B-KEY
codes	I-KEY
that	O
for	O
large	B-KEY
block	I-KEY
sizes	I-KEY
are	O
orders	O
of	O
magnitude	O
faster	O
than	O
standard	O
erasure	B-KEY
codes	I-KEY
.	O
We	O
provide	O
performance	B-KEY
measurements	I-KEY
that	O
demonstrate	O
the	O
feasibility	O
of	O
our	O
approach	O
and	O
discuss	O
the	O
design	O
,	O
implementation	O
,	O
and	O
performance	O
of	O
an	O
experimental	O
system	O

An	O
agent-oriented	O
and	O
service-oriented	B-KEY
environment	I-KEY
for	O
deploying	O
dynamic	B-KEY
distributed	I-KEY
systems	I-KEY
This	O
paper	O
presents	O
JASE	B-KEY
,	O
a	O
Java-based	O
Agent-oriented	O
and	O
Service-oriented	B-KEY
Environment	I-KEY
for	O
deploying	O
dynamic	O
distributed	O
systems	O
.	O
JASE	B-KEY
utilizes	O
two	O
important	O
concepts	O
in	O
the	O
field	O
of	O
distributed	O
computing	O
:	O
the	O
concept	O
of	O
services	O
and	O
remote	B-KEY
programming	I-KEY
with	O
mobile	B-KEY
agents	I-KEY
.	O
In	O
JASE	B-KEY
,	O
mobile	B-KEY
agents	I-KEY
are	O
used	O
to	O
support	O
applications	O
,	O
and	O
service	O
interface	O
agents	O
are	O
used	O
to	O
wrap	O
services	O
.	O
Service	O
inter-face	O
agents	O
can	O
dynamically	O
register	O
their	O
services	O
in	O
Service	O
Server	O
.	O
Mobile	B-KEY
agent	I-KEY
locates	O
a	O
specific	O
service	O
interface	O
agent	O
by	O
submitting	O
requests	O
to	O
the	O
Service	O
Server	O
with	O
descriptions	O
of	O
required	O
services	O
.	O
JASE	B-KEY
uses	O
XML	O
to	O
describe	O
both	O
service	O
descriptions	O
and	O
the	O
mobile	B-KEY
agent	I-KEY
's	O
queries	O
.	O
JASE	B-KEY
supports	O
two	O
kinds	O
of	O
communication	O
facility	O
:	O
tuple	O
space	O
and	O
asynchronous	O
messages	O
.	O
In	O
this	O
paper	O
,	O
the	O
design	O
and	O
implementation	O
of	O
JASE	B-KEY
are	O
described	O
.	O
An	O
application	O
shows	O
the	O
suitability	O
and	O
the	O
effectiveness	O
of	O
the	O
JASE	B-KEY
and	O
performance	B-KEY
evaluation	I-KEY
is	O
also	O
made	O
.	O
Finally	O
,	O
related	O
work	O
and	O
some	O
conclusions	O
are	O
given	O

Optimization	O
of	O
the	O
memory	B-KEY
weighting	I-KEY
function	I-KEY
in	O
stochastic	O
functional	O
self-organized	O
sorting	B-KEY
performed	O
by	O
a	O
team	O
of	O
autonomous	B-KEY
mobile	I-KEY
agents	I-KEY
The	O
activity	O
of	O
a	O
team	O
of	O
autonomous	B-KEY
mobile	I-KEY
agents	I-KEY
formed	O
by	O
identical	O
``	O
robot-like-ant	O
''	O
individuals	O
capable	O
of	O
performing	B-KEY
a	I-KEY
random	O
walk	O
through	O
an	O
environment	O
that	O
are	O
able	O
to	O
recognize	O
and	O
move	O
different	O
``	O
objects	O
''	O
is	O
modeled	O
.	O
The	O
emergent	O
desired	O
behavior	O
is	O
a	O
distributed	O
sorting	B-KEY
and	O
clustering	B-KEY
based	O
only	O
on	O
local	O
information	O
and	O
a	O
memory	O
register	O
that	O
records	O
the	O
past	O
objects	O
encountered	O
.	O
An	O
optimum	O
weighting	O
function	O
for	O
the	O
memory	O
registers	O
is	O
theoretically	O
derived	O
.	O
The	O
optimum	O
time-dependent	O
weighting	O
function	O
allows	O
sorting	B-KEY
and	O
clustering	B-KEY
of	O
the	O
randomly	O
distributed	O
objects	O
in	O
the	O
shortest	O
time	O
.	O
By	O
maximizing	O
the	O
average	O
speed	O
of	O
a	O
texture	O
feature	O
-LRB-	O
the	O
contrast	O
-RRB-	O
we	O
check	O
the	O
central	O
assumption	O
,	O
the	O
intermediate	O
steady-states	O
hypothesis	O
,	O
of	O
our	O
theoretical	O
result	O
.	O
It	O
is	O
proved	O
that	O
the	O
algorithm	B-KEY
optimization	I-KEY
based	O
on	O
maximum	O
speed	O
variation	O
of	O
the	O
contrast	O
feature	O
gives	O
relationships	O
similar	O
to	O
the	O
theoretically	O
derived	O
annealing	O
law	O

Novel	O
approach	O
to	O
super-resolution	B-KEY
pits	I-KEY
readout	I-KEY
We	O
proposed	O
a	O
novel	O
method	O
to	O
realize	O
the	O
readout	O
of	O
super-resolution	O
pits	O
by	O
using	O
a	O
super-resolution	B-KEY
reflective	I-KEY
film	I-KEY
to	O
replace	O
the	O
reflective	O
layer	O
of	O
the	O
conventional	O
ROM	O
.	O
At	O
the	O
same	O
time	O
,	O
by	O
using	O
Sb	O
as	O
the	O
super-resolution	O
reflective	O
layer	O
and	O
SiN	O
as	O
a	O
dielectric	O
layer	O
,	O
the	O
super-resolution	O
pits	O
with	O
diameters	O
of	O
380	B-KEY
nm	I-KEY
were	O
read	O
out	O
by	O
a	O
setup	O
whose	O
laser	O
wavelength	O
is	O
632.8	B-KEY
nm	I-KEY
and	O
numerical	B-KEY
aperture	I-KEY
is	O
0.40	O
.	O
In	O
addition	O
,	O
the	O
influence	O
of	O
the	O
Sb	B-KEY
thin	I-KEY
film	I-KEY
thickness	I-KEY
on	O
the	O
readout	B-KEY
signal	I-KEY
was	O
investigated	O
,	O
the	O
results	O
showed	O
that	O
the	O
optimum	O
Sb	B-KEY
thin	I-KEY
film	I-KEY
thickness	I-KEY
is	O
28	B-KEY
to	I-KEY
30	I-KEY
nm	I-KEY
,	O
and	O
the	O
maximum	B-KEY
CNR	I-KEY
is	O
38	O
to	O
40	O
dB	O

Integrate-and-fire	B-KEY
neurons	I-KEY
driven	O
by	O
correlated	B-KEY
stochastic	I-KEY
input	I-KEY
Neurons	O
are	O
sensitive	O
to	O
correlations	O
among	O
synaptic	O
inputs	O
.	O
However	O
,	O
analytical	O
models	O
that	O
explicitly	O
include	O
correlations	O
are	O
hard	O
to	O
solve	O
analytically	O
,	O
so	O
their	O
influence	O
on	O
a	O
neuron	O
's	O
response	O
has	O
been	O
difficult	O
to	O
ascertain	O
.	O
To	O
gain	O
some	O
intuition	O
on	O
this	O
problem	O
,	O
we	O
studied	O
the	O
firing	B-KEY
times	I-KEY
of	O
two	O
simple	O
integrate-and-fire	O
model	O
neurons	O
driven	O
by	O
a	O
correlated	B-KEY
binary	I-KEY
variable	I-KEY
that	O
represents	O
the	O
total	O
input	O
current	O
.	O
Analytic	O
expressions	O
were	O
obtained	O
for	O
the	O
average	O
firing	O
rate	O
and	O
coefficient	B-KEY
of	I-KEY
variation	I-KEY
-LRB-	O
a	O
measure	O
of	O
spike-train	B-KEY
variability	I-KEY
-RRB-	O
as	O
functions	O
of	O
the	O
mean	O
,	O
variance	O
,	O
and	O
correlation	O
time	O
of	O
the	O
stochastic	O
input	O
.	O
The	O
results	O
of	O
computer	B-KEY
simulations	I-KEY
were	O
in	O
excellent	O
agreement	O
with	O
these	O
expressions	O
.	O
In	O
these	O
models	O
,	O
an	O
increase	O
in	O
correlation	O
time	O
in	O
general	O
produces	O
an	O
increase	O
in	O
both	O
the	O
average	O
firing	O
rate	O
and	O
the	O
variability	O
of	O
the	O
output	B-KEY
spike	I-KEY
trains	I-KEY
.	O
However	O
,	O
the	O
magnitude	O
of	O
the	O
changes	O
depends	O
differentially	O
on	O
the	O
relative	O
values	O
of	O
the	O
input	O
mean	O
and	O
variance	O
:	O
the	O
increase	O
in	O
firing	O
rate	O
is	O
higher	O
when	O
the	O
variance	O
is	O
large	O
relative	O
to	O
the	O
mean	O
,	O
whereas	O
the	O
increase	O
in	O
variability	O
is	O
higher	O
when	O
the	O
variance	O
is	O
relatively	O
small	O
.	O
In	O
addition	O
,	O
the	O
firing	O
rate	O
always	O
tends	O
to	O
a	O
finite	O
limit	O
value	O
as	O
the	O
correlation	O
time	O
increases	O
toward	O
infinity	O
,	O
whereas	O
the	O
coefficient	B-KEY
of	I-KEY
variation	I-KEY
typically	O
diverges	O
.	O
These	O
results	O
suggest	O
that	O
temporal	B-KEY
correlations	I-KEY
may	O
play	O
a	O
major	O
role	O
in	O
determining	O
the	O
variability	O
as	O
well	O
as	O
the	O
intensity	O
of	O
neuronal	O
spike	O
trains	O

Lattice	B-KEY
Boltzmann	I-KEY
schemes	I-KEY
for	O
quantum	B-KEY
applications	I-KEY
We	O
review	O
the	O
basic	O
ideas	O
behind	O
the	O
quantum	O
lattice	O
Boltzmann	O
equation	O
-LRB-	O
LBE	O
-RRB-	O
,	O
and	O
present	O
a	O
few	O
thoughts	O
on	O
the	O
possible	O
use	O
of	O
such	O
an	O
equation	O
for	O
simulating	O
quantum	B-KEY
many-body	I-KEY
problems	I-KEY
on	O
both	O
-LRB-	O
parallel	O
-RRB-	O
electronic	O
and	O
quantum	B-KEY
computers	I-KEY

Applied	B-KEY
ethics	I-KEY
in	O
business	B-KEY
information	I-KEY
units	I-KEY
The	O
primary	O
thesis	O
of	O
this	O
paper	O
is	O
that	O
business	B-KEY
information	I-KEY
professionals	I-KEY
commonly	O
overlook	O
ethical	B-KEY
dilemmas	I-KEY
in	O
the	O
workplace	O
.	O
Although	O
the	O
thesis	O
remains	O
unproven	O
,	O
the	O
author	O
highlights	O
,	O
by	O
way	O
of	O
real	O
and	O
hypothetical	O
case	O
studies	O
,	O
a	O
number	O
of	O
situations	O
in	O
which	O
ethical	O
tensions	O
can	O
be	O
identified	O
,	O
and	O
suggests	O
that	O
information	O
professionals	O
need	O
to	O
be	O
more	O
aware	O
of	O
the	O
moral	B-KEY
context	I-KEY
of	O
their	O
actions	O
.	O
Resolving	O
ethical	B-KEY
dilemmas	I-KEY
should	O
be	O
one	O
of	O
the	O
aims	O
of	O
competent	O
information	O
professionals	O
and	O
their	O
managers	O
,	O
although	O
it	O
is	O
recognized	O
that	O
dilemmas	O
often	O
can	O
not	O
easily	O
be	O
resolved	O
.	O
A	O
background	O
to	O
the	O
main	O
theories	O
of	O
applied	B-KEY
ethics	I-KEY
forms	O
the	O
framework	O
for	O
later	O
discussion	O

MPEG-4	O
video	O
object-based	B-KEY
rate	I-KEY
allocation	I-KEY
with	O
variable	B-KEY
temporal	I-KEY
rates	I-KEY
In	O
object-based	O
coding	O
,	O
bit	B-KEY
allocation	I-KEY
is	O
performed	O
at	O
the	O
object	O
level	O
and	O
temporal	O
rates	O
of	O
different	O
objects	O
may	O
vary	O
.	O
The	O
proposed	O
algorithm	O
deals	O
with	O
these	O
two	O
issues	O
when	O
coding	O
multiple	B-KEY
video	I-KEY
objects	I-KEY
-LRB-	O
MVOs	O
-RRB-	O
.	O
The	O
proposed	O
algorithm	O
is	O
able	O
to	O
successfully	O
achieve	O
the	O
target	O
bit	O
rate	O
,	O
effectively	O
code	O
arbitrarily	O
shaped	O
MVOs	O
with	O
different	O
temporal	O
rates	O
,	O
and	O
maintain	O
a	O
stable	O
buffer	O
level	O

Speaker	B-KEY
adaptive	I-KEY
modeling	I-KEY
by	O
vocal	B-KEY
tract	I-KEY
normalization	I-KEY
This	O
paper	O
presents	O
methods	O
for	O
speaker	B-KEY
adaptive	I-KEY
modeling	I-KEY
using	O
vocal	B-KEY
tract	I-KEY
normalization	I-KEY
-LRB-	O
VTN	O
-RRB-	O
along	O
with	O
experimental	O
tests	O
on	O
three	O
databases	B-KEY
.	O
We	O
propose	O
a	O
new	O
training	B-KEY
method	I-KEY
for	O
VTN	O
:	O
By	O
using	O
single-density	B-KEY
acoustic	I-KEY
models	I-KEY
per	O
HMM	B-KEY
state	I-KEY
for	O
selecting	O
the	O
scale	O
factor	O
of	O
the	O
frequency	O
axis	O
,	O
we	O
avoid	O
the	O
problem	O
that	O
a	O
mixture-density	O
tends	O
to	O
learn	O
the	O
scale	O
factors	O
of	O
the	O
training	B-KEY
speakers	I-KEY
and	O
thus	O
can	O
not	O
be	O
used	O
for	O
selecting	O
the	O
scale	O
factor	O
.	O
We	O
show	O
that	O
using	O
single	B-KEY
Gaussian	I-KEY
densities	I-KEY
for	O
selecting	O
the	O
scale	O
factor	O
in	O
training	B-KEY
results	I-KEY
in	O
lower	O
error	O
rates	O
than	O
using	O
mixture	O
densities	O
.	O
For	O
the	O
recognition	O
phase	O
,	O
we	O
propose	O
an	O
improvement	O
of	O
the	O
well-known	O
two-pass	B-KEY
strategy	I-KEY
:	O
by	O
using	O
a	O
non-normalized	O
acoustic	O
model	O
for	O
the	O
first	O
recognition	O
pass	O
instead	O
of	O
a	O
normalized	O
model	O
,	O
lower	O
error	O
rates	O
are	O
obtained	O
.	O
In	O
recognition	O
tests	O
,	O
this	O
method	O
is	O
compared	O
with	O
a	O
fast	O
variant	O
of	O
VTN	O
.	O
The	O
two-pass	B-KEY
strategy	I-KEY
is	O
an	O
efficient	O
method	O
,	O
but	O
it	O
is	O
suboptimal	O
because	O
the	O
scale	O
factor	O
and	O
the	O
word	B-KEY
sequence	I-KEY
are	O
determined	O
sequentially	O
.	O
We	O
found	O
that	O
for	O
telephone	B-KEY
digit	I-KEY
string	I-KEY
recognition	I-KEY
this	O
suboptimality	O
reduces	O
the	O
VTN	O
gain	O
in	O
recognition	O
performance	O
by	O
30	O
%	O
relative	O
.	O
In	O
summary	O
,	O
on	O
the	O
German	B-KEY
spontaneous	I-KEY
speech	I-KEY
task	I-KEY
Verbmobil	O
,	O
the	O
WSJ	B-KEY
task	I-KEY
and	O
the	O
German	B-KEY
telephone	I-KEY
digit	I-KEY
string	I-KEY
corpus	I-KEY
SieTill	B-KEY
,	O
the	O
proposed	O
methods	O
for	O
VTN	O
reduce	O
the	O
error	O
rates	O
significantly	O

On	O
the	O
contractivity	B-KEY
of	O
implicit-explicit	B-KEY
linear	I-KEY
multistep	I-KEY
methods	I-KEY
This	O
paper	O
is	O
concerned	O
with	O
the	O
class	O
of	O
implicit-explicit	B-KEY
linear	I-KEY
multistep	I-KEY
methods	I-KEY
for	O
the	O
numerical	B-KEY
solution	I-KEY
of	O
initial	B-KEY
value	I-KEY
problems	I-KEY
for	O
ordinary	B-KEY
differential	I-KEY
equations	I-KEY
which	O
are	O
composed	O
of	O
stiff	O
and	O
nonstiff	O
parts	O
.	O
We	O
study	O
the	O
contractivity	B-KEY
of	O
such	O
methods	O
,	O
with	O
regard	O
to	O
linear	B-KEY
autonomous	I-KEY
systems	I-KEY
of	O
ordinary	B-KEY
differential	I-KEY
equations	I-KEY
and	O
a	O
-LRB-	O
scaled	O
-RRB-	O
Euclidean	B-KEY
norm	I-KEY
.	O
In	O
addition	O
,	O
we	O
derive	O
a	O
strong	O
stability	B-KEY
result	I-KEY
based	O
on	O
the	O
stability	O
regions	O
of	O
these	O
methods	O

Reply	O
to	O
``	O
Comment	O
on	O
:	O
Teleportation	B-KEY
of	O
an	O
unknown	B-KEY
state	I-KEY
by	O
W	O
state	O
''	O
-LSB-	O
Phys	O
.	O
Lett	O
.	O
A	O
300	O
-LRB-	O
2002	O
-RRB-	O
324	O
-RSB-	O
In	O
our	O
letter	O
-LRB-	O
see	O
ibid.	O
,	O
vol	O
.	O
296	O
,	O
p.	O
161	O
-LRB-	O
2002	O
-RRB-	O
-RRB-	O
,	O
the	O
main	O
question	O
we	O
consider	O
is	O
whether	O
a	O
general	O
three-particle	B-KEY
W	I-KEY
state	I-KEY
can	O
be	O
used	O
to	O
realize	O
the	O
teleportation	B-KEY
of	O
an	O
unknown	O
qubit	B-KEY
state	I-KEY
.	O
We	O
give	O
the	O
positive	O
answer	O
to	O
this	O
question	O
in	O
our	O
letter	O
,	O
and	O
show	O
that	O
W	O
state	O
can	O
be	O
used	O
to	O
realize	O
to	O
do	O
that	O
probabilistically	O
.	O
We	O
also	O
discuss	O
how	O
to	O
do	O
it	O
in	O
detail	O
in	O
our	O
letter	O
.	O
In	O
the	O
previous	O
comment	O
-LRB-	O
see	O
ibid.	O
,	O
vol	O
.	O
300	O
,	O
p.	O
324	O
-LRB-	O
2002	O
-RRB-	O
-RRB-	O
,	O
authors	O
check	O
carefully	O
the	O
mathematics	O
calculation	O
of	O
our	O
letter	O
,	O
find	O
and	O
point	O
out	O
a	O
simple	O
mathematics	O
error	O
about	O
normalization	B-KEY
coefficient	I-KEY
of	O
Eq	O
.	O
-LRB-	O
1	O
-RRB-	O
.	O
This	O
mathematics	O
error	O
induces	O
the	O
incorrect	O
probability	B-KEY
calculation	I-KEY
of	O
Eq	O
.	O
-LRB-	O
6	O
-RRB-	O
,	O
and	O
also	O
an	O
incorrect	O
claim	O
in	O
first	O
part	O
of	O
our	O
letter	O

VoIP	O
makeover	O
transforms	O
ugly	O
duckling	O
network	O
Surrey	B-KEY
County	I-KEY
Council	I-KEY
's	O
Swan	B-KEY
project	O
is	O
Europe	O
's	O
biggest	O
implementation	O
of	O
voice	B-KEY
over	I-KEY
IP	I-KEY
.	O
Six	O
Wans	B-KEY
and	O
countless	O
Lans	B-KEY
are	O
are	O
being	O
consolidated	O
into	O
a	O
single	O
network	O
covering	O
6,000	O
users	O
at	O
200	O
sites	O
.	O
The	O
contract	O
was	O
signed	O
in	O
October	O
2001	O
for	O
Pounds	O
13m	O
over	O
five	O
years	O
and	O
rollout	O
will	O
be	O
completed	O
in	O
May	O
2003	O

Five-axis	B-KEY
NC	I-KEY
milling	I-KEY
of	O
ruled	B-KEY
surfaces	I-KEY
:	O
optimal	B-KEY
geometry	I-KEY
of	O
a	O
conical	B-KEY
tool	O
The	O
side	B-KEY
milling	I-KEY
of	O
ruled	B-KEY
surfaces	I-KEY
using	O
a	O
conical	B-KEY
milling	O
cutter	O
was	O
studied	O
.	O
This	O
is	O
a	O
field	O
that	O
has	O
largely	O
been	O
ignored	O
by	O
research	O
scientists	O
,	O
but	O
it	O
is	O
much	O
used	O
in	O
industry	O
,	O
especially	O
to	O
machine	O
turbine	O
blades	O
.	O
We	O
first	O
suggest	O
an	O
improved	O
positioning	B-KEY
with	O
respect	O
to	O
the	O
directrices	O
of	O
the	O
ruled	B-KEY
surface	I-KEY
.	O
As	O
compared	O
with	O
the	O
methods	O
already	O
developed	O
for	O
the	O
cylindrical	O
cutter	O
,	O
this	O
positioning	B-KEY
enables	O
the	O
error	O
between	O
the	O
cutter	O
and	O
the	O
work-piece	O
to	O
be	O
reduced	O
.	O
An	O
algorithm	O
is	O
then	O
introduced	O
to	O
calculate	O
error	O
so	O
one	O
can	O
determine	O
the	O
cutter	B-KEY
dimensions	I-KEY
-LRB-	O
cone	O
radius	O
and	O
angle	O
-RRB-	O
in	O
order	O
to	O
respect	O
the	O
tolerance	B-KEY
interval	I-KEY
imposed	O
by	O
the	O
design	O
office	O
.	O
This	O
study	O
provides	O
an	O
opportunity	O
to	O
determine	O
cutters	O
with	O
greater	O
dimensions	O
,	O
thus	O
alleviating	O
bending	O
problems	O
during	O
milling	O

Novel	O
line	O
conditioner	O
with	O
voltage	O
up/down	O
capability	O
In	O
this	O
paper	O
,	O
a	O
novel	O
pulsewidth-modulated	B-KEY
line	I-KEY
conditioner	I-KEY
with	O
fast	O
output	B-KEY
voltage	I-KEY
control	I-KEY
is	O
proposed	O
.	O
The	O
line	O
conditioner	O
is	O
made	O
up	O
of	O
an	O
AC	B-KEY
chopper	I-KEY
with	O
reversible	B-KEY
voltage	I-KEY
control	I-KEY
and	O
a	O
transformer	O
for	O
series	O
voltage	O
compensation	O
.	O
In	O
the	O
AC	B-KEY
chopper	I-KEY
,	O
a	O
proper	O
switching	B-KEY
operation	I-KEY
is	O
achieved	O
without	O
the	O
commutation	B-KEY
problem	O
.	O
To	O
absorb	O
energy	O
stored	O
in	O
line	B-KEY
stray	I-KEY
inductance	I-KEY
,	O
a	O
regenerative	B-KEY
DC	I-KEY
snubber	I-KEY
can	O
be	O
utilized	O
which	O
has	O
only	O
one	O
capacitor	O
without	O
discharging	O
resistors	O
or	O
complicated	O
regenerative	O
circuit	O
for	O
snubber	O
energy	O
.	O
Therefore	O
,	O
the	O
proposed	O
AC	B-KEY
chopper	I-KEY
gives	O
high	O
efficiency	O
and	O
reliability	O
.	O
The	O
output	O
voltage	O
of	O
the	O
line	O
conditioner	O
is	O
controlled	O
using	O
a	O
fast	O
sensing	O
technique	O
of	O
the	O
output	O
voltage	O
.	O
It	O
is	O
also	O
shown	O
via	O
some	O
experimental	O
results	O
that	O
the	O
presented	O
line	O
conditioner	O
gives	O
good	O
dynamic	O
and	O
steady-state	B-KEY
performance	I-KEY
for	O
high	O
quality	O
of	O
the	O
output	O
voltage	O

Impossible	O
choice	O
-LSB-	O
web	B-KEY
hosting	I-KEY
service	I-KEY
provider	I-KEY
-RSB-	O
Selecting	B-KEY
a	O
telecoms	O
and	O
web	B-KEY
hosting	I-KEY
service	I-KEY
provider	I-KEY
has	O
become	O
a	O
high-stakes	O
game	O
of	O
chance	O

A	O
knowledge	B-KEY
intensive	I-KEY
multi-agent	I-KEY
framework	I-KEY
for	O
cooperative/collaborative	O
design	O
modeling	O
and	O
decision	O
support	O
of	O
assemblies	O
Multi-agent	O
modeling	O
has	O
emerged	O
as	O
a	O
promising	O
discipline	O
for	O
dealing	O
with	O
the	O
decision	O
making	O
process	O
in	O
distributed	B-KEY
information	I-KEY
system	I-KEY
applications	I-KEY
.	O
One	O
of	O
such	O
applications	O
is	O
the	O
modeling	O
of	O
distributed	B-KEY
design	I-KEY
or	O
manufacturing	O
processes	O
which	O
can	O
link	O
up	O
various	O
designs	O
or	O
manufacturing	O
processes	O
to	O
form	O
a	O
virtual	B-KEY
consortium	I-KEY
on	O
a	O
global	O
basis	O
.	O
This	O
paper	O
proposes	O
a	O
novel	O
knowledge	O
intensive	O
multi-agent	O
cooperative/collaborative	O
framework	O
for	O
concurrent	B-KEY
intelligent	I-KEY
design	I-KEY
and	O
assembly	B-KEY
planning	I-KEY
,	O
which	O
integrates	O
product	B-KEY
design	I-KEY
,	O
design	B-KEY
for	I-KEY
assembly	I-KEY
,	O
assembly	B-KEY
planning	I-KEY
,	O
assembly	O
system	O
design	O
,	O
and	O
assembly	B-KEY
simulation	I-KEY
subjected	O
to	O
econo-technical	O
evaluations	O
.	O
An	O
AI	B-KEY
protocol	I-KEY
based	O
method	O
is	O
proposed	O
to	O
facilitate	O
the	O
integration	O
of	O
intelligent	O
agents	O
for	O
assembly	O
design	O
,	O
planning	O
,	O
evaluation	O
and	O
simulation	O
processes	O
.	O
A	O
unified	O
class	O
of	O
knowledge	B-KEY
intensive	I-KEY
Petri	I-KEY
nets	I-KEY
is	O
defined	O
using	O
the	O
OO	O
knowledge-based	O
Petri	O
net	O
approach	O
and	O
used	O
as	O
an	O
AI	B-KEY
protocol	I-KEY
for	O
handling	O
both	O
the	O
integration	O
and	O
the	O
negotiation	O
problems	O
among	O
multi-agents	O
.	O
The	O
detailed	O
cooperative/collaborative	O
mechanism	O
and	O
algorithms	O
are	O
given	O
based	O
on	O
the	O
knowledge	B-KEY
object	I-KEY
cooperation	I-KEY
formalisms	O
.	O
As	O
such	O
,	O
the	O
assembly-oriented	O
design	O
system	O
can	O
easily	O
be	O
implemented	O
under	O
the	O
multi-agent-based	O
knowledge-intensive	O
Petri	O
net	O
framework	O
with	O
concurrent	O
integration	O
of	O
multiple	O
cooperative	O
knowledge	O
sources	O
and	O
software	O
.	O
Thus	O
,	O
product	B-KEY
design	I-KEY
and	O
assembly	B-KEY
planning	I-KEY
can	O
be	O
carried	O
out	O
simultaneously	O
and	O
intelligently	O
in	O
an	O
entirely	O
computer-aided	O
concurrent	O
design	O
and	O
assembly	B-KEY
planning	I-KEY
system	O

Games	O
machines	O
play	O
Individual	B-KEY
rationality	I-KEY
,	O
or	O
doing	O
what	O
is	O
best	O
for	O
oneself	O
,	O
is	O
a	O
standard	O
model	O
used	O
to	O
explain	O
and	O
predict	O
human	B-KEY
behavior	I-KEY
,	O
and	O
von	O
Neumann-Morgenstern	O
game	B-KEY
theory	I-KEY
is	O
the	O
classical	O
mathematical	O
formalization	O
of	O
this	O
theory	O
in	O
multiple-agent	B-KEY
settings	O
.	O
Individual	B-KEY
rationality	I-KEY
,	O
however	O
,	O
is	O
an	O
inadequate	O
model	O
for	O
the	O
synthesis	O
of	O
artificial	B-KEY
social	I-KEY
systems	I-KEY
where	O
cooperation	B-KEY
is	O
essential	O
,	O
since	O
it	O
does	O
not	O
permit	O
the	O
accommodation	O
of	O
group	O
interests	O
other	O
than	O
as	O
aggregations	O
of	O
individual	O
interests	O
.	O
Satisficing	O
game	B-KEY
theory	I-KEY
is	O
based	O
upon	O
a	O
well-defined	O
notion	O
of	O
being	O
good	O
enough	O
,	O
and	O
does	O
accommodate	O
group	O
as	O
well	O
as	O
individual	O
interests	O
through	O
the	O
use	O
of	O
conditional	B-KEY
preference	I-KEY
relationships	I-KEY
,	O
whereby	O
a	O
decision	O
maker	O
is	O
able	O
to	O
adjust	O
its	O
preferences	O
as	O
a	O
function	O
of	O
the	O
preferences	O
,	O
and	O
not	O
just	O
the	O
options	O
,	O
of	O
others	O
.	O
This	O
new	O
theory	O
is	O
offered	O
as	O
an	O
alternative	O
paradigm	O
to	O
construct	O
artificial	B-KEY
societies	I-KEY
that	O
are	O
capable	O
of	O
complex	O
behavior	O
that	O
goes	O
beyond	O
exclusive	O
self	B-KEY
interest	I-KEY

Modeling	O
of	O
torsional	B-KEY
vibration	I-KEY
induced	O
by	O
extension-twisting	O
coupling	O
of	O
anisotropic	O
composite	O
laminates	O
with	O
piezoelectric	O
actuators	O
In	O
this	O
paper	O
we	O
present	O
a	O
dynamic	B-KEY
analytical	I-KEY
model	I-KEY
for	O
the	O
torsional	B-KEY
vibration	I-KEY
of	O
an	O
anisotropic	B-KEY
piezoelectric	I-KEY
laminate	I-KEY
induced	O
by	O
the	O
extension-twisting	O
coupling	O
effect	O
.	O
In	O
the	O
present	O
approach	O
,	O
we	O
use	O
the	O
Hamilton	B-KEY
principle	I-KEY
and	O
a	O
reduced	B-KEY
bending	I-KEY
stiffness	I-KEY
method	O
for	O
the	O
derivation	O
of	O
equations	B-KEY
of	I-KEY
motion	I-KEY
.	O
As	O
a	O
result	O
,	O
the	O
in-plane	B-KEY
displacements	I-KEY
are	O
not	O
involved	O
and	O
the	O
out-of-plane	B-KEY
displacement	I-KEY
of	O
the	O
laminate	O
is	O
the	O
only	O
quantity	O
to	O
be	O
calculated	O
.	O
Therefore	O
,	O
the	O
proposed	O
method	O
turns	O
the	O
twisting	B-KEY
of	O
a	O
laminate	O
with	O
structural	B-KEY
coupling	I-KEY
into	O
a	O
simplified	O
problem	O
without	O
losing	O
its	O
features	O
.	O
We	O
give	O
analytical	O
solutions	O
of	O
the	O
present	O
model	O
with	O
harmonic	B-KEY
excitation	I-KEY
.	O
A	O
parametric	B-KEY
study	I-KEY
is	O
performed	O
to	O
demonstrate	O
the	O
present	O
approach	O

Influence	O
of	O
advertising	O
expenses	O
on	O
the	O
characteristics	O
of	O
functioning	O
of	O
an	O
insurance	O
company	O
The	O
basic	O
characteristics	O
of	O
the	O
functioning	O
of	O
an	O
insurance	O
company	O
,	O
including	O
the	O
average	B-KEY
capital	I-KEY
,	O
ruin	O
and	O
survival	B-KEY
probabilities	I-KEY
,	O
and	O
the	O
conditional	B-KEY
time	I-KEY
before	O
ruin	O
,	O
are	O
examined	O
with	O
allowance	O
for	O
advertising	O
expenses	O

What	O
do	O
you	O
say	O
?	O
Open	O
letters	O
to	O
women	B-KEY
considering	O
a	O
computer	B-KEY
science	I-KEY
major	I-KEY
In	O
the	O
last	O
decade	O
we	O
have	O
both	O
monitored	O
with	O
great	O
interest	O
the	O
ratio	O
of	O
female	B-KEY
to	O
male	B-KEY
computer	B-KEY
science	I-KEY
majors	I-KEY
at	O
our	O
respective	O
institutions	O
.	O
With	O
each	O
entering	O
class	O
,	O
we	O
think	O
:	O
``	O
Surely	O
,	O
now	O
is	O
the	O
time	O
when	O
the	O
numbers	O
will	O
become	O
more	O
balanced	O
.	O
''	O
Logic	O
tells	O
us	O
that	O
this	O
must	O
eventually	O
happen	O
,	O
because	O
the	O
opportunities	O
in	O
computing	O
are	O
simply	O
too	O
attractive	O
for	O
an	O
entire	O
segment	O
of	O
our	O
population	O
to	O
routinely	O
pass	O
up	O
.	O
But	O
each	O
year	O
we	O
are	O
again	O
disappointed	O
in	O
the	O
number	O
of	O
women	B-KEY
students	O
,	O
as	O
they	O
continue	O
to	O
be	O
woefully	O
under-represented	O
among	O
computer	B-KEY
science	I-KEY
majors	I-KEY
.	O
So	O
,	O
what	O
do	O
you	O
say	O
to	O
a	O
young	O
woman	O
who	O
is	O
considering	O
a	O
college	O
choice	O
and	O
a	O
choice	O
of	O
major	O
in	O
order	O
to	O
make	O
computer	O
science	O
a	O
more	O
attractive	O
option	O
?	O
We	O
have	O
organized	O
some	O
thoughts	O
on	O
that	O
subject	O
into	O
open	O
letters	O

Reproducibility	O
of	O
mammary	O
gland	O
structure	O
during	O
repeat	B-KEY
setups	I-KEY
in	O
a	O
supine	B-KEY
position	I-KEY
Purpose	O
:	O
In	O
breast	B-KEY
conserving	I-KEY
therapy	I-KEY
,	O
complete	O
excision	O
of	O
the	O
tumor	O
with	O
an	O
acceptable	O
cosmetic	O
outcome	O
depends	O
on	O
accurate	O
localization	O
in	O
terms	O
of	O
both	O
the	O
position	O
of	O
the	O
lesion	O
and	O
its	O
extent	O
.	O
We	O
hypothesize	O
that	O
preoperative	O
contrast-enhanced	O
magnetic	O
resonance	O
-LRB-	O
MR	O
-RRB-	O
imaging	O
of	O
the	O
patient	O
in	O
a	O
supine	B-KEY
position	I-KEY
may	O
be	O
used	O
for	O
accurate	B-KEY
tumor	I-KEY
localization	I-KEY
and	O
marking	O
of	O
its	O
extent	O
immediately	O
prior	O
to	O
surgery	O
.	O
Our	O
aims	O
in	O
this	O
study	O
are	O
to	O
assess	O
the	O
reproducibility	O
of	O
mammary	O
gland	O
structure	O
during	O
repeat	B-KEY
setups	I-KEY
in	O
a	O
supine	B-KEY
position	I-KEY
,	O
to	O
evaluate	O
the	O
effect	O
of	O
a	O
breast	B-KEY
immobilization	I-KEY
device	I-KEY
,	O
and	O
to	O
derive	O
reproducibility	B-KEY
margins	I-KEY
that	O
take	O
internal	B-KEY
tissue	I-KEY
shifts	I-KEY
into	O
account	O
occurring	O
between	O
repeat	B-KEY
setups	I-KEY
.	O
Materials	O
Methods	O
:	O
The	O
reproducibility	O
of	O
mammary	O
gland	O
structure	O
during	O
repeat	B-KEY
setups	I-KEY
in	O
a	O
supine	B-KEY
position	I-KEY
is	O
estimated	O
by	O
quantification	O
of	O
tissue	O
shifts	O
in	O
the	O
breasts	O
of	O
healthy	O
volunteers	O
between	O
repeat	O
MR	O
setups	O
.	O
For	O
each	O
volunteer	O
fiducials	O
are	O
identified	O
and	O
registered	O
with	O
their	O
counter	O
locations	O
in	O
corresponding	O
MR	O
volumes	O
.	O
The	O
difference	O
in	O
position	O
denotes	O
the	O
shift	O
of	O
breast	O
tissue	O
.	O
The	O
dependence	O
on	O
breast	O
volume	O
and	O
the	O
part	O
of	O
the	O
breast	O
,	O
as	O
well	O
as	O
the	O
effect	O
of	O
a	O
breast	O
immobilization	O
cast	O
are	O
studied	O
.	O
Results	O
:	O
The	O
tissue	O
shifts	O
are	O
small	O
with	O
a	O
mean	O
standard	O
deviation	O
on	O
the	O
order	O
of	O
1.5	O
mm	O
,	O
being	O
slightly	O
larger	O
in	O
large	O
breasts	O
-LRB-	O
V	O
>	O
1000	O
cm/sup	O
3	O
/	O
-RRB-	O
,	O
and	O
in	O
the	O
posterior	O
part	O
-LRB-	O
toward	O
the	O
pectoral	O
muscle	O
-RRB-	O
of	O
both	O
small	O
and	O
large	O
breasts	O
.	O
The	O
application	O
of	O
a	O
breast	O
immobilization	O
cast	O
reduces	O
the	O
tissue	O
shifts	O
in	O
large	O
breasts	O
.	O
A	O
reproducibility	B-KEY
margin	I-KEY
on	O
the	O
order	O
of	O
5	O
mm	O
will	O
take	O
the	O
internal	B-KEY
tissue	I-KEY
shifts	I-KEY
into	O
account	O
that	O
occur	O
between	O
repeat	B-KEY
setups	I-KEY
.	O
Conclusion	O
:	O
The	O
results	O
demonstrate	O
a	O
high	O
reproducibility	O
of	O
mammary	O
gland	O
structure	O
during	O
repeat	B-KEY
setups	I-KEY
in	O
a	O
supine	B-KEY
position	I-KEY

Reachability	O
in	O
contextual	O
nets	O
Contextual	O
nets	O
,	O
or	O
Petri	B-KEY
nets	I-KEY
with	O
read	O
arcs	O
,	O
are	O
models	O
of	O
concurrent	B-KEY
systems	I-KEY
with	O
context	B-KEY
dependent	I-KEY
actions	I-KEY
.	O
The	O
problem	O
of	O
reachability	O
in	O
such	O
nets	O
consists	O
in	O
finding	O
a	O
sequence	O
of	O
transitions	O
that	O
leads	O
from	O
the	O
initial	O
marking	O
of	O
a	O
given	O
contextual	O
net	O
to	O
a	O
given	O
goal	B-KEY
marking	I-KEY
.	O
The	O
solution	O
to	O
this	O
problem	O
that	O
is	O
presented	O
in	O
this	O
paper	O
consists	O
in	O
constructing	O
a	O
finite	O
complete	O
prefix	O
of	O
the	O
unfolding	O
of	O
the	O
given	O
contextual	O
net	O
,	O
that	O
is	O
a	O
finite	B-KEY
prefix	I-KEY
in	O
which	O
all	O
the	O
markings	O
that	O
are	O
reachable	O
from	O
the	O
initial	O
marking	O
are	O
present	O
,	O
and	O
in	O
searching	O
in	O
each	O
branch	O
of	O
this	O
prefix	O
for	O
the	O
goal	B-KEY
marking	I-KEY
by	O
solving	O
an	O
appropriate	O
linear	B-KEY
programming	I-KEY
problem	O

On	O
average	B-KEY
depth	I-KEY
of	O
decision	B-KEY
trees	I-KEY
implementing	O
Boolean	B-KEY
functions	I-KEY
The	O
article	O
considers	O
the	O
representation	O
of	O
Boolean	B-KEY
functions	I-KEY
in	O
the	O
form	O
of	O
decision	B-KEY
trees	I-KEY
.	O
It	O
presents	O
the	O
bounds	O
on	O
average	B-KEY
time	I-KEY
complexity	I-KEY
of	O
decision	B-KEY
trees	I-KEY
for	O
all	O
classes	O
of	O
Boolean	B-KEY
functions	I-KEY
that	O
are	O
closed	O
over	O
substitution	O
,	O
and	O
the	O
insertion	O
and	O
deletion	O
of	O
unessential	O
variables	O
.	O
The	O
obtained	O
results	O
are	O
compared	O
with	O
the	O
results	O
developed	O
by	O
M.Ju	O
.	O
Moshkov	O
-LRB-	O
1995	O
-RRB-	O
that	O
describe	O
the	O
worst	B-KEY
case	I-KEY
time	I-KEY
complexity	I-KEY
of	O
decision	B-KEY
trees	I-KEY

Simulation	O
study	O
of	O
the	O
cardiovascular	B-KEY
functional	I-KEY
status	I-KEY
in	O
hypertensive	B-KEY
situation	I-KEY
An	O
extended	B-KEY
cardiovascular	I-KEY
model	I-KEY
was	O
established	O
based	O
on	O
our	O
previous	O
work	O
to	O
study	O
the	O
consequences	O
of	O
physiological	O
or	O
pathological	B-KEY
changes	I-KEY
to	O
the	O
homeostatic	B-KEY
functions	I-KEY
of	O
the	O
cardiovascular	O
system	O
.	O
To	O
study	O
hemodynamic	B-KEY
changes	O
in	O
hypertensive	B-KEY
situations	I-KEY
,	O
the	O
impacts	O
of	O
cardiovascular	B-KEY
parameter	I-KEY
variations	I-KEY
-LRB-	O
peripheral	B-KEY
vascular	I-KEY
resistance	I-KEY
,	O
arterial	B-KEY
vessel	I-KEY
wall	I-KEY
stiffness	I-KEY
and	O
baroreflex	B-KEY
gain	I-KEY
-RRB-	O
upon	O
hemodynamics	B-KEY
and	O
the	O
short-term	B-KEY
regulation	I-KEY
of	O
the	O
cardiovascular	O
system	O
were	O
investigated	O
.	O
For	O
the	O
purpose	O
of	O
analyzing	O
baroregulation	O
function	O
,	O
the	O
short-term	B-KEY
regulation	I-KEY
of	O
arterial	B-KEY
pressure	I-KEY
in	O
response	O
to	O
moderate	B-KEY
dynamic	I-KEY
exercise	I-KEY
for	O
normotensive	O
and	O
hypertensive	O
cases	O
was	O
studied	O
through	O
computer	B-KEY
simulation	I-KEY
and	O
clinical	B-KEY
experiments	I-KEY
.	O
The	O
simulation	O
results	O
agree	O
well	O
with	O
clinical	O
data	O
.	O
The	O
results	O
of	O
this	O
work	O
suggest	O
that	O
the	O
model	O
presented	O
in	O
this	O
paper	O
provides	O
a	O
useful	O
tool	O
to	O
investigate	O
the	O
functional	O
status	O
of	O
the	O
cardiovascular	O
system	O
in	O
normal	O
or	O
pathological	O
conditions	O

Sliding	B-KEY
mode	I-KEY
control	I-KEY
of	O
chaos	B-KEY
in	O
the	O
cubic	O
Chua	O
's	O
circuit	O
system	O
In	O
this	O
paper	O
,	O
a	O
sliding	B-KEY
mode	I-KEY
controller	I-KEY
is	O
applied	O
to	O
control	O
the	O
cubic	O
Chua	O
's	O
circuit	O
system	O
.	O
The	O
sliding	B-KEY
surface	I-KEY
of	O
this	O
paper	O
used	O
is	O
one	O
dimension	O
higher	O
than	O
the	O
traditional	O
surface	O
and	O
guarantees	O
its	O
passage	O
through	O
the	O
initial	O
states	O
of	O
the	O
controlled	O
system	O
.	O
Therefore	O
,	O
using	O
the	O
characteristic	O
of	O
this	O
sliding	O
mode	O
we	O
aim	O
to	O
design	O
a	O
controller	O
that	O
can	O
meet	O
the	O
desired	O
specification	O
and	O
use	O
less	O
control	O
energy	O
by	O
comparing	O
with	O
the	O
result	O
in	O
the	O
current	O
existing	O
literature	O
.	O
The	O
results	O
show	O
that	O
the	O
proposed	O
controller	O
can	O
steer	O
Chua	O
's	O
circuit	O
system	O
to	O
the	O
desired	O
state	O
without	O
the	O
chattering	B-KEY
phenomenon	O
and	O
abrupt	O
state	B-KEY
change	I-KEY

Gearing	O
up	O
for	O
CLS	O
bank	O
Continuous-Linked	B-KEY
Settlement	I-KEY
,	O
a	O
dream	O
of	O
the	O
foreign-exchange	B-KEY
community	O
for	O
years	O
,	O
may	O
finally	O
become	O
a	O
reality	O
by	O
the	O
end	O
of	O
2002	O

IT	O
:	O
Utilities	B-KEY
A	O
look	O
at	O
five	O
utilities	B-KEY
to	O
make	O
your	O
PCs	B-KEY
more	O
,	O
efficient	O
,	O
effective	O
,	O
and	O
efficacious	O

A	O
static	B-KEY
semantics	I-KEY
for	O
Haskell	O
This	O
paper	O
gives	O
a	O
static	B-KEY
semantics	I-KEY
for	O
Haskell	B-KEY
98	I-KEY
,	O
a	O
non-strict	O
purely	O
functional	O
programming	O
language	O
.	O
The	O
semantics	O
formally	O
specifies	O
nearly	O
all	O
the	O
details	O
of	O
the	O
Haskell	B-KEY
98	I-KEY
type	B-KEY
system	I-KEY
,	O
including	O
the	O
resolution	O
of	O
overloading	B-KEY
,	O
kind	B-KEY
inference	I-KEY
-LRB-	O
including	O
defaulting	O
-RRB-	O
and	O
polymorphic	B-KEY
recursion	I-KEY
,	O
the	O
only	O
major	O
omission	O
being	O
a	O
proper	O
treatment	O
of	O
ambiguous	O
overloading	B-KEY
and	O
its	O
resolution	O
.	O
Overloading	B-KEY
is	O
translated	O
into	O
explicit	B-KEY
dictionary	I-KEY
passing	I-KEY
,	O
as	O
in	O
all	O
current	O
implementations	O
of	O
Haskell	O
.	O
The	O
target	O
language	O
of	O
this	O
translation	O
is	O
a	O
variant	O
of	O
the	O
Girard-Reynolds	O
polymorphic	B-KEY
lambda	I-KEY
calculus	I-KEY
featuring	O
higher	B-KEY
order	I-KEY
polymorphism	I-KEY
.	O
and	O
explicit	B-KEY
type	I-KEY
abstraction	I-KEY
and	O
application	O
in	O
the	O
term	B-KEY
language	I-KEY
.	O
Translated	O
programs	O
can	O
thus	O
still	O
be	O
type	B-KEY
checked	I-KEY
,	O
although	O
the	O
implicit	O
version	O
of	O
this	O
system	O
is	O
impredicative	O
.	O
A	O
surprising	O
result	O
of	O
this	O
formalization	O
effort	O
is	O
that	O
the	O
monomorphism	B-KEY
restriction	I-KEY
,	O
when	O
rendered	O
in	O
a	O
system	O
of	O
inference	B-KEY
rules	I-KEY
,	O
compromises	O
the	O
principal	O
type	O
property	O

ECG-gated	O
/	O
sup	O
18/F-FDG	O
positron	O
emission	O
tomography	O
.	O
Single	O
test	O
evaluation	O
of	O
segmental	O
metabolism	O
,	O
function	B-KEY
and	O
contractile	O
reserve	O
in	O
patients	B-KEY
with	O
coronary	B-KEY
artery	I-KEY
disease	I-KEY
and	O
regional	B-KEY
dysfunction	I-KEY
/	O
sup	O
18/F-fluorodeoxyglucose	O
-LRB-	O
/	O
sup	O
18/F-FDG	O
-RRB-	O
-	O
positron	O
emission	O
tomography	O
-LRB-	O
PET	O
-RRB-	O
provides	O
information	O
about	O
myocardial	B-KEY
glucose	I-KEY
metabolism	I-KEY
to	O
diagnose	O
myocardial	B-KEY
viability	I-KEY
.	O
Additional	O
information	O
about	O
the	O
functional	B-KEY
status	O
is	O
necessary	O
.	O
Comparison	O
of	O
tomographic	O
metabolic	O
PET	O
with	O
data	O
from	O
other	O
imaging	O
techniques	O
is	O
always	O
hampered	O
by	O
some	O
transfer	B-KEY
uncertainty	I-KEY
and	O
scatter	O
.	O
We	O
wanted	O
to	O
evaluate	O
a	O
new	O
Fourier-based	B-KEY
ECG-gated	I-KEY
PET	I-KEY
technique	I-KEY
using	O
a	O
high	B-KEY
resolution	I-KEY
scanner	I-KEY
providing	O
both	O
metabolic	O
and	O
functional	B-KEY
data	O
with	O
respect	O
to	O
feasibility	O
in	O
patients	B-KEY
with	O
diseased	B-KEY
left	I-KEY
ventricles	I-KEY
.	O
Forty-five	O
patients	B-KEY
with	O
coronary	B-KEY
artery	I-KEY
disease	I-KEY
and	O
at	O
least	O
one	O
left	B-KEY
ventricular	I-KEY
segment	I-KEY
with	O
severe	B-KEY
hypokinesis	I-KEY
or	O
akinesis	B-KEY
at	O
biplane	B-KEY
cineventriculography	I-KEY
were	O
included	O
.	O
A	O
new	O
Fourier-based	O
ECG-gated	O
metabolic	O
/	O
sup	O
18/F-FDG-PET	O
was	O
performed	O
in	O
these	O
patients	B-KEY
.	O
Function	B-KEY
at	O
rest	O
and	O
/	O
sup	O
18/F-FDG	O
uptake	O
were	O
examined	O
in	O
the	O
PET	O
study	O
using	O
a	O
36-segment	O
model	O
.	O
Segmental	O
comparison	O
with	O
ventriculography	B-KEY
revealed	O
a	O
high	O
reliability	O
in	O
identifying	O
dysfunctional	B-KEY
segments	I-KEY
-LRB-	O
>	O
96	O
%	O
-RRB-	O
.	O
/	O
sup	O
18/F-FDG	O
uptake	O
of	O
normokinetic/hypokinetic/akinetic	O
segments	O
was	O
75.4	O
+	O
or-7	O
.5	O
,	O
65.3	O
+	O
or-10	O
.5	O
,	O
and	O
35.9	O
+	O
or-15	O
.2	O
%	O
-LRB-	O
p	O
<	O
0.001	O
-RRB-	O
.	O
In	O
segments	O
>	O
or	O
=	O
70	O
%	O
/	O
sup	O
18/F-FDG	O
uptake	O
no	O
akinesia	O
was	O
observed	O
.	O
No	O
residual	B-KEY
function	I-KEY
was	O
found	O
below	O
40	O
%	O
/	O
sup	O
18/F-FDG	O
uptake	O
.	O
An	O
additional	O
dobutamine	B-KEY
test	I-KEY
was	O
performed	O
and	O
revealed	O
inotropic	B-KEY
reserve	I-KEY
-LRB-	O
viability	O
-RRB-	O
in	O
42	O
akinetic	B-KEY
segments	I-KEY
and	O
45	O
hypokinetic	B-KEY
segments	I-KEY
.	O
ECG-gated	O
metabolic	O
PET	O
with	O
pixel-based	B-KEY
Fourier	I-KEY
smoothing	I-KEY
provides	O
reliable	O
data	O
on	O
regional	B-KEY
function	I-KEY
.	O
Assessment	O
of	O
metabolism	O
and	O
function	B-KEY
makes	O
complete	O
judgement	O
of	O
segmental	B-KEY
status	I-KEY
feasible	O
within	O
a	O
single	O
study	O
without	O
any	O
transfer	B-KEY
artefacts	I-KEY
or	O
test-to-test	O
variability	O
.	O
The	O
results	O
indicate	O
the	O
presence	O
of	O
considerable	O
amounts	O
of	O
viable	B-KEY
myocardium	I-KEY
in	O
regions	O
with	O
an	O
uptake	O
of	O
40-50	O
%	O
/	O
sup	O
18/F-FDG	O

An	O
interlingua-based	O
Chinese-English	O
MT	O
system	O
Chinese-English	O
machine	O
translation	O
is	O
a	O
significant	O
and	O
challenging	O
problem	O
in	O
information	B-KEY
processing	I-KEY
.	O
The	O
paper	O
presents	O
an	O
interlingua-based	O
Chinese-English	O
natural	B-KEY
language	I-KEY
translation	I-KEY
system	I-KEY
-LRB-	O
ICENT	O
-RRB-	O
.	O
It	O
introduces	O
the	O
realization	O
mechanism	O
of	O
Chinese	O
language	O
analysis	O
,	O
which	O
contains	O
syntactic	B-KEY
parsing	I-KEY
and	O
semantic	B-KEY
analyzing	I-KEY
and	O
gives	O
the	O
design	O
of	O
interlingua	O
in	O
details	O
.	O
Experimental	O
results	O
and	O
system	O
evaluation	O
are	O
given	O
.	O
The	O
result	O
is	O
satisfying	O

A	O
new	O
identification	B-KEY
approach	I-KEY
for	O
FIR	B-KEY
models	I-KEY
The	O
identification	O
of	O
stochastic	B-KEY
discrete	I-KEY
systems	I-KEY
disturbed	O
with	O
noise	O
is	O
discussed	O
in	O
this	O
brief	O
.	O
The	O
concept	O
of	O
general	O
prediction	O
error	O
-LRB-	O
GPE	O
-RRB-	O
criterion	O
is	O
introduced	O
for	O
the	O
time-domain	B-KEY
estimate	I-KEY
with	O
optimal	B-KEY
frequency	I-KEY
estimation	I-KEY
-LRB-	O
OFE	O
-RRB-	O
introduced	O
for	O
the	O
frequency-domain	B-KEY
estimate	I-KEY
.	O
The	O
two	O
estimation	O
methods	O
are	O
combined	O
to	O
form	O
a	O
new	O
identification	O
algorithm	O
,	O
which	O
is	O
called	O
the	O
empirical	O
frequency-domain	O
optimal	O
parameter	O
-LRB-	O
EFOP	O
-RRB-	O
estimate	O
,	O
for	O
the	O
finite	O
impulse	O
response	O
-LRB-	O
FIR	O
-RRB-	O
model	O
interfered	O
by	O
noise	O
.	O
The	O
algorithm	O
theoretically	O
provides	O
the	O
global	O
optimum	O
of	O
the	O
model	O
frequency-domain	B-KEY
estimate	I-KEY
.	O
Some	O
simulation	O
examples	O
are	O
given	O
to	O
illustrate	O
the	O
new	O
identification	O
method	O

Yet	O
some	O
more	O
complexity	B-KEY
results	I-KEY
for	O
default	B-KEY
logic	I-KEY
We	O
identify	O
several	O
new	O
tractable	B-KEY
subsets	I-KEY
and	O
several	O
new	O
intractable	O
simple	O
cases	O
for	O
reasoning	B-KEY
in	O
the	O
propositional	O
version	O
of	O
Reiter	O
's	O
default	B-KEY
logic	I-KEY
.	O
The	O
majority	O
of	O
our	O
findings	O
are	O
related	O
to	O
brave	O
reasoning	B-KEY
.	O
By	O
making	O
some	O
intuitive	O
observations	O
,	O
most	O
classes	O
that	O
we	O
identify	O
can	O
be	O
derived	O
quite	O
easily	O
from	O
some	O
subsets	O
of	O
default	B-KEY
logic	I-KEY
already	O
known	O
in	O
the	O
literature	O
.	O
Some	O
of	O
the	O
subsets	O
we	O
discuss	O
are	O
subclasses	O
of	O
the	O
so-called	O
``	O
extended	B-KEY
logic	I-KEY
programs	I-KEY
''	O
.	O
All	O
the	O
tractable	B-KEY
subsets	I-KEY
presented	O
in	O
this	O
paper	O
can	O
be	O
recognized	O
in	O
linear	O
time	O

Work	O
in	O
progress	O
:	O
Developing	O
policies	O
for	O
access	O
to	O
government	B-KEY
information	I-KEY
in	O
the	O
New	O
South	B-KEY
Africa	I-KEY
Following	O
South	B-KEY
Africa	I-KEY
's	O
transition	O
to	O
democracy	O
in	O
1994	O
,	O
the	O
SA	O
government	O
has	O
adopted	O
policies	O
supporting	O
freedom	B-KEY
of	I-KEY
expression	I-KEY
and	O
freedom	B-KEY
of	I-KEY
access	I-KEY
to	I-KEY
information	I-KEY
.	O
The	O
Bill	B-KEY
of	I-KEY
Rights	I-KEY
in	O
the	O
new	O
Constitution	O
includes	O
a	O
constitutional	B-KEY
right	I-KEY
of	I-KEY
access	I-KEY
to	O
information	O
held	O
by	O
the	O
state	O
.	O
Since	O
1994	O
various	O
initiatives	O
have	O
been	O
taken	O
by	O
government	O
and	O
other	O
bodies	O
to	O
promote	O
such	O
access	O
.	O
These	O
include	O
moves	O
to	O
reorganize	O
government	B-KEY
printing	I-KEY
and	O
publishing	O
,	O
restructure	O
the	O
government	O
's	O
public	B-KEY
information	I-KEY
services	I-KEY
,	O
make	O
government	B-KEY
information	I-KEY
available	O
on	O
the	O
Internet	B-KEY
,	O
and	O
extend	O
telephony	O
and	O
Internet	B-KEY
access	O
to	O
poor	O
communities	O
.	O
SA	O
's	O
new	O
Legal	O
Deposit	O
Act	O
,	O
-LRB-	O
1997	O
-RRB-	O
makes	O
provision	O
for	O
the	O
creation	O
of	O
official	B-KEY
publications	I-KEY
depositories	I-KEY
.	O
The	O
Promotion	O
of	O
Access	O
to	O
Information	O
Act	O
,	O
-LRB-	O
2000	O
-RRB-	O
was	O
enacted	O
to	O
ensure	O
access	O
to	O
information	O
held	O
by	O
the	O
state	O
and	O
public	B-KEY
bodies	I-KEY
.	O
However	O
,	O
despite	O
much	O
activity	O
,	O
it	O
has	O
proved	O
difficult	O
to	O
translate	O
principles	O
into	O
practical	O
and	O
well-coordinated	O
measures	O
to	O
improve	O
access	O
to	O
government	B-KEY
information	I-KEY
.	O
A	O
specific	O
concern	O
is	O
the	O
failure	O
of	O
policy-makers	O
to	O
visualize	O
a	O
role	O
for	O
libraries	B-KEY

Using	O
fractional	B-KEY
order	I-KEY
adjustment	I-KEY
rules	I-KEY
and	O
fractional	B-KEY
order	I-KEY
reference	I-KEY
models	I-KEY
in	O
model-reference	B-KEY
adaptive	I-KEY
control	I-KEY
This	O
paper	O
investigates	O
the	O
use	O
of	O
Fractional	O
Order	O
Calculus	O
-LRB-	O
FOC	B-KEY
-RRB-	O
in	O
conventional	O
Model	O
Reference	O
Adaptive	O
Control	O
-LRB-	O
MRAC	B-KEY
-RRB-	O
systems	O
.	O
Two	O
modifications	O
to	O
the	O
conventional	O
MRAC	B-KEY
are	O
presented	O
,	O
i.e.	O
,	O
the	O
use	O
of	O
fractional	O
order	O
parameter	O
adjustment	O
rule	O
and	O
the	O
employment	O
of	O
fractional	B-KEY
order	I-KEY
reference	I-KEY
model	I-KEY
.	O
Through	O
examples	O
,	O
benefits	O
from	O
the	O
use	O
of	O
FOC	B-KEY
are	O
illustrated	O
together	O
with	O
some	O
remarks	O
for	O
further	O
research	O

The	O
disconnect	O
continues	O
-LSB-	O
digital	B-KEY
content	I-KEY
providers	O
-RSB-	O
The	O
relationships	O
between	O
the	O
people	O
who	O
buy	O
digital	B-KEY
content	I-KEY
and	O
those	O
who	O
sell	O
it	O
are	O
probably	O
more	O
acrimonious	O
than	O
ever	O
before	O
,	O
says	O
Dick	O
Curtis	O
,	O
a	O
director	O
and	O
lead	O
analyst	O
for	O
the	O
research	O
firm	O
Outsell	O
Inc.	O
,	O
where	O
he	O
covers	O
econtent	O
contract	B-KEY
and	I-KEY
negotiation	O
strategies	O
.	O
Several	O
buyers	O
agree	O
with	O
his	O
observation	O
.	O
They	O
cite	O
aggressive	O
sales	B-KEY
tactics	I-KEY
,	O
an	O
unwillingness	O
to	O
deliver	O
content	O
in	O
formats	O
buyers	O
need	O
,	O
a	O
reluctance	O
to	O
provide	O
licensing	O
terms	O
that	O
take	O
into	O
account	O
the	O
structure	O
of	O
today	O
's	O
corporations	O
,	O
and	O
inadequate	O
service	O
and	O
support	O
as	O
a	O
few	O
of	O
the	O
factors	O
underlying	O
the	O
acrimony	O
.	O
Still	O
,	O
many	O
buyers	O
remain	O
optimistic	O
that	O
compromises	O
can	O
be	O
reached	O
on	O
some	O
of	O
these	O
issues	O
.	O
But	O
first	O
,	O
they	O
say	O
,	O
sellers	O
must	O
truly	O
understand	O
the	O
econtent	O
needs	O
of	O
today	O
's	O
enterprises	O

Bit-serial	O
AB/sup	O
2	O
/	O
multiplier	O
using	O
modified	B-KEY
inner	I-KEY
product	I-KEY
This	O
paper	O
presents	O
a	O
new	O
multiplication	B-KEY
algorithm	I-KEY
and	O
,	O
based	O
on	O
this	O
algorithm	O
,	O
proposes	O
a	O
hardware	B-KEY
architecture	I-KEY
,	O
called	O
modified	B-KEY
inner-product	I-KEY
multiplier	I-KEY
-LRB-	O
MIPM	O
-RRB-	O
,	O
which	O
computes	O
AB/sup	O
2	O
/	O
multiplication	O
based	O
on	O
a	O
linear	B-KEY
feedback	I-KEY
shift	I-KEY
register	I-KEY
-LRB-	O
LFSR	O
-RRB-	O
.	O
The	O
algorithm	O
is	O
based	O
on	O
the	O
property	O
of	O
the	O
irreducible	B-KEY
all	I-KEY
one	I-KEY
polynomial	I-KEY
-LRB-	O
AOP	O
-RRB-	O
over	O
the	O
finite	O
field	O
GF	O
-LRB-	O
2/sup	O
m	O
/	O
-RRB-	O
.	O
The	O
proposed	O
architecture	O
reduces	O
the	O
time	O
and	O
space	B-KEY
complexity	I-KEY
for	O
computing	O
AB/sup	O
2	O
/	O
.	O
The	O
proposed	O
architecture	O
has	O
a	O
potential	O
application	O
to	O
implementing	O
exponentiation	O
architecture	O
for	O
a	O
public-key	B-KEY
cryptosystem	I-KEY

SBC	O
gets	O
more	O
serious	O
on	O
regulatory	B-KEY
compliance	I-KEY
With	O
one	O
eye	O
on	O
the	O
past	O
and	O
the	O
other	O
on	O
its	O
future	O
,	O
SBC	B-KEY
Communications	I-KEY
last	O
week	O
created	O
a	O
unit	O
it	O
hopes	O
will	O
bring	O
a	O
cohesiveness	O
and	O
efficiency	O
to	O
its	O
regulatory	B-KEY
compliance	I-KEY
efforts	O
that	O
previously	O
had	O
been	O
lacking	O
.	O
The	O
carrier	O
also	O
hopes	O
the	O
new	O
regulatory	B-KEY
compliance	I-KEY
unit	O
will	O
help	O
it	O
accomplish	O
its	O
short-term	O
goal	O
of	O
landing	O
FCC	O
approval	O
.	O
to	O
provide	O
long-distance	O
service	O
throughout	O
its	O
region	O
,	O
and	O
its	O
longer-term	O
,	O
goal	O
of	O
reducing	O
the	O
regulatory	O
burdens	O
under	O
which	O
it	O
and	O
currently	O
operate	O

Time-integration	O
of	O
multiphase	B-KEY
chemistry	I-KEY
in	O
size-resolved	B-KEY
cloud	I-KEY
models	I-KEY
The	O
existence	O
of	O
cloud	B-KEY
drops	I-KEY
leads	O
to	O
a	O
transfer	O
of	O
chemical	B-KEY
species	I-KEY
between	O
the	O
gas	O
and	O
aqueous	O
phases	O
.	O
Species	O
concentrations	O
in	O
both	O
phases	O
are	O
modified	O
by	O
chemical	B-KEY
reactions	I-KEY
and	O
by	O
this	O
phase	O
transfer	O
.	O
The	O
model	O
equations	O
resulting	O
from	O
such	O
multiphase	B-KEY
chemical	I-KEY
systems	I-KEY
are	O
nonlinear	O
,	O
highly	O
coupled	O
and	O
extremely	O
stiff	O
.	O
In	O
the	O
paper	O
we	O
investigate	O
several	O
numerical	O
approaches	O
for	O
treating	O
such	O
processes	O
.	O
The	O
droplets	O
are	O
subdivided	O
into	O
several	O
classes	O
.	O
This	O
decomposition	O
of	O
the	O
droplet	O
spectrum	O
into	O
classes	O
is	O
based	O
on	O
their	O
droplet	O
size	O
and	O
the	O
amount	O
of	O
scavenged	O
material	O
inside	O
the	O
drops	O
,	O
respectively	O
.	O
The	O
very	O
fast	O
dissociations	O
in	O
the	O
aqueous	B-KEY
phase	I-KEY
chemistry	I-KEY
are	O
treated	O
as	O
forward	O
and	O
backward	O
reactions	O
.	O
The	O
aqueous	O
phase	O
and	O
gas	B-KEY
phase	I-KEY
chemistry	I-KEY
,	O
the	O
mass	O
transfer	O
between	O
the	O
different	O
droplet	O
classes	O
among	O
themselves	O
and	O
with	O
the	O
gas	O
phase	O
are	O
integrated	O
in	O
an	O
implicit	O
and	O
coupled	O
manner	O
by	O
the	O
second	O
order	O
BDF	O
method	O
.	O
For	O
this	O
part	O
we	O
apply	O
a	O
modification	O
of	O
the	O
code	O
LSODE	O
with	O
special	O
linear	O
system	O
solvers	O
.	O
These	O
direct	O
sparse	O
techniques	O
exploit	O
the	O
special	O
block	O
structure	O
of	O
the	O
corresponding	O
Jacobian	O
.	O
Furthermore	O
we	O
investigate	O
an	O
approximate	B-KEY
matrix	I-KEY
factorization	I-KEY
which	O
is	O
related	O
to	O
operator	B-KEY
splitting	I-KEY
at	O
the	O
linear	B-KEY
algebra	I-KEY
level	O
.	O
The	O
sparse	B-KEY
Jacobians	I-KEY
are	O
generated	O
explicitly	O
and	O
stored	O
in	O
a	O
sparse	O
form	O
.	O
The	O
efficiency	O
and	O
accuracy	O
of	O
our	O
time-integration	B-KEY
schemes	I-KEY
is	O
discussed	O
for	O
four	O
multiphase	B-KEY
chemistry	I-KEY
systems	O
of	O
different	O
complexity	O
and	O
for	O
a	O
different	O
number	O
of	O
droplet	O
classes	O

Recruitment	B-KEY
and	O
retention	B-KEY
of	O
women	B-KEY
graduate	I-KEY
students	I-KEY
in	O
computer	B-KEY
science	I-KEY
and	O
engineering	B-KEY
:	O
results	O
of	O
a	O
workshop	O
organized	O
by	O
the	O
Computing	B-KEY
Research	I-KEY
Association	I-KEY
This	O
document	O
is	O
the	O
report	O
of	O
a	O
workshop	O
that	O
convened	O
a	O
group	O
of	O
experts	O
to	O
discuss	O
the	O
recruitment	B-KEY
and	O
retention	B-KEY
of	O
women	O
in	O
computer	B-KEY
science	I-KEY
and	O
engineering	B-KEY
-LRB-	O
CSE	O
-RRB-	O
graduate	O
programs	O
.	O
Participants	O
included	O
long-time	O
members	O
of	O
the	O
CSE	O
academic	O
and	O
research	B-KEY
communities	I-KEY
,	O
social	B-KEY
scientists	I-KEY
engaged	O
in	O
relevant	O
research	O
,	O
and	O
directors	B-KEY
of	O
successful	O
retention	B-KEY
efforts	O
.	O
The	O
report	O
is	O
a	O
compendium	O
of	O
the	O
experience	O
and	O
expertise	O
of	O
workshop	B-KEY
participants	I-KEY
,	O
rather	O
than	O
the	O
result	O
of	O
a	O
full-scale	O
,	O
scholarly	O
study	O
into	O
the	O
range	O
of	O
issues	O
.	O
Its	O
goal	O
is	O
to	O
provide	O
departments	O
with	O
practical	O
advice	O
on	O
recruitment	B-KEY
and	O
retention	B-KEY
in	O
the	O
form	O
of	O
a	O
set	O
of	O
specific	O
recommendations	O

Hysteretic	O
threshold	O
logic	O
and	O
quasi-delay	O
insensitive	O
asynchronous	O
design	O
We	O
introduce	O
the	O
class	O
of	O
hysteretic	O
linear-threshold	O
-LRB-	O
HLT	O
-RRB-	O
logic	O
functions	O
as	O
a	O
novel	O
extension	O
of	O
linear	O
threshold	O
logic	O
,	O
and	O
prove	O
their	O
general	O
applicability	O
for	O
constructing	O
state-holding	B-KEY
Boolean	I-KEY
functions	I-KEY
.	O
We	O
then	O
demonstrate	O
a	O
fusion	O
of	O
HLT	B-KEY
logic	I-KEY
with	O
the	O
quasi-delay	B-KEY
insensitive	I-KEY
style	I-KEY
of	O
asynchronous	B-KEY
circuit	I-KEY
design	I-KEY
,	O
complete	O
with	O
logical	B-KEY
design	I-KEY
examples	O
.	O
Future	O
research	O
directions	O
are	O
also	O
identified	O

Evaluation	O
of	O
existing	O
and	O
new	O
feature	B-KEY
recognition	I-KEY
algorithms	I-KEY
.	O
2	O
.	O
Experimental	O
results	O
For	O
pt	O
.1	O
see	O
ibid.	O
,	O
p.839-851	O
.	O
This	O
is	O
the	O
second	O
of	O
two	O
papers	O
investigating	O
the	O
performance	O
of	O
general-purpose	B-KEY
feature	I-KEY
detection	I-KEY
techniques	I-KEY
.	O
The	O
first	O
paper	O
describes	O
the	O
development	O
of	O
a	O
methodology	O
to	O
synthesize	O
possible	O
general	O
feature	O
detection	O
face	B-KEY
sets	I-KEY
.	O
Six	O
algorithms	O
resulting	O
from	O
the	O
synthesis	O
have	O
been	O
designed	O
and	O
implemented	O
on	O
a	O
SUN	O
Workstation	O
in	O
C++	O
using	O
ACIS	O
as	O
the	O
geometric	O
modelling	O
system	O
.	O
In	O
this	O
paper	O
,	O
extensive	O
tests	O
and	O
comparative	O
analysis	O
are	O
conducted	O
on	O
the	O
feature	O
detection	O
algorithms	O
,	O
using	O
carefully	O
selected	O
components	O
from	O
the	O
public	O
domain	O
,	O
mostly	O
from	O
the	O
National	B-KEY
Design	I-KEY
Repository	I-KEY
.	O
The	O
results	O
show	O
that	O
the	O
new	O
and	O
enhanced	O
algorithms	O
identify	O
face	B-KEY
sets	I-KEY
that	O
previously	O
published	O
algorithms	O
can	O
not	O
detect	O
.	O
The	O
tests	O
also	O
show	O
that	O
each	O
algorithm	O
can	O
detect	O
,	O
among	O
other	O
types	O
,	O
a	O
certain	O
type	O
of	O
feature	O
that	O
is	O
unique	O
to	O
it	O
.	O
Hence	O
,	O
most	O
of	O
the	O
algorithms	O
discussed	O
in	O
this	O
paper	O
would	O
have	O
to	O
be	O
combined	O
to	O
obtain	O
complete	O
coverage	O

Control	O
of	O
thin	B-KEY
film	I-KEY
growth	I-KEY
in	O
chemical	B-KEY
vapor	I-KEY
deposition	I-KEY
manufacturing	O
systems	O
:	O
a	O
feasibility	O
study	O
A	O
study	O
is	O
carried	O
out	O
to	O
design	O
and	O
optimize	B-KEY
chemical	B-KEY
vapor	I-KEY
deposition	I-KEY
-LRB-	O
CVD	O
-RRB-	O
systems	O
for	O
material	B-KEY
fabrication	I-KEY
.	O
Design	O
and	O
optimization	B-KEY
of	O
the	O
CVD	O
process	O
is	O
necessary	O
to	O
satisfying	O
strong	O
global	O
demand	O
and	O
ever	O
increasing	O
quality	O
requirements	O
for	O
thin	O
film	O
production	O
.	O
Advantages	O
of	O
computer	O
aided	O
optimization	B-KEY
include	O
high	O
design	O
turnaround	O
time	O
,	O
flexibility	O
to	O
explore	O
a	O
larger	O
design	O
space	O
and	O
the	O
development	O
and	O
adaptation	O
of	O
automation	O
techniques	O
for	O
design	O
and	O
optimization	B-KEY
.	O
A	O
CVD	O
reactor	O
consisting	O
of	O
a	O
vertical	O
impinging	O
jet	O
at	O
atmospheric	O
pressure	O
,	O
for	O
growing	O
titanium	B-KEY
nitride	I-KEY
films	I-KEY
,	O
is	O
studied	O
for	O
thin	O
film	O
deposition	O
.	O
Numerical	O
modeling	O
and	O
simulation	O
are	O
used	O
to	O
determine	O
the	O
rate	O
of	O
deposition	O
and	O
film	O
uniformity	O
over	O
a	O
wide	O
range	O
of	O
design	O
variables	O
and	O
operating	O
conditions	O
.	O
These	O
results	O
are	O
used	O
for	O
system	O
design	O
and	O
optimization	B-KEY
.	O
The	O
optimization	B-KEY
procedure	O
employs	O
an	O
objective	O
function	O
characterizing	O
film	B-KEY
quality	I-KEY
,	O
productivity	O
and	O
operational	B-KEY
costs	I-KEY
based	O
on	O
reactor	B-KEY
gas	I-KEY
flow	I-KEY
rate	I-KEY
,	O
susceptor	B-KEY
temperature	I-KEY
and	O
precursor	B-KEY
concentration	I-KEY
.	O
Parameter	B-KEY
space	I-KEY
mappings	I-KEY
are	O
used	O
to	O
determine	O
the	O
design	O
space	O
,	O
while	O
a	O
minimization	O
algorithm	O
,	O
such	O
as	O
the	O
steepest	O
descent	O
method	O
,	O
is	O
used	O
to	O
determine	O
optimal	B-KEY
operating	O
conditions	O
for	O
the	O
system	O
.	O
The	O
main	O
features	O
of	O
computer	O
aided	O
design	O
and	O
optimization	B-KEY
using	O
these	O
techniques	O
are	O
discussed	O
in	O
detail	O

Semantic	B-KEY
data	I-KEY
broadcast	I-KEY
for	O
a	O
mobile	O
environment	O
based	O
on	O
dynamic	O
and	O
adaptive	B-KEY
chunking	I-KEY
Database	O
broadcast	O
is	O
an	O
effective	O
and	O
scalable	O
approach	O
to	O
disseminate	O
information	O
of	O
high	O
affinity	O
to	O
a	O
large	O
collection	O
of	O
mobile	B-KEY
clients	I-KEY
.	O
A	O
common	O
problem	O
of	O
existing	O
broadcast	O
approaches	O
is	O
the	O
lack	O
of	O
knowledge	O
for	O
a	O
client	O
to	O
determine	O
if	O
all	O
data	O
items	O
satisfying	O
its	O
query	O
could	O
be	O
obtained	O
from	O
the	O
broadcast	O
.	O
We	O
therefore	O
propose	O
a	O
semantic-based	O
broadcast	O
approach	O
.	O
A	O
semantic	B-KEY
descriptor	I-KEY
is	O
attached	O
to	O
each	O
broadcast	O
unit	O
,	O
called	O
a	O
data	B-KEY
chunk	I-KEY
.	O
This	O
semantic	B-KEY
descriptor	I-KEY
allows	O
a	O
client	O
to	O
determine	O
if	O
a	O
query	O
can	O
be	O
answered	B-KEY
entirely	O
based	O
on	O
broadcast	O
items	O
and	O
,	O
if	O
needed	O
,	O
identify	O
the	O
precise	O
definition	O
of	O
the	O
remaining	O
items	O
in	O
the	O
form	O
of	O
a	O
``	O
supplementary	O
''	O
query	O
.	O
Data	B-KEY
chunks	I-KEY
can	O
be	O
of	O
static	O
or	O
dynamic	O
sizes	O
and	O
organized	O
hierarchically	O
.	O
Their	O
boundary	O
can	O
be	O
determined	O
on-the-fly	O
,	O
adaptive	O
to	O
the	O
nature	O
of	O
client	O
queries	O
.	O
We	O
investigate	O
different	O
ways	O
of	O
organizing	O
the	O
data	B-KEY
chunks	I-KEY
over	O
a	O
broadcast	O
channel	O
to	O
improve	O
access	O
performance	O
.	O
We	O
introduce	O
the	O
data	B-KEY
affinity	I-KEY
index	I-KEY
metric	O
,	O
which	O
more	O
accurately	O
reflects	O
client-perceived	O
performance	O
.	O
A	O
simulation	O
model	O
is	O
built	O
to	O
evaluate	O
our	O
semantic-based	O
broadcast	O
schemes	O

Dot-Net	O
makes	O
slow	O
progress	O
Microsoft	O
's	O
Windows	O
.	O
Net	O
Enterprise	O
Server	O
Release	O
Candidate	O
I	O
,	O
which	O
was	O
released	O
at	O
the	O
end	O
of	O
last	O
month	O
,	O
provides	O
an	O
early	O
glimpse	O
of	O
the	O
system	O
that	O
will	O
eventually	O
replace	O
Windows	O
200	O
Advanced	O
Server	O
.	O
The	O
software	O
has	O
been	O
improved	O
so	O
that	O
Active	B-KEY
Directory	I-KEY
is	O
more	O
flexible	O
and	O
easier	O
to	O
deploy	O
;	O
and	O
security	B-KEY
,	O
scalability	B-KEY
and	O
management	O
have	O
also	O
been	O
enhanced	O

Cooperative	B-KEY
mutation	I-KEY
based	I-KEY
evolutionary	I-KEY
programming	I-KEY
for	O
continuous	B-KEY
function	I-KEY
optimization	I-KEY
An	O
evolutionary	O
programming	O
-LRB-	O
EP	O
-RRB-	O
algorithm	O
adapting	O
a	O
new	O
mutation	O
operator	O
is	O
presented	O
.	O
Unlike	O
most	O
previous	O
EPs	O
,	O
in	O
which	O
each	O
individual	O
is	O
mutated	O
on	O
its	O
own	O
,	O
each	O
individual	O
in	O
the	O
proposed	O
algorithm	O
is	O
mutated	O
in	O
cooperation	O
with	O
the	O
other	O
individuals	O
.	O
This	O
not	O
only	O
enhances	O
convergence	B-KEY
speed	I-KEY
but	O
also	O
gives	O
more	O
chance	O
to	O
escape	O
from	O
local	B-KEY
minima	I-KEY

Dynamic	O
testing	O
of	O
inflatable	B-KEY
structures	I-KEY
using	O
smart	B-KEY
materials	I-KEY
In	O
this	O
paper	O
we	O
present	O
experimental	O
investigations	O
of	O
the	O
vibration	O
testing	O
of	O
an	O
inflated	O
,	O
thin-film	B-KEY
torus	I-KEY
using	O
smart	B-KEY
materials	I-KEY
.	O
Lightweight	O
,	O
inflatable	B-KEY
structures	I-KEY
are	O
very	O
attractive	O
in	O
satellite	B-KEY
applications	I-KEY
.	O
However	O
,	O
the	O
lightweight	O
,	O
flexible	O
and	O
highly	O
damped	O
nature	O
of	O
inflated	B-KEY
structures	I-KEY
poses	O
difficulties	O
in	O
ground	B-KEY
vibration	I-KEY
testing	I-KEY
.	O
In	O
this	O
study	O
,	O
we	O
show	O
that	O
polyvinylidene	O
fluoride	O
-LRB-	O
PVDF	O
-RRB-	O
patches	O
and	O
recently	O
developed	O
macro-fiber	B-KEY
composite	I-KEY
actuators	I-KEY
may	O
be	O
used	O
as	O
sensors	O
and	O
actuators	O
in	O
identifying	O
modal	B-KEY
parameters	I-KEY
.	O
Both	O
smart	B-KEY
materials	I-KEY
can	O
be	O
integrated	O
unobtrusively	O
into	O
the	O
skin	O
of	O
a	O
torus	O
or	O
space	B-KEY
device	I-KEY
forming	O
an	O
attractive	O
testing	O
arrangement	O
.	O
The	O
addition	O
of	O
actuators	O
and	O
PVDF	B-KEY
sensors	I-KEY
to	O
the	O
torus	O
does	O
not	O
significantly	O
interfere	O
with	O
the	O
suspension	O
modes	O
of	O
a	O
free-free	O
boundary	B-KEY
condition	I-KEY
,	O
and	O
can	O
be	O
considered	O
an	O
integral	O
part	O
of	O
the	O
inflated	B-KEY
structure	I-KEY
.	O
The	O
results	O
indicate	O
the	O
potential	O
of	O
using	O
smart	B-KEY
materials	I-KEY
to	O
measure	O
and	O
control	O
the	O
dynamic	B-KEY
response	I-KEY
of	O
inflated	B-KEY
structures	I-KEY

Using	O
constructed	O
types	O
in	O
C++	O
unions	O
The	O
C++	B-KEY
Standard	I-KEY
states	O
that	O
a	O
union	B-KEY
type	I-KEY
can	O
not	O
have	O
a	O
member	O
with	O
a	O
nontrivial	O
constructor	B-KEY
or	O
destructor	B-KEY
.	O
While	O
at	O
first	O
this	O
seems	O
unreasonable	O
,	O
further	O
thought	O
makes	O
it	O
clear	O
why	O
this	O
is	O
the	O
case	O
:	O
The	O
crux	O
of	O
the	O
problem	O
is	O
that	O
unions	O
do	O
n't	O
have	O
built-in	O
semantics	O
for	O
denoting	O
when	O
a	O
member	O
is	O
the	O
``	O
current	O
''	O
member	O
of	O
the	O
union	O
.	O
Therefore	O
,	O
the	O
compiler	O
ca	O
n't	O
know	O
when	O
it	O
's	O
appropriate	O
to	O
call	O
constructors	B-KEY
or	O
destructors	B-KEY
on	O
the	O
union	B-KEY
members	I-KEY
.	O
Still	O
,	O
there	O
are	O
good	O
reasons	O
for	O
wanting	O
to	O
use	O
constructed	O
object	O
types	O
in	O
a	O
union	O
.	O
For	O
example	O
,	O
you	O
might	O
want	O
to	O
implement	O
a	O
scripting	B-KEY
language	I-KEY
with	O
a	O
single	O
variable	O
type	O
that	O
can	O
either	O
be	O
an	O
integer	O
,	O
a	O
string	O
,	O
or	O
a	O
list	O
.	O
A	O
union	O
is	O
the	O
perfect	O
candidate	O
for	O
implementing	O
such	O
a	O
composite	O
type	O
,	O
but	O
the	O
restriction	O
on	O
constructed	O
union	B-KEY
members	I-KEY
may	O
prevent	O
you	O
from	O
using	O
an	O
existing	O
string	O
or	O
list	O
class	O
-LRB-	O
for	O
example	O
,	O
from	O
the	O
STL	O
-RRB-	O
to	O
provide	O
the	O
underlying	O
functionality	O
.	O
Luckily	O
,	O
a	O
feature	O
of	O
C++	O
called	O
placement	B-KEY
new	I-KEY
can	O
provide	O
a	O
workaround	O

Evolving	O
receptive-field	B-KEY
controllers	I-KEY
for	O
mobile	B-KEY
robots	I-KEY
The	O
use	O
of	O
evolutionary	B-KEY
methods	I-KEY
to	O
generate	O
controllers	O
for	O
real-world	B-KEY
autonomous	I-KEY
agents	I-KEY
has	O
attracted	O
attention	O
.	O
Most	O
of	O
the	O
pertinent	O
research	O
has	O
employed	O
genetic	O
algorithms	O
or	O
variations	O
thereof	O
.	O
Research	O
has	O
applied	O
an	O
alternative	O
evolutionary	B-KEY
method	I-KEY
,	O
evolution	B-KEY
strategies	I-KEY
,	O
to	O
the	O
generation	O
of	O
simple	B-KEY
Braitenberg	I-KEY
vehicles	I-KEY
.	O
This	O
application	O
accelerates	O
the	O
development	O
of	O
such	O
controllers	O
by	O
more	O
than	O
an	O
order	O
of	O
magnitude	O
-LRB-	O
a	O
few	O
hours	O
compared	O
to	O
more	O
than	O
two	O
days	O
-RRB-	O
.	O
Motivated	O
by	O
this	O
useful	O
speedup	O
,	O
the	O
paper	O
investigates	O
the	O
evolution	O
of	O
more	O
complex	O
architectures	O
,	O
receptive-field	B-KEY
controllers	I-KEY
,	O
that	O
can	O
employ	O
nonlinear	B-KEY
interactions	I-KEY
and	O
,	O
therefore	O
,	O
can	O
yield	O
more	O
complex	B-KEY
behavior	I-KEY
.	O
It	O
is	O
interesting	O
to	O
note	O
that	O
the	O
evolution	B-KEY
strategy	I-KEY
yields	O
the	O
same	O
efficacy	O
in	O
terms	O
of	O
function	O
evaluations	O
,	O
even	O
though	O
the	O
second	O
class	O
of	O
controllers	O
requires	O
up	O
to	O
10	O
times	O
more	O
parameters	O
than	O
the	O
simple	O
Braitenberg	O
architecture	O
.	O
In	O
addition	O
to	O
the	O
speedup	O
,	O
there	O
is	O
an	O
important	O
theoretical	O
reason	O
for	O
preferring	O
an	O
evolution	B-KEY
strategy	I-KEY
over	O
a	O
genetic	O
algorithm	O
for	O
this	O
problem	O
,	O
namely	O
the	O
presence	O
of	O
epistasis	O

Portal	B-KEY
dose	I-KEY
image	I-KEY
prediction	I-KEY
for	O
dosimetric	B-KEY
treatment	I-KEY
verification	I-KEY
in	O
radiotherapy	B-KEY
.	O
II	O
.	O
An	O
algorithm	O
for	O
wedged	O
beams	O
A	O
method	O
is	O
presented	O
for	O
calculation	O
of	O
a	O
two-dimensional	B-KEY
function	I-KEY
,	O
T/sub	O
wedge	O
/	O
-LRB-	O
x	O
,	O
y	O
-RRB-	O
,	O
describing	O
the	O
transmission	O
of	O
a	O
wedged	B-KEY
photon	I-KEY
beam	I-KEY
through	O
a	O
patient	O
.	O
This	O
in	O
an	O
extension	O
of	O
the	O
method	O
that	O
we	O
have	O
published	O
for	O
open	O
-LRB-	O
nonwedged	O
-RRB-	O
fields	O
-LSB-	O
Med	O
.	O
Phys	O
.	O
25	O
,	O
830-840	O
-LRB-	O
1998	O
-RRB-	O
-RSB-	O
.	O
Transmission	O
functions	O
for	O
open	O
fields	O
are	O
being	O
used	O
in	O
our	O
clinic	O
for	O
prediction	O
of	O
portal	O
dose	O
images	O
-LRB-	O
PDI	O
,	O
i.e.	O
,	O
a	O
dose	O
distribution	O
behind	O
the	O
patient	O
in	O
a	O
plane	O
normal	O
to	O
the	O
beam	O
axis	O
-RRB-	O
,	O
which	O
are	O
compared	O
with	O
PDIs	O
measured	O
with	O
an	O
electronic	B-KEY
portal	I-KEY
imaging	I-KEY
device	I-KEY
-LRB-	O
EPID	O
-RRB-	O
.	O
The	O
calculations	O
are	O
based	O
on	O
the	O
planning	B-KEY
CT	I-KEY
scan	I-KEY
of	O
the	O
patient	O
and	O
on	O
the	O
irradiation	B-KEY
geometry	I-KEY
as	O
determined	O
in	O
the	O
treatment	O
planning	O
process	O
.	O
Input	O
data	O
for	O
the	O
developed	O
algorithm	O
for	O
wedged	O
beams	O
are	O
derived	O
from	O
-LRB-	O
the	O
already	O
available	O
-RRB-	O
measured	O
input	O
data	O
set	O
for	O
transmission	O
prediction	O
in	O
open	B-KEY
beams	I-KEY
,	O
which	O
is	O
extended	O
with	O
only	O
a	O
limited	O
set	O
of	O
measurements	O
in	O
the	O
wedged	O
beam	O
.	O
The	O
method	O
has	O
been	O
tested	O
for	O
a	O
PDI	O
plane	O
at	O
160	O
cm	O
from	O
the	O
focus	O
,	O
in	O
agreement	O
with	O
the	O
applied	O
focus-to-detector	O
distance	O
of	O
our	O
fluoroscopic	O
EPIDs	O
.	O
For	O
low	O
and	O
high	B-KEY
energy	I-KEY
photon	I-KEY
beams	I-KEY
-LRB-	O
6	O
and	O
23	B-KEY
MV	I-KEY
-RRB-	O
good	O
agreement	O
-LRB-	O
~	O
1	O
%	O
-RRB-	O
has	O
been	O
found	O
between	O
calculated	O
and	O
measured	O
transmissions	O
for	O
a	O
slab	O
and	O
a	O
thorax	B-KEY
phantom	I-KEY

The	O
decision	B-KEY
procedure	I-KEY
for	O
profitability	B-KEY
of	O
investment	O
projects	O
using	O
the	O
internal	B-KEY
rate	I-KEY
of	I-KEY
return	I-KEY
of	O
single-period	B-KEY
projects	I-KEY
The	O
internal	B-KEY
rate	I-KEY
of	I-KEY
return	I-KEY
-LRB-	O
IRR	O
-RRB-	O
criterion	O
is	O
often	O
used	O
to	O
evaluate	O
profitability	B-KEY
of	O
investment	O
projects	O
.	O
In	O
this	O
paper	O
,	O
we	O
focus	O
on	O
a	O
single-period	B-KEY
project	I-KEY
which	O
consists	O
of	O
two	O
types	O
of	O
cash	B-KEY
flows	I-KEY
;	O
an	O
investment	O
at	O
one	O
period	O
and	O
a	O
return	O
at	O
a	O
succeeding	O
period	O
,	O
and	O
a	O
financing	O
at	O
one	O
period	O
and	O
a	O
repayment	O
at	O
a	O
succeeding	O
period	O
.	O
We	O
decompose	O
the	O
given	O
investment	O
project	O
into	O
a	O
series	O
of	O
the	O
single-period	B-KEY
projects	I-KEY
.	O
From	O
the	O
viewpoint	O
of	O
the	O
single-period	B-KEY
project	I-KEY
,	O
we	O
point	O
out	O
the	O
applicability	O
issue	O
of	O
the	O
IRR	B-KEY
criterion	I-KEY
,	O
namely	O
the	O
IRR	B-KEY
criterion	I-KEY
can	O
not	O
be	O
applied	O
in	O
which	O
a	O
project	O
is	O
composed	O
of	O
both	O
investment	O
type	O
and	O
financing	O
type	O
.	O
Investigating	O
the	O
properties	O
of	O
a	O
series	O
of	O
the	O
single-period	B-KEY
projects	I-KEY
,	O
we	O
resolve	O
the	O
applicability	O
issue	O
of	O
the	O
IRR	B-KEY
criterion	I-KEY
and	O
propose	O
the	O
decision	B-KEY
procedure	I-KEY
for	O
profitability	B-KEY
judgment	O
toward	O
any	O
type	O
of	O
investment	O
project	O
based	O
on	O
the	O
comparison	O
between	O
the	O
IRR	O
and	O
the	O
capital	O
cost	O
.	O
We	O
develop	O
a	O
new	O
algorithm	O
to	O
obtain	O
the	O
value	O
of	O
the	O
project	B-KEY
investment	I-KEY
rate	I-KEY
-LRB-	O
PIR	B-KEY
-RRB-	O
for	O
the	O
given	O
project	O
,	O
which	O
is	O
a	O
function	O
of	O
the	O
capital	O
cost	O
,	O
only	O
using	O
the	O
standard	O
IRR	O
computing	O
routine	O
.	O
This	O
outcome	O
is	O
a	O
theoretical	O
breakthrough	O
to	O
widen	O
the	O
utilization	O
of	O
IRR	O
in	O
practical	O
applications	O

Taking	O
back	O
control	O
-LSB-	O
SCADA	B-KEY
system	O
-RSB-	O
Most	O
common	O
way	O
to	O
implement	O
a	O
SCADA	B-KEY
system	O
is	O
to	O
go	O
outside	O
.	O
However	O
,	O
in	O
the	O
author	O
's	O
opinion	O
,	O
to	O
truly	O
take	O
control	O
of	O
a	O
SCADA	B-KEY
project	O
,	O
in-house	O
personnel	O
should	O
handle	O
as	O
much	O
of	O
the	O
job	O
as	O
possible	O
.	O
This	O
includes	O
design	O
,	O
equipment	O
specification	O
,	O
installation	O
,	O
and	O
programming	O
.	O
The	O
more	O
of	O
these	O
tasks	O
one	O
does	O
in-house	O
,	O
the	O
more	O
control	O
and	O
ownership	O
one	O
has	O
.	O
To	O
accomplish	O
this	O
,	O
we	O
first	O
evaluated	O
the	O
existing	O
SCADA	B-KEY
system	O
and	O
investigated	O
new	O
technologies	O
to	O
establish	O
a	O
list	O
of	O
features	O
the	O
new	O
system	O
needed	O
to	O
incorporate	O

Making	O
the	O
MIS	O
integration	O
process	O
work	O
Focused	O
,	O
cross-functional	O
teams	O
that	O
implement	O
flexible	O
and	O
scalable	B-KEY
information	I-KEY
systems	I-KEY
-LRB-	O
IS	O
-RRB-	O
can	O
deliver	O
a	O
smooth	O
,	O
lean	B-KEY
manufacturing	I-KEY
process	I-KEY
.	O
When	O
integrating	O
new	O
technology	O
into	O
an	O
existing	O
facility	O
,	O
one	O
should	O
always	O
consider	O
three	O
things	O
:	O
the	O
hard	O
infrastructure	O
,	O
the	O
soft	O
infrastructure	O
,	O
and	O
information	B-KEY
flow	I-KEY
.	O
Hard	O
infrastructure	O
includes	O
client	O
and	O
server	O
hardware	O
and	O
network	B-KEY
infrastructure	I-KEY
.	O
Soft	O
infrastructure	O
includes	O
operating	O
systems	O
,	O
existing	O
or	O
legacy	B-KEY
software	I-KEY
,	O
needed	O
code	O
customizations	O
,	O
and	O
the	O
human	B-KEY
resources	I-KEY
to	O
run/support	O
the	O
system	O
.	O
Information	B-KEY
flow	I-KEY
includes	O
how	O
data	O
in	O
the	O
new	O
system	O
interacts	O
with	O
legacy	O
systems	O
and	O
what	O
legacy	O
data	O
the	O
new	O
system	O
will	O
require	O
,	O
as	O
well	O
as	O
who	O
will	O
want	O
to	O
receive/access	O
the	O
information	O
that	O
is	O
held	O
by	O
the	O
system	O

A	O
fundamental	O
investigation	O
into	O
large	O
strain	B-KEY
recovery	I-KEY
of	O
one-way	B-KEY
shape	I-KEY
memory	I-KEY
alloy	B-KEY
wires	I-KEY
embedded	O
in	O
flexible	B-KEY
polyurethanes	I-KEY
Shape	O
memory	O
alloys	O
-LRB-	O
SMAs	O
-RRB-	O
are	O
being	O
embedded	O
in	O
or	O
externally	O
attached	O
to	O
smart	B-KEY
structures	I-KEY
because	O
of	O
the	O
large	O
amount	O
of	O
actuation	B-KEY
deformation	I-KEY
and	O
force	O
that	O
these	O
materials	O
are	O
capable	O
of	O
producing	O
when	O
they	O
are	O
heated	O
.	O
Previous	O
investigations	O
have	O
focused	O
primarily	O
on	O
using	O
single	O
or	O
opposing	O
SMA	B-KEY
wires	I-KEY
exhibiting	O
the	O
two-way	B-KEY
shape	I-KEY
memory	I-KEY
effect	I-KEY
-LRB-	O
SME	O
-RRB-	O
because	O
of	O
the	O
simplicity	O
with	O
which	O
the	O
repeatable	O
actuation	O
behavior	O
of	O
the	O
structure	O
can	O
be	O
predicted	O
.	O
This	O
repeatable	O
actuation	O
behavior	O
is	O
achieved	O
at	O
the	O
expense	O
of	O
reduced	O
levels	O
of	O
recoverable	O
deformation	B-KEY
.	O
Alternatively	O
,	O
many	O
potential	O
smart	B-KEY
structure	I-KEY
applications	O
will	O
employ	O
multiple	O
SMA	B-KEY
wires	I-KEY
exhibiting	O
a	O
permanent	O
one-way	O
SME	O
to	O
simplify	O
fabrication	O
and	O
increase	O
the	O
recoverable	B-KEY
strains	I-KEY
in	O
the	O
structure	O
.	O
To	O
employ	O
the	O
one-way	O
wires	O
,	O
it	O
is	O
necessary	O
to	O
investigate	O
how	O
they	O
affect	O
the	O
recovery	O
of	O
large	O
strains	O
when	O
they	O
are	O
embedded	O
in	O
a	O
structure	O
.	O
In	O
this	O
investigation	O
,	O
the	O
large	O
strain	B-KEY
recovery	I-KEY
of	O
a	O
one-way	O
SMA	B-KEY
wire	I-KEY
embedded	O
in	O
a	O
flexible	B-KEY
polyurethane	I-KEY
is	O
characterized	O
using	O
the	O
novel	O
deformation	B-KEY
measurement	O
technique	O
known	O
as	O
digital	O
image	O
correlation	O
.	O
These	O
results	O
are	O
compared	O
with	O
a	O
simple	O
actuation	O
model	O
and	O
a	O
three-dimensional	B-KEY
finite	I-KEY
element	I-KEY
analysis	I-KEY
of	O
the	O
structure	O
using	O
the	O
Brinson	O
model	O
for	O
describing	O
the	O
thermomechanical	O
behavior	O
of	O
the	O
SMA	O
.	O
Results	O
indicate	O
that	O
the	O
level	O
of	O
actuation	B-KEY
strain	I-KEY
in	O
the	O
structure	O
is	O
substantially	O
reduced	O
by	O
the	O
inelastic	O
behavior	O
of	O
the	O
one-way	O
SMA	B-KEY
wires	I-KEY
,	O
and	O
there	O
are	O
significant	O
differences	O
between	O
the	O
deformations	B-KEY
of	O
the	O
matrix	B-KEY
material	I-KEY
adjacent	O
to	O
the	O
SMA	B-KEY
wires	I-KEY
and	O
in	O
the	O
region	O
surrounding	O
it	O
.	O
The	O
transformation	O
behavior	O
of	O
the	O
SMA	B-KEY
wires	I-KEY
was	O
also	O
determined	O
to	O
be	O
volume	O
preserving	O
,	O
which	O
had	O
a	O
significant	O
effect	O
on	O
the	O
transverse	B-KEY
strain	I-KEY
fields	I-KEY

C	O
and	O
C++	O
:	O
a	O
case	O
for	O
compatibility	O
Modern	O
C	O
and	O
C++	O
are	O
sibling	O
languages	O
descended	O
from	O
Classic	O
C	O
.	O
In	O
many	O
people	O
's	O
minds	O
,	O
they	O
are	O
-LRB-	O
wrongly	O
,	O
but	O
understandably	O
-RRB-	O
fused	O
into	O
the	O
mythical	O
C/C	O
+	O
+	O
programming	O
language	O
.	O
There	O
is	O
no	O
C/C	O
+	O
+	O
language	O
,	O
but	O
there	O
is	O
a	O
C/C	O
+	O
+	O
community	O
.	O
Previously	O
the	O
author	O
described	O
some	O
of	O
the	O
incompatibilities	B-KEY
that	O
complicate	O
the	O
work	O
of	O
developers	O
within	O
that	O
C/C	O
+	O
+	O
community	O
.	O
In	O
this	O
article	O
,	O
he	O
discusses	O
some	O
of	O
the	O
underlying	O
myths	O
that	O
help	O
perpetuate	O
these	O
incompatibilities	B-KEY
.	O
He	O
also	O
shows	O
why	O
more	O
compatibility	O
-LRB-	O
ideally	O
,	O
full	O
compatibility	O
-RRB-	O
is	O
in	O
the	O
best	O
interest	O
of	O
the	O
C/C	O
+	O
+	O
community	O
.	O
In	O
the	O
next	O
paper	O
,	O
he	O
presents	O
some	O
examples	O
of	O
how	O
the	O
incompatibilities	B-KEY
in	O
C	O
and	O
C++	O
might	O
be	O
resolved	O

A	O
scalable	B-KEY
intelligent	O
takeoff	O
controller	O
for	O
a	O
simulated	O
running	O
jointed	O
leg	O
Running	O
with	O
jointed	O
legs	O
poses	O
a	O
difficult	O
control	O
problem	O
in	O
robotics	O
.	O
Neural	B-KEY
controllers	I-KEY
are	O
attractive	O
because	O
they	O
allow	O
the	O
robot	O
to	O
adapt	O
to	O
changing	B-KEY
environmental	I-KEY
conditions	I-KEY
.	O
However	O
,	O
scalability	B-KEY
is	O
an	O
issue	O
with	O
many	O
neural	B-KEY
controllers	I-KEY
.	O
The	O
paper	O
describes	O
the	O
development	O
of	O
a	O
scalable	B-KEY
neurofuzzy	O
controller	O
for	O
the	O
takeoff	O
phase	O
of	O
the	O
running	O
stride	O
.	O
Scalability	B-KEY
is	O
achieved	O
by	O
selecting	O
a	O
controller	O
whose	O
size	O
does	O
not	O
grow	O
with	O
the	O
dimensionality	O
of	O
the	O
problem	O
.	O
Empirical	O
results	O
show	O
that	O
with	O
proper	O
design	O
the	O
takeoff	O
controller	O
scales	O
from	O
a	O
leg	O
with	O
a	O
single	O
movable	O
link	O
to	O
one	O
with	O
three	O
movable	O
links	O
without	O
a	O
corresponding	O
growth	O
in	O
size	O
and	O
without	O
a	O
loss	O
of	O
accuracy	O

Favorable	O
noise	B-KEY
uniformity	I-KEY
properties	I-KEY
of	O
Fourier-based	B-KEY
interpolation	I-KEY
and	O
reconstruction	B-KEY
approaches	I-KEY
in	O
single-slice	B-KEY
helical	I-KEY
computed	I-KEY
tomography	I-KEY
Volumes	O
reconstructed	O
by	O
standard	O
methods	O
from	O
single-slice	B-KEY
helical	I-KEY
computed	I-KEY
tomography	I-KEY
-LRB-	O
CT	O
-RRB-	O
data	O
have	O
been	O
shown	O
to	O
have	O
noise	O
levels	O
that	O
are	O
highly	O
nonuniform	O
relative	O
to	O
those	O
in	O
conventional	O
CT.	O
.	O
These	O
noise	O
nonuniformities	O
can	O
affect	O
low-contrast	B-KEY
object	I-KEY
detectability	I-KEY
and	O
have	O
also	O
been	O
identified	O
as	O
the	O
cause	O
of	O
the	O
zebra	B-KEY
artifacts	I-KEY
that	O
plague	O
maximum	O
intensity	O
projection	O
-LRB-	O
MIP	O
-RRB-	O
images	O
of	O
such	O
volumes	O
.	O
While	O
these	O
spatially	O
variant	O
noise	O
levels	O
have	O
their	O
root	O
in	O
the	O
peculiarities	O
of	O
the	O
helical	O
scan	O
geometry	O
,	O
there	O
is	O
also	O
a	O
strong	O
dependence	O
on	O
the	O
interpolation	O
and	O
reconstruction	O
algorithms	O
employed	O
.	O
In	O
this	O
paper	O
,	O
we	O
seek	O
to	O
develop	O
image	O
reconstruction	O
strategies	O
that	O
eliminate	O
or	O
reduce	O
,	O
at	O
its	O
source	O
,	O
the	O
nonuniformity	O
of	O
noise	O
levels	O
in	O
helical	O
CT	O
relative	O
to	O
that	O
in	O
conventional	O
CT.	O
.	O
We	O
pursue	O
two	O
approaches	O
,	O
independently	O
and	O
in	O
concert	O
.	O
We	O
argue	O
,	O
and	O
verify	O
,	O
that	O
Fourier-based	O
longitudinal	O
interpolation	O
approaches	O
lead	O
to	O
more	B-KEY
uniform	I-KEY
noise	I-KEY
ratios	I-KEY
than	O
do	O
the	O
standard	O
360LI	O
and	O
180LI	O
approaches	O
.	O
We	O
also	O
demonstrate	O
that	O
a	O
Fourier-based	B-KEY
fan-to-parallel	I-KEY
rebinning	I-KEY
algorithm	I-KEY
,	O
used	O
as	O
an	O
alternative	O
to	O
fanbeam	O
filtered	O
backprojection	O
for	O
slice	O
reconstruction	O
,	O
also	O
leads	O
to	O
more	B-KEY
uniform	I-KEY
noise	I-KEY
ratios	I-KEY
,	O
even	O
when	O
making	O
use	O
of	O
the	O
180LI	O
and	O
360LI	O
interpolation	O
approaches	O

`	O
Virtual	B-KEY
Family	I-KEY
'	O
:	O
an	O
approach	O
to	O
introducing	O
Java	O
programming	O
This	O
paper	O
introduces	O
and	O
discusses	O
Virtual	B-KEY
Family	I-KEY
-LRB-	O
VF	O
-RRB-	O
:	O
a	O
gender-neutral	B-KEY
game-based	I-KEY
software	I-KEY
that	O
introduces	O
Java	O
programming	O
.	O
VF	O
provides	O
a	O
completely	O
functioning	O
game	O
that	O
students	O
extend	O
and	O
enhance	O
via	O
programming	O
.	O
We	O
discuss	O
the	O
background	O
and	O
context	O
within	O
which	O
Virtual	B-KEY
Family	I-KEY
was	O
developed	O
and	O
other	O
available	O
multimedia	B-KEY
resources	I-KEY
for	O
teaching	O
programming	O
.	O
The	O
paper	O
then	O
goes	O
on	O
to	O
describe	O
Virtual	B-KEY
Family	I-KEY
's	O
concept	O
and	O
design	O
.	O
Finally	O
,	O
feedback	O
received	O
from	O
Virtual	B-KEY
Family	I-KEY
teaching	B-KEY
workshops	I-KEY
is	O
related	O
,	O
as	O
well	O
as	O
preliminary	O
results	O
from	O
using	O
VF	O
in	O
high-school	B-KEY
teaching	I-KEY
units	I-KEY
.	O
Virtual	B-KEY
Family	I-KEY
is	O
under	O
development	O
in	O
a	O
research	O
lab	O
at	O
the	O
University	O
of	O
British	O
Columbia	O
and	O
is	O
an	O
initiative	O
of	O
Supporting	B-KEY
Women	I-KEY
in	I-KEY
Information	I-KEY
Technology	I-KEY
-LRB-	O
SWIFT	O
-RRB-	O
.	O
SWIFT	O
is	O
a	O
five-year	O
research	O
action	O
and	O
implementation	O
project	O
to	O
increase	O
the	O
participation	O
of	O
women	O
in	O
information	O
technology	O

Loop	B-KEY
restructuring	I-KEY
for	O
data	B-KEY
I/O	I-KEY
minimization	I-KEY
on	O
limited	O
on-chip	B-KEY
memory	I-KEY
embedded	B-KEY
processors	I-KEY
In	O
this	O
paper	O
,	O
we	O
propose	O
a	O
framework	O
for	O
analyzing	O
the	O
flow	O
of	O
values	O
and	O
their	O
reuse	O
in	O
loop	O
nests	O
to	O
minimize	O
data	B-KEY
traffic	I-KEY
under	O
the	O
constraints	O
of	O
limited	O
on-chip	B-KEY
memory	I-KEY
capacity	O
and	O
dependences	O
.	O
Our	O
analysis	O
first	O
undertakes	O
fusion	O
of	O
possible	O
loop	O
nests	O
intra-procedurally	O
and	O
then	O
performs	O
loop	O
distribution	O
.	O
The	O
analysis	O
discovers	O
the	O
closeness	B-KEY
factor	I-KEY
of	O
two	O
statements	O
which	O
is	O
a	O
quantitative	O
measure	O
of	O
data	B-KEY
traffic	I-KEY
saved	O
per	O
unit	O
memory	O
occupied	O
if	O
the	O
statements	O
were	O
under	O
the	O
same	O
loop	O
nest	O
over	O
the	O
case	O
where	O
they	O
are	O
under	O
different	O
loop	O
nests	O
.	O
We	O
then	O
develop	O
a	O
greedy	O
algorithm	O
which	O
traverses	O
the	O
program	B-KEY
dependence	I-KEY
graph	I-KEY
to	O
group	O
statements	O
together	O
under	O
the	O
same	O
loop	O
nest	O
legally	O
to	O
promote	O
maximal	O
reuse	O
per	O
unit	O
of	O
memory	O
occupied	O
.	O
We	O
implemented	O
our	O
framework	O
in	O
Petit	B-KEY
,	O
a	O
tool	O
for	O
dependence	O
analysis	O
and	O
loop	O
transformations	O
.	O
We	O
compared	O
our	O
method	O
with	O
one	O
based	O
on	O
tiling	O
of	O
fused	B-KEY
loop	I-KEY
nest	I-KEY
and	O
one	O
based	O
on	O
a	O
greedy	O
strategy	O
to	O
purely	O
maximize	O
reuse	O
.	O
We	O
show	O
that	O
our	O
methods	O
work	O
better	O
than	O
both	O
of	O
these	O
strategies	O
in	O
most	O
cases	O
for	O
processors	O
such	O
as	O
TMS320Cxx	O
,	O
which	O
have	O
a	O
very	O
limited	O
amount	O
of	O
on-chip	B-KEY
memory	I-KEY
.	O
The	O
improvements	O
in	O
data	O
I/O	O
range	O
from	O
10	O
to	O
30	O
percent	O
over	O
tiling	O
and	O
from	O
10	O
to	O
40	O
percent	O
over	O
maximal	O
reuse	O
for	O
JPEG	O
loops	O

Outsourced	B-KEY
backup	O
saves	O
time	O
To	O
increase	O
the	O
efficiency	O
of	O
its	O
data	B-KEY
backup	I-KEY
and	O
to	O
free	O
staff	O
to	O
concentrate	O
on	O
core	O
business	O
,	O
The	B-KEY
Gadget	I-KEY
Shop	I-KEY
is	O
relying	O
on	O
a	O
secure	O
,	O
automated	O
system	O
hosted	O
by	O
a	O
third	O
party	O

Sufficient	B-KEY
conditions	I-KEY
on	O
nonemptiness	B-KEY
and	O
boundedness	B-KEY
of	O
the	O
solution	B-KEY
set	I-KEY
of	O
the	O
P/sub	O
0	O
/	O
function	O
nonlinear	O
complementarity	O
problem	O
The	O
P/sub	O
0	O
/	O
function	O
nonlinear	O
complementarity	O
,	O
problem	O
-LRB-	O
NCP	O
-RRB-	O
has	O
attracted	O
a	O
lot	O
of	O
attention	O
among	O
researchers	O
.	O
Various	O
assumed	O
conditions	O
,	O
which	O
ensure	O
that	O
the	O
NCP	O
has	O
a	O
solution	O
have	O
been	O
proposed	O
.	O
In	O
this	O
paper	O
,	O
by	O
using	O
the	O
notion	O
of	O
an	O
exceptional	O
family	O
of	O
elements	O
we	O
develop	O
a	O
sufficient	B-KEY
condition	I-KEY
which	O
ensures	O
that	O
the	O
solution	B-KEY
set	I-KEY
of	O
the	O
P/sub	O
0	O
/	O
function	O
NCP	O
is	O
nonempty	B-KEY
and	O
bounded	O
.	O
In	O
particular	O
,	O
we	O
prove	O
that	O
many	O
existing	O
assumed	O
conditions	O
imply	O
this	O
sufficient	B-KEY
condition	I-KEY
.	O
Thus	O
,	O
these	O
conditions	O
imply	O
that	O
the	O
solution	B-KEY
set	I-KEY
of	O
the	O
P/sub	O
0	O
/	O
function	O
NCP	O
is	O
nonempty	B-KEY
and	O
bounded	O
.	O
In	O
addition	O
,	O
we	O
also	O
prove	O
directly	O
that	O
a	O
few	O
existence	O
conditions	O
imply	O
that	O
the	O
solution	B-KEY
set	I-KEY
of	O
the	O
P/sub	O
0	O
/	O
function	O
NCP	O
is	O
bounded	O

Analogue	B-KEY
realizations	I-KEY
of	O
fractional-order	B-KEY
controllers	I-KEY
An	O
approach	O
to	O
the	O
design	O
of	O
analogue	O
circuits	O
,	O
implementing	O
fractional-order	B-KEY
controllers	I-KEY
,	O
is	O
presented	O
.	O
The	O
suggested	O
approach	O
is	O
based	O
on	O
the	O
use	O
of	O
continued	B-KEY
fraction	I-KEY
expansions	I-KEY
;	O
in	O
the	O
case	O
of	O
negative	O
coefficients	O
in	O
a	O
continued	O
fraction	O
expansion	O
,	O
the	O
use	O
of	O
negative	O
impedance	O
converters	O
is	O
proposed	O
.	O
Several	O
possible	O
methods	O
for	O
obtaining	O
suitable	O
rational	B-KEY
approximations	I-KEY
and	O
continued	B-KEY
fraction	I-KEY
expansions	I-KEY
are	O
discussed	O
.	O
An	O
example	O
of	O
realization	O
of	O
a	O
fractional-order	O
I/sup	O
lambda	O
/	O
controller	O
is	O
presented	O
and	O
illustrated	O
by	O
obtained	O
measurements	O
.	O
The	O
suggested	O
approach	O
can	O
be	O
used	O
for	O
the	O
control	O
of	O
very	O
fast	B-KEY
processes	I-KEY
,	O
where	O
the	O
use	O
of	O
digital	B-KEY
controllers	I-KEY
is	O
difficult	O
or	O
impossible	O

Survey	B-KEY
says	O
!	O
-LSB-	O
online	O
world	O
of	O
polls	B-KEY
and	O
surveys	B-KEY
-RSB-	O
Many	O
content	B-KEY
managers	I-KEY
miss	O
the	O
fundamental	O
interactivity	O
of	O
the	O
Web	O
by	O
not	O
using	O
polls	B-KEY
and	O
surveys	B-KEY
.	O
Using	O
interactive	O
features-like	O
a	O
poll	B-KEY
or	O
quiz-offers	O
your	O
readers	O
an	O
opportunity	O
to	O
become	O
more	O
engaged	O
in	O
your	O
content	O
.	O
Using	O
a	O
survey	B-KEY
to	O
gather	O
feedback	O
about	O
your	O
content	O
provides	O
cost-effective	O
data	O
to	O
help	O
make	O
modifications	O
or	O
plot	O
the	O
appropriate	O
course	O
of	O
action	O
.	O
The	O
Web	O
has	O
allowed	O
us	O
to	O
take	O
traditional	O
market	O
research	O
and	O
turn	O
it	O
on	O
its	O
ear	O
.	O
Surveys	B-KEY
and	O
polls	B-KEY
can	O
be	O
conducted	O
faster	O
and	O
cheaper	O
than	O
with	O
telephone	O
and	O
mail	O
.	O
But	O
if	O
you	O
are	O
running	O
a	O
Web	O
site	O
,	O
should	O
you	O
care	O
about	O
polls	B-KEY
and	O
surveys	B-KEY
?	O
Do	O
you	O
know	O
the	O
difference	O
between	O
the	O
two	O
in	O
Web-speak	O
?	O

VoIP	B-KEY
:	O
leveraging	O
existing	O
cable	B-KEY
architecture	I-KEY
As	O
operators	O
prepare	O
to	O
enter	O
the	O
voice-over-IP	B-KEY
fray	O
,	O
they	O
are	O
searching	O
for	O
ways	O
to	O
leverage	O
their	O
existing	O
two-way	O
,	O
interactive	O
infrastructure	O
.	O
There	O
are	O
several	O
approaches	O
for	O
supporting	O
VoIP	B-KEY
on	O
top	O
of	O
the	O
core	B-KEY
IP	I-KEY
transport	I-KEY
network	I-KEY
.	O
The	O
one	O
garnering	O
the	O
most	O
interest	O
,	O
especially	O
in	O
the	O
United	B-KEY
States	I-KEY
,	O
is	O
based	O
on	O
the	O
PacketCable	B-KEY
1.x	I-KEY
architecture	I-KEY
.	O
This	O
article	O
discusses	O
the	O
PacketCable-based	B-KEY
approach	I-KEY

Anti-spam	O
suit	O
attempts	O
to	O
hold	O
carriers	O
accountable	O
A	O
lawsuit	B-KEY
alleges	O
that	O
Sprint	B-KEY
has	O
violated	O
Utah	O
's	O
new	O
anti-spam	B-KEY
act	I-KEY
.	O
The	O
action	O
could	O
open	O
the	O
door	O
to	O
new	O
regulations	B-KEY
on	O
telecom	B-KEY
service	I-KEY
providers	I-KEY

A	O
distributed	B-KEY
mobile	I-KEY
agent	I-KEY
framework	I-KEY
for	O
maintaining	O
persistent	B-KEY
distance	I-KEY
education	I-KEY
Mobile	O
agent	O
techniques	O
involve	O
distributed	B-KEY
control	I-KEY
if	O
communication	O
is	O
required	O
among	O
different	O
types	O
of	O
agents	O
,	O
especially	O
when	O
mobile	O
agents	O
can	O
migrate	O
from	O
station	O
to	O
station	O
.	O
This	O
technique	O
can	O
be	O
implemented	O
in	O
a	O
distributed	O
distance	O
learning	O
environment	O
,	O
which	O
allows	O
students	O
or	O
instructors	O
to	O
login	O
from	O
anywhere	O
to	O
a	O
central	B-KEY
server	I-KEY
in	O
an	O
education	O
center	O
while	O
still	O
retaining	O
the	O
look-and-feel	O
of	O
personal	O
setups	O
.	O
In	O
this	O
research	O
paper	O
,	O
we	O
propose	O
a	O
distributed	B-KEY
agent	I-KEY
framework	I-KEY
along	O
with	O
its	O
communication	O
messages	O
to	O
facilitate	O
mobile	O
personal	O
agents	O
,	O
which	O
serve	O
three	O
different	O
groups	O
of	O
distance	O
education	O
users	O
:	O
instructors	O
,	O
students	O
,	O
and	O
system	O
administrators	O
.	O
We	O
propose	O
an	O
agent	O
communication	O
framework	O
as	O
well	O
as	O
agent	O
evolution	O
states	O
of	O
mobile	O
agents	O
.	O
The	O
communication	O
architecture	O
and	O
message	B-KEY
transmission	I-KEY
protocols	I-KEY
are	O
illustrated	O
.	O
The	O
system	O
is	O
implemented	O
on	O
the	O
Windows	O
platform	O
to	O
support	O
nomadic	O
accessibility	O
of	O
remote	O
distance	O
learning	O
users	O
.	O
Personal	O
data	O
also	O
migrate	O
with	O
the	O
mobile	O
agents	O
,	O
allowing	O
users	O
to	O
maintain	O
accessibility	O
to	O
some	O
extent	O
even	O
when	O
the	O
Internet	O
connection	O
is	O
temperately	O
disconnected	O
.	O
Using	O
user-friendly	B-KEY
personal	I-KEY
agents	I-KEY
,	O
a	O
distance	O
education	O
platform	O
can	O
include	O
different	O
tools	O
to	O
meet	O
different	O
needs	O
for	O
users	O

Ten	O
suggestions	O
for	O
a	O
gender-equitable	O
CS	O
classroom	O
Though	O
considerable	O
attention	O
has	O
been	O
paid	O
to	O
the	O
creation	O
of	O
a	O
nurturing	B-KEY
environment	I-KEY
for	O
women	O
in	O
the	O
field	O
of	O
computer	B-KEY
science	I-KEY
,	O
proposed	O
solutions	O
have	O
primarily	O
focused	O
on	O
activities	O
outside	O
of	O
the	O
classroom	O
.	O
This	O
paper	O
presents	O
a	O
list	O
of	O
suggestions	O
for	O
modifications	O
to	O
both	O
the	O
pedagogy	B-KEY
and	O
content	O
of	O
CS	O
courses	O
designed	O
to	O
make	O
the	O
CS	B-KEY
classroom	I-KEY
environment	I-KEY
more	O
inviting	O
for	O
women	B-KEY
students	I-KEY

A	O
model	B-KEY
of	O
periodic	B-KEY
oscillation	I-KEY
for	O
genetic	B-KEY
regulatory	I-KEY
systems	I-KEY
In	O
this	O
paper	O
,	O
we	O
focus	O
on	O
modeling	B-KEY
and	O
explaining	O
periodic	B-KEY
oscillations	I-KEY
in	O
gene-protein	B-KEY
systems	I-KEY
with	O
a	O
simple	O
nonlinear	B-KEY
model	I-KEY
and	O
on	O
analyzing	O
effects	O
of	O
time	O
delay	O
on	O
the	O
stability	O
of	O
oscillations	O
.	O
Our	O
main	O
model	B-KEY
of	O
genetic	B-KEY
regulation	I-KEY
comprises	O
of	O
a	O
two-gene	B-KEY
system	I-KEY
with	O
an	O
autoregulatory	B-KEY
feedback	I-KEY
loop	I-KEY
.	O
We	O
exploit	O
multiple	O
time	O
scales	O
and	O
hysteretic	B-KEY
properties	I-KEY
of	O
the	O
model	B-KEY
to	O
construct	O
periodic	B-KEY
oscillations	I-KEY
with	O
jumping	B-KEY
dynamics	I-KEY
and	O
analyze	O
the	O
possible	O
mechanism	O
according	O
to	O
the	O
singular	B-KEY
perturbation	I-KEY
theory	I-KEY
.	O
As	O
shown	O
in	O
this	O
paper	O
,	O
periodic	B-KEY
oscillations	I-KEY
are	O
mainly	O
generated	O
by	O
nonlinearly	O
negative	O
and	O
positive	O
feedback	O
loops	O
in	O
gene	O
regulatory	O
systems	O
,	O
whereas	O
the	O
jumping	B-KEY
dynamics	I-KEY
is	O
generally	O
caused	O
by	O
time	O
scale	O
differences	O
among	O
biochemical	B-KEY
reactions	I-KEY
.	O
This	O
simple	O
model	B-KEY
may	O
actually	O
act	O
as	O
a	O
genetic	O
oscillator	O
or	O
switch	O
in	O
gene-protein	O
networks	O
because	O
the	O
dynamics	O
are	O
robust	O
for	O
parameter	O
perturbations	O
or	O
environment	O
variations	O
.	O
We	O
also	O
explore	O
effects	O
of	O
time	B-KEY
delay	I-KEY
on	O
the	O
stability	O
of	O
the	O
dynamics	O
,	O
showing	O
that	O
the	O
time	B-KEY
delay	I-KEY
generally	O
increases	O
the	O
stability	B-KEY
region	I-KEY
of	O
the	O
oscillations	O
,	O
thereby	O
making	O
the	O
oscillations	O
robust	O
to	O
parameter	O
changes	O
.	O
Two	O
examples	O
are	O
also	O
provided	O
to	O
numerically	O
demonstrate	O
our	O
theoretical	O
results	O

Evaluation	O
of	O
existing	O
and	O
new	O
feature	B-KEY
recognition	I-KEY
algorithms	I-KEY
.	O
1	O
.	O
Theory	O
and	O
implementation	O
This	O
is	O
the	O
first	O
of	O
two	O
papers	O
evaluating	O
the	O
performance	O
of	O
general-purpose	B-KEY
feature	I-KEY
detection	I-KEY
techniques	I-KEY
for	O
geometric	B-KEY
models	I-KEY
.	O
In	O
this	O
paper	O
,	O
six	O
different	O
methods	O
are	O
described	O
to	O
identify	O
sets	B-KEY
of	I-KEY
faces	I-KEY
that	O
bound	O
depression	O
and	O
protrusion	B-KEY
faces	I-KEY
.	O
Each	O
algorithm	O
has	O
been	O
implemented	O
and	O
tested	O
on	O
eight	O
components	O
from	O
the	O
National	B-KEY
Design	I-KEY
Repository	I-KEY
.	O
The	O
algorithms	O
studied	O
include	O
previously	O
published	O
general-purpose	O
feature	O
detection	O
algorithms	O
such	O
as	O
the	O
single-face	O
inner-loop	O
and	O
concavity	B-KEY
techniques	I-KEY
.	O
Others	O
are	O
improvements	O
to	O
existing	O
algorithms	O
such	O
as	O
extensions	O
of	O
the	O
two-dimensional	B-KEY
convex	I-KEY
hull	I-KEY
method	I-KEY
to	O
handle	O
curved	B-KEY
faces	I-KEY
as	O
well	O
as	O
protrusions	O
.	O
Lastly	O
,	O
new	O
algorithms	O
based	O
on	O
the	O
three-dimensional	B-KEY
convex	I-KEY
hull	I-KEY
,	O
minimum	B-KEY
concave	I-KEY
,	O
visible	O
and	O
multiple-face	B-KEY
inner-loop	I-KEY
face	I-KEY
sets	I-KEY
are	O
described	O

Development	O
of	O
a	O
computer-aided	O
manufacturing	O
system	O
for	O
profiled	B-KEY
edge	I-KEY
lamination	I-KEY
tooling	I-KEY
Profiled	O
edge	O
lamination	O
-LRB-	O
PEL	O
-RRB-	O
tooling	O
is	O
a	O
promising	O
rapid	B-KEY
tooling	I-KEY
-LRB-	O
RT	O
-RRB-	O
method	O
involving	O
the	O
assembly	O
of	O
an	O
array	O
of	O
laminations	O
whose	O
top	O
edges	O
are	O
simultaneously	O
profiled	O
and	O
beveled	O
based	O
on	O
a	O
CAD	O
model	O
of	O
the	O
intended	O
tool	O
surface	O
.	O
To	O
facilitate	O
adoption	O
of	O
this	O
RT	O
method	O
by	O
industry	O
,	O
a	O
comprehensive	O
PEL	O
tooling	O
development	O
system	O
is	O
proposed	O
.	O
The	O
two	O
main	O
parts	O
of	O
this	O
system	O
are	O
:	O
-LRB-	O
1	O
-RRB-	O
iterative	O
tool	O
design	O
based	O
on	O
thermal	O
and	O
structural	O
models	O
;	O
and	O
-LRB-	O
2	O
-RRB-	O
fabrication	O
of	O
the	O
tool	O
using	O
a	O
computer-aided	O
manufacturing	O
-LRB-	O
CAM	O
-RRB-	O
software	O
and	O
abrasive	B-KEY
water	I-KEY
jet	I-KEY
cutting	I-KEY
.	O
CAM	B-KEY
software	I-KEY
has	O
been	O
developed	O
to	O
take	O
lamination	O
slice	O
data	O
-LRB-	O
profiles	O
-RRB-	O
from	O
any	O
proprietary	O
RP	O
software	O
in	O
the	O
form	O
of	O
polylines	O
and	O
create	O
smooth	O
,	O
kinematically	O
desirable	O
cutting	O
trajectories	O
for	O
each	O
tool	O
lamination	O
.	O
Two	O
cutting	B-KEY
trajectory	I-KEY
algorithms	I-KEY
,	O
called	O
identical	B-KEY
equidistant	I-KEY
profile	I-KEY
segmentation	I-KEY
and	O
adaptively	B-KEY
vector	I-KEY
profiles	I-KEY
projection	I-KEY
-LRB-	O
AVPP	O
-RRB-	O
,	O
were	O
created	O
for	O
this	O
purpose	O
.	O
By	O
comparing	O
the	O
performance	O
of	O
both	O
algorithms	O
with	O
a	O
benchmark	O
part	O
shape	O
,	O
the	O
AVPP	O
algorithm	O
provided	O
better	O
cutting	O
trajectories	O
for	O
complicated	O
tool	O
geometries	O
.	O
A	O
15-layer	O
aluminum	O
PEL	O
tool	O
was	O
successfully	O
fabricated	O
using	O
a	O
5-axis	O
CNC	O
AWJ	O
cutter	O
and	O
NC	O
code	O
generated	O
by	O
the	O
CAM	B-KEY
software	I-KEY

Subject	O
access	O
to	O
government	B-KEY
documents	I-KEY
in	O
an	O
era	O
of	O
globalization	B-KEY
:	O
intellectual	B-KEY
bundling	I-KEY
of	O
entities	O
affected	O
by	O
the	O
decisions	O
of	O
supranational	B-KEY
organizations	I-KEY
As	O
a	O
result	O
of	O
the	O
growing	O
influence	O
of	O
supranational	B-KEY
organizations	I-KEY
,	O
there	O
is	O
a	O
need	O
for	O
a	O
new	O
model	O
for	O
subject	O
access	O
to	O
government	O
information	O
in	O
academic	B-KEY
libraries	I-KEY
.	O
Rulings	O
made	O
by	O
supranational	O
bodies	O
such	O
as	O
the	O
World	B-KEY
Trade	I-KEY
Organization	I-KEY
-LRB-	O
WTO	O
-RRB-	O
and	O
rulings	O
determined	O
under	O
the	O
auspices	O
of	O
transnational	B-KEY
economic	I-KEY
agreements	I-KEY
such	O
as	O
the	O
North	B-KEY
American	I-KEY
Free	I-KEY
Trade	I-KEY
Agreement	I-KEY
-LRB-	O
NAFTA	O
-RRB-	O
often	O
supersede	O
existing	O
law	O
,	O
resulting	O
in	O
obligatory	O
changes	O
to	O
national	O
,	O
provincial	O
,	O
state	O
,	O
and	O
municipal	B-KEY
legislation	I-KEY
.	O
Just	O
as	O
important	O
is	O
the	O
relationship	O
among	O
private	O
sector	O
companies	O
,	O
third	O
party	O
actors	O
such	O
as	O
nongovernmental	O
organizations	O
-LRB-	O
NGOs	O
-RRB-	O
,	O
and	O
governments	O
.	O
The	O
interaction	O
among	O
the	O
various	O
entities	O
affected	O
by	O
supranational	O
rulings	O
could	O
potentially	O
form	O
the	O
basis	O
of	O
a	O
new	O
model	O
for	O
subject	O
access	O
to	O
government	O
information	O

SIA	O
shelves	O
T	O
+1	O
decision	O
till	O
2004	O
The	O
Securities	B-KEY
Industry	I-KEY
Association	I-KEY
has	O
decided	O
that	O
a	O
move	O
to	O
T	O
+1	O
is	O
more	O
than	O
the	O
industry	O
can	O
handle	O
right	O
now	O
.	O
STP	O
,	O
however	O
,	O
will	O
remain	O
a	O
focus	O

Modeling	O
group	B-KEY
foraging	I-KEY
:	O
individual	B-KEY
suboptimality	I-KEY
,	O
interference	O
,	O
and	O
a	O
kind	O
of	O
matching	O
A	O
series	O
of	O
agent-based	B-KEY
models	I-KEY
support	O
the	O
hypothesis	O
that	O
behaviors	O
adapted	O
to	O
a	O
group	B-KEY
situation	I-KEY
may	O
be	O
suboptimal	O
-LRB-	O
or	O
``	O
irrational	O
''	O
-RRB-	O
when	O
expressed	O
by	O
an	O
isolated	B-KEY
individual	I-KEY
.	O
These	O
models	O
focus	O
on	O
two	O
areas	O
of	O
current	O
concern	O
in	O
behavioral	B-KEY
ecology	I-KEY
and	O
experimental	B-KEY
psychology	I-KEY
:	O
the	O
``	O
interference	B-KEY
function	I-KEY
''	O
-LRB-	O
which	O
relates	O
the	O
intake	O
rate	O
of	O
a	O
focal	B-KEY
forager	I-KEY
to	O
the	O
density	O
of	O
conspecifics	O
-RRB-	O
and	O
the	O
``	O
matching	B-KEY
law	I-KEY
''	O
-LRB-	O
which	O
formalizes	O
the	O
observation	O
that	O
many	O
animals	O
match	O
the	O
frequency	O
of	O
their	O
response	O
to	O
different	O
stimuli	O
in	O
proportion	O
to	O
the	O
reward	O
obtained	O
from	O
each	O
stimulus	O
type	O
-RRB-	O
.	O
Each	O
model	O
employs	O
genetic	B-KEY
algorithms	I-KEY
to	O
evolve	O
foraging	O
behaviors	O
for	O
multiple	B-KEY
agents	I-KEY
in	O
spatially	B-KEY
explicit	I-KEY
environments	I-KEY
,	O
structured	O
at	O
the	O
level	O
of	O
situated	B-KEY
perception	I-KEY
and	O
action	O
.	O
A	O
second	O
concern	O
of	O
the	O
article	O
is	O
to	O
extend	O
the	O
understanding	O
of	O
both	O
matching	O
and	O
interference	O
per	O
se	O
by	O
modeling	O
at	O
this	O
level	O

Web	O
talk	O
is	O
cheap	O
Web	O
technology	O
provides	O
a	O
wealth	O
of	O
opportunities	O
for	O
reaching	O
potential	O
customers	O
.	O
So	O
how	O
do	O
you	O
make	O
it	O
work	O
for	O
your	O
business	O
?	O

Developing	O
a	O
high-performance	B-KEY
web	I-KEY
server	I-KEY
in	O
Concurrent	B-KEY
Haskell	I-KEY
Server	O
applications	O
,	O
and	O
in	O
particular	O
network-based	B-KEY
server	I-KEY
applications	I-KEY
,	O
place	O
a	O
unique	O
combination	O
of	O
demands	O
on	O
a	O
programming	O
language	O
:	O
lightweight	B-KEY
concurrency	I-KEY
,	O
high	B-KEY
I/O	I-KEY
throughput	I-KEY
,	O
and	O
fault	B-KEY
tolerance	I-KEY
are	O
all	O
important	O
.	O
This	O
paper	O
describes	O
a	O
prototype	O
Web	O
server	O
written	O
in	O
Concurrent	B-KEY
Haskell	I-KEY
-LRB-	O
with	O
extensions	O
-RRB-	O
,	O
and	O
presents	O
two	O
useful	O
results	O
:	O
firstly	O
,	O
a	O
conforming	B-KEY
server	I-KEY
could	O
be	O
written	O
with	O
minimal	O
effort	O
,	O
leading	O
to	O
an	O
implementation	O
in	O
less	O
than	O
1500	O
lines	O
of	O
code	O
,	O
and	O
secondly	O
the	O
naive	O
implementation	O
produced	O
reasonable	O
performance	O
.	O
Furthermore	O
,	O
making	O
minor	O
modifications	O
to	O
a	O
few	O
time-critical	B-KEY
components	I-KEY
improved	O
performance	O
to	O
a	O
level	O
acceptable	O
for	O
anything	O
but	O
the	O
most	O
heavily	O
loaded	O
Web	O
servers	O

Characterization	O
of	O
sheet	B-KEY
buckling	I-KEY
subjected	O
to	O
controlled	O
boundary	B-KEY
constraints	I-KEY
A	O
wedge	B-KEY
strip	I-KEY
test	I-KEY
is	O
designed	O
to	O
study	O
the	O
onset	O
and	O
post-buckling	O
behavior	O
of	O
a	O
sheet	O
under	O
various	O
boundary	B-KEY
constraints	I-KEY
.	O
The	O
device	O
can	O
be	O
easily	O
incorporated	O
into	O
a	O
conventional	O
tensile	B-KEY
test	I-KEY
machine	I-KEY
,	O
and	O
material	O
resistance	O
to	O
buckling	O
is	O
measured	O
as	O
the	O
buckling	O
height	O
versus	O
the	O
in-plane	O
strain	B-KEY
state	I-KEY
.	O
The	O
design	O
yields	O
different	O
but	O
consistent	O
buckling	O
modes	O
with	O
easy	O
changes	O
of	O
boundary	O
conditions	O
-LRB-	O
either	O
clamped	O
or	O
freed	O
-RRB-	O
and	O
sample	O
geometry	O
.	O
Experimental	O
results	O
are	O
then	O
used	O
to	O
verify	O
a	O
hybrid	O
approach	O
to	O
buckling	O
prediction	O
,	O
i.e.	O
,	O
the	O
combination	O
of	O
the	O
FEM	O
analysis	O
and	O
an	O
energy-based	B-KEY
analytical	I-KEY
wrinkling	I-KEY
criterion	I-KEY
.	O
The	O
FEM	O
analysis	O
is	O
used	O
to	O
obtain	O
the	O
stress	B-KEY
field	I-KEY
and	O
deformed	B-KEY
geometry	I-KEY
in	O
a	O
complex	O
forming	O
condition	O
,	O
while	O
the	O
analytical	O
solution	O
is	O
to	O
provide	O
the	O
predictions	O
less	O
sensitive	O
to	O
artificial	O
numerical	O
parameters	O
.	O
A	O
good	O
agreement	O
between	O
experimental	O
data	O
and	O
numerical	O
predictions	O
is	O
obtained	O

An	O
attack-finding	B-KEY
algorithm	I-KEY
for	O
security	B-KEY
protocols	I-KEY
This	O
paper	O
proposes	O
an	O
automatic	O
attack	O
construction	O
algorithm	O
in	O
order	O
to	O
find	O
potential	O
attacks	O
on	O
security	B-KEY
protocols	I-KEY
.	O
It	O
is	O
based	O
on	O
a	O
dynamic	B-KEY
strand	I-KEY
space	I-KEY
model	I-KEY
,	O
which	O
enhances	O
the	O
original	O
strand	O
space	O
model	O
by	O
introducing	O
active	O
nodes	O
on	O
strands	O
so	O
as	O
to	O
characterize	O
the	O
dynamic	O
procedure	O
of	O
protocol	O
execution	O
.	O
With	O
exact	O
causal	O
dependency	O
relations	O
between	O
messages	O
considered	O
in	O
the	O
model	O
,	O
this	O
algorithm	O
can	O
avoid	O
state	B-KEY
space	I-KEY
explosion	I-KEY
caused	O
by	O
asynchronous	B-KEY
composition	I-KEY
.	O
In	O
order	O
to	O
get	O
a	O
finite	O
state	O
space	O
,	O
a	O
new	O
method	O
called	O
strand-added	B-KEY
on	I-KEY
demand	I-KEY
is	O
exploited	O
,	O
which	O
extends	O
a	O
bundle	O
in	O
an	O
incremental	O
manner	O
without	O
requiring	O
explicit	O
configuration	O
of	O
protocol	O
execution	O
parameters	O
.	O
A	O
finer	O
granularity	O
model	O
of	O
term	O
structure	O
is	O
also	O
introduced	O
,	O
in	O
which	O
subterms	O
are	O
divided	O
into	O
check	B-KEY
subterms	I-KEY
and	O
data	B-KEY
subterms	I-KEY
.	O
Moreover	O
,	O
data	B-KEY
subterms	I-KEY
can	O
be	O
further	O
classified	O
based	O
on	O
the	O
compatible	O
data	B-KEY
subterm	I-KEY
relation	O
to	O
obtain	O
automatically	O
the	O
finite	O
set	O
of	O
valid	O
acceptable	O
terms	O
for	O
an	O
honest	O
principal	O
.	O
In	O
this	O
algorithm	O
,	O
terms	O
core	O
is	O
designed	O
to	O
represent	O
the	O
intruder	O
's	O
knowledge	O
compactly	O
,	O
and	O
forward	O
search	O
technology	O
is	O
used	O
to	O
simulate	O
attack	O
patterns	O
easily	O
.	O
Using	O
this	O
algorithm	O
,	O
a	O
new	O
attack	O
on	O
the	O
Dolve-Yao	B-KEY
protocol	I-KEY
can	O
be	O
found	O
,	O
which	O
is	O
even	O
more	O
harmful	O
because	O
the	O
secret	O
is	O
revealed	O
before	O
the	O
session	O
terminates	O

A	O
formal	B-KEY
framework	I-KEY
for	O
viewpoint	B-KEY
consistency	I-KEY
Multiple	B-KEY
viewpoint	I-KEY
models	I-KEY
of	O
system	B-KEY
development	I-KEY
are	O
becoming	O
increasingly	O
important	O
.	O
Each	O
viewpoint	O
offers	O
a	O
different	O
perspective	O
on	O
the	O
target	O
system	O
and	O
system	B-KEY
development	I-KEY
involves	O
parallel	O
refinement	O
of	O
the	O
multiple	O
views	O
.	O
Viewpoint	O
related	O
approaches	O
have	O
been	O
considered	O
in	O
a	O
number	O
of	O
different	O
guises	O
by	O
a	O
spectrum	O
of	O
researchers	O
.	O
Our	O
work	O
particularly	O
focuses	O
on	O
the	O
use	O
of	O
viewpoints	O
in	O
open	B-KEY
distributed	I-KEY
processing	I-KEY
-LRB-	O
ODP	B-KEY
-RRB-	O
which	O
is	O
an	O
ISO/ITU	B-KEY
standardisation	I-KEY
framework	I-KEY
.	O
The	O
requirements	O
of	O
viewpoint	O
modelling	O
in	O
ODP	B-KEY
are	O
very	O
broad	O
and	O
,	O
hence	O
,	O
demanding	O
.	O
Multiple	O
viewpoints	O
,	O
though	O
,	O
prompt	O
the	O
issue	O
of	O
consistency	O
between	O
viewpoints	O
.	O
This	O
paper	O
describes	O
a	O
very	O
general	O
interpretation	O
of	O
consistency	O
which	O
we	O
argue	O
is	O
broad	O
enough	O
to	O
meet	O
the	O
requirements	O
of	O
consistency	O
in	O
ODP	B-KEY
.	O
We	O
present	O
a	O
formal	B-KEY
framework	I-KEY
for	O
this	O
general	O
interpretation	O
;	O
highlight	O
basic	O
properties	O
of	O
the	O
interpretation	O
and	O
locate	O
restricted	O
classes	O
of	O
consistency	O
.	O
Strategies	O
for	O
checking	O
consistency	O
are	O
also	O
investigated	O
.	O
Throughout	O
we	O
illustrate	O
our	O
theory	O
using	O
the	O
formal	B-KEY
description	I-KEY
technique	I-KEY
LOTOS	B-KEY
.	O
Thus	O
,	O
the	O
paper	O
also	O
characterises	O
the	O
nature	O
of	O
and	O
options	O
for	O
consistency	B-KEY
checking	I-KEY
in	O
LOTOS	B-KEY

Digital	O
rights	O
-LRB-	O
and	O
wrongs	O
-RRB-	O
Attempting	O
to	O
grasp	O
the	O
many	O
conflicts	O
and	O
proposed	O
safeguards	O
for	O
intellectual	B-KEY
property	I-KEY
is	O
extremely	O
difficult	O
.	O
Legal	O
,	O
political	O
,	O
economic	O
,	O
and	O
cultural	O
issues-both	O
domestic	O
and	O
international-loom	O
large	O
,	O
almost	O
dwarfing	O
the	O
daunting	O
technological	O
challenges	O
.	O
Solutions	O
devised	O
by	O
courts	O
and	O
legislatures	O
and	O
regulatory	O
agencies	O
are	O
always	O
late	O
out	O
of	O
the	O
blocks	O
and	O
fall	O
ever	O
farther	O
behind	O
.	O
Recently	O
proposed	O
legislation	O
only	O
illustrates	O
the	O
depth	O
and	O
complexity	O
of	O
the	O
problem	O

Minimizing	O
the	O
number	O
of	O
successor	O
states	O
in	O
the	O
stubborn	B-KEY
set	I-KEY
method	I-KEY
Combinatorial	B-KEY
explosion	I-KEY
which	O
occurs	O
in	O
parallel	O
compositions	O
of	O
LTSs	O
can	O
be	O
alleviated	O
by	O
letting	O
the	O
stubborn	B-KEY
set	I-KEY
method	I-KEY
construct	O
on-the-fly	O
a	O
reduced	O
LTS	O
that	O
is	O
CFFD	O
-	O
or	O
CSP-equivalent	B-KEY
to	O
the	O
actual	O
parallel	O
composition	O
.	O
This	O
article	O
considers	O
the	O
problem	O
of	O
minimizing	O
the	O
number	O
of	O
successor	O
states	O
of	O
a	O
given	O
state	O
in	O
the	O
reduced	O
LTS	O
.	O
The	O
problem	O
can	O
be	O
solved	O
by	O
constructing	O
an	O
and/or-graph	O
with	O
weighted	B-KEY
vertices	I-KEY
and	O
by	O
finding	O
a	O
set	O
of	O
vertices	O
that	O
satisfies	O
a	O
certain	O
constraint	O
such	O
that	O
no	O
set	O
of	O
vertices	O
satisfying	O
the	O
constraint	O
has	O
a	O
smaller	O
sum	O
of	O
weights	O
.	O
Without	O
weights	O
,	O
the	O
and/or-graph	O
can	O
be	O
constructed	O
in	O
low-degree	B-KEY
polynomial	I-KEY
time	I-KEY
w.r.t.	O
the	O
length	O
of	O
the	O
input	O
of	O
the	O
problem	O
.	O
However	O
,	O
since	O
actions	O
can	O
be	O
nondeterministic	O
and	O
transitions	O
can	O
share	O
target	O
states	O
,	O
it	O
is	O
not	O
known	O
whether	O
the	O
weights	O
are	O
generally	O
computable	O
in	O
polynomial	O
time	O
.	O
Consequently	O
,	O
it	O
is	O
an	O
open	O
problem	O
whether	O
minimizing	O
the	O
number	O
of	O
successor	O
states	O
is	O
as	O
``	O
easy	O
''	O
as	O
minimizing	O
the	O
number	O
of	O
successor	O
transitions	O

Numerical	B-KEY
modeling	I-KEY
of	O
the	O
flow	O
in	O
stenosed	O
coronary	O
artery	O
.	O
The	O
relationship	O
between	O
main	O
hemodynamic	B-KEY
parameters	I-KEY
The	O
severity	O
of	O
coronary	B-KEY
arterial	I-KEY
stenosis	I-KEY
is	O
usually	O
measured	O
by	O
either	O
simple	O
geometrical	O
parameters	O
,	O
such	O
as	O
percent	O
diameter	O
stenosis	O
,	O
or	O
hemodynamically	O
based	O
parameters	O
,	O
such	O
as	O
the	O
fractional	O
flow	O
reserve	O
-LRB-	O
FFR	O
-RRB-	O
or	O
coronary	O
flow	O
reserve	O
-LRB-	O
CFR	O
-RRB-	O
.	O
The	O
present	O
study	O
aimed	O
to	O
establish	O
a	O
relationship	O
between	O
actual	O
hemodynamic	O
conditions	O
and	O
the	O
parameters	O
that	O
define	O
stenosis	B-KEY
severity	I-KEY
in	O
the	O
clinical	B-KEY
setting	I-KEY
.	O
We	O
used	O
a	O
computational	B-KEY
model	I-KEY
of	O
the	O
blood	B-KEY
flow	I-KEY
in	O
a	O
vessel	O
with	O
a	O
blunt	B-KEY
stenosis	I-KEY
and	O
an	O
autoregulated	B-KEY
vascular	I-KEY
bed	I-KEY
to	O
simulate	B-KEY
a	O
stenosed	B-KEY
blood	I-KEY
vessel	I-KEY
.	O
A	O
key	O
point	O
in	O
creating	O
realistic	O
simulations	B-KEY
is	O
to	O
properly	O
model	O
arterial	B-KEY
autoregulation	I-KEY
.	O
A	O
constant	B-KEY
flow	I-KEY
regulation	I-KEY
mechanism	I-KEY
resulted	O
in	O
CFR	O
and	O
FFR	O
values	O
that	O
were	O
within	O
the	O
physiological	B-KEY
range	I-KEY
,	O
while	O
a	O
constant	O
wall-shear	O
stress	O
model	O
yielded	O
unrealistic	O
values	O
.	O
The	O
simulation	B-KEY
tools	O
developed	O
in	O
the	O
present	O
study	O
may	O
be	O
useful	O
in	O
the	O
clinical	O
assessment	O
of	O
single	O
and	O
multiple	O
stenoses	O
by	O
means	O
of	O
minimally	B-KEY
invasive	I-KEY
methods	I-KEY

Self-describing	B-KEY
Turing	O
machines	O
After	O
a	O
sketchy	O
historical	O
account	O
on	O
the	O
question	O
of	O
self-describeness	B-KEY
and	O
self-reproduction	B-KEY
,	O
and	O
after	O
discussing	O
the	O
definition	O
of	O
suitable	O
encodings	B-KEY
for	O
self-describeness	B-KEY
,	O
we	O
give	O
the	O
construction	O
of	O
several	O
self-describing	B-KEY
Turing	O
machines	O
,	O
namely	O
self-describing	O
machines	O
with	O
,	O
respectively	O
,	O
350	O
,	O
267	O
,	O
224	O
and	O
206	O
instructions	O

The	O
average-case	B-KEY
identifiability	I-KEY
and	O
controllability	O
of	O
large	B-KEY
scale	I-KEY
systems	I-KEY
Needs	O
for	O
increased	O
product	O
quality	O
,	O
reduced	O
pollution	O
,	O
and	O
reduced	O
energy	O
and	O
material	O
consumption	O
are	O
driving	O
enhanced	B-KEY
process	I-KEY
integration	I-KEY
.	O
This	O
increases	O
the	O
number	O
of	O
manipulated	O
and	O
measured	B-KEY
variables	I-KEY
required	O
by	O
the	O
control	O
system	O
to	O
achieve	O
its	O
objectives	O
.	O
This	O
paper	O
addresses	O
the	O
question	O
of	O
whether	O
processes	O
tend	O
to	O
become	O
increasingly	O
more	O
difficult	O
to	O
identify	O
and	O
control	O
as	O
the	O
process	O
dimension	O
increases	O
.	O
Tools	O
and	O
results	O
of	O
multivariable	B-KEY
statistics	I-KEY
are	O
used	O
to	O
show	O
that	O
,	O
under	O
a	O
variety	O
of	O
assumed	O
distributions	O
on	O
the	O
elements	O
,	O
square	O
processes	O
of	O
higher	O
dimension	O
tend	O
to	O
be	O
more	O
difficult	O
to	O
identify	O
and	O
control	O
,	O
whereas	O
the	O
expected	O
controllability	O
and	O
identifiability	O
of	O
nonsquare	B-KEY
processes	I-KEY
depends	O
on	O
the	O
relative	O
numbers	O
of	O
measured	O
and	O
manipulated	B-KEY
variables	I-KEY
.	O
These	O
results	O
suggest	O
that	O
the	O
procedure	O
of	O
simplifying	O
the	O
control	O
problem	O
so	O
that	O
only	O
a	O
square	O
process	O
is	O
considered	O
is	O
a	O
poor	O
practice	O
for	O
large	B-KEY
scale	I-KEY
systems	I-KEY

A	O
wizard	O
idea	O
-LSB-	O
Internet	B-KEY
in	O
finance	B-KEY
-RSB-	O
New	O
technology	O
is	O
set	O
to	O
become	O
an	O
ever-more	O
important	O
area	O
of	O
work	O
for	O
brokers	B-KEY
.	O
Lawrie	O
Holmes	O
looks	O
at	O
how	O
the	O
Internet	B-KEY
is	O
driving	O
change	O
and	O
opportunity	O

Real-time	B-KEY
estimations	I-KEY
of	O
multi-modal	B-KEY
frequencies	I-KEY
for	O
smart	B-KEY
structures	I-KEY
In	O
this	O
paper	O
,	O
various	O
methods	O
for	O
the	O
real-time	B-KEY
estimation	I-KEY
of	O
multi-modal	B-KEY
frequencies	I-KEY
are	O
realized	O
in	O
real	O
time	O
and	O
compared	O
through	O
numerical	O
and	O
experimental	O
tests	O
.	O
These	O
parameter-based	O
frequency	B-KEY
estimation	I-KEY
methods	O
can	O
be	O
applied	O
to	O
various	O
engineering	O
fields	O
such	O
as	O
communications	O
,	O
radar	O
and	O
adaptive	O
vibration	O
and	O
noise	B-KEY
control	I-KEY
.	O
Well-known	O
frequency	B-KEY
estimation	I-KEY
methods	O
are	O
introduced	O
and	O
explained	O
.	O
The	O
Bairstow	B-KEY
method	I-KEY
is	O
introduced	O
to	O
find	O
the	O
roots	O
of	O
a	O
characteristic	B-KEY
equation	I-KEY
for	O
estimations	O
of	O
multi-modal	B-KEY
frequencies	I-KEY
,	O
and	O
the	O
computational	O
efficiency	O
of	O
the	O
Bairstow	B-KEY
method	I-KEY
is	O
shown	O
quantitatively	O
.	O
For	O
a	O
simple	O
numerical	O
test	O
,	O
we	O
consider	O
two	O
sinusoids	O
of	O
the	O
same	O
amplitudes	O
mixed	O
with	O
various	O
amounts	O
of	O
white	O
noise	O
.	O
The	O
test	O
results	O
show	O
that	O
the	O
auto	O
regressive	O
-LRB-	O
AR	O
-RRB-	O
and	O
auto	O
regressive	O
and	O
moving	O
average	O
-LRB-	O
ARMA	B-KEY
-RRB-	O
methods	O
are	O
unsuitable	O
in	O
noisy	O
environments	O
.	O
The	O
other	O
methods	O
apart	O
from	O
the	O
AR	O
method	O
have	O
fast	O
tracking	O
capability	O
.	O
From	O
the	O
point	O
of	O
view	O
of	O
computational	O
efficiency	O
,	O
the	O
results	O
reveal	O
that	O
the	O
ARMA	B-KEY
method	O
is	O
inefficient	O
,	O
while	O
the	O
cascade	B-KEY
notch	I-KEY
filter	I-KEY
method	O
is	O
very	O
effective	O
.	O
The	O
linearized	B-KEY
adaptive	I-KEY
notch	I-KEY
filter	I-KEY
and	O
recursive	B-KEY
maximum	I-KEY
likelihood	I-KEY
methods	I-KEY
have	O
average	O
performances	O
.	O
Experimental	O
tests	O
are	O
devised	O
to	O
confirm	O
the	O
feasibility	O
of	O
real-time	B-KEY
computations	I-KEY
and	O
to	O
impose	O
the	O
severe	O
conditions	O
of	O
drastically	O
different	O
amplitudes	O
and	O
of	O
considerable	O
changes	O
of	O
natural	O
frequencies	O
.	O
We	O
have	O
performed	O
experiments	O
to	O
extract	O
the	O
natural	O
frequencies	O
from	O
the	O
vibration	B-KEY
signal	I-KEY
of	O
wing-like	B-KEY
composite	I-KEY
plates	I-KEY
in	O
real	O
time	O
.	O
The	O
natural	O
frequencies	O
of	O
the	O
specimen	O
are	O
changed	O
by	O
added	O
masses	O
.	O
Especially	O
,	O
the	O
AR	O
method	O
exhibits	O
a	O
remarkable	O
performance	O
in	O
spite	O
of	O
the	O
severe	O
conditions	O
.	O
This	O
study	O
will	O
be	O
helpful	O
to	O
anyone	O
who	O
needs	O
a	O
frequency	B-KEY
estimation	I-KEY
algorithm	O
for	O
real-time	O
applications	O

Mathematical	B-KEY
model	I-KEY
of	O
functioning	O
of	O
an	O
insurance	O
company	O
with	O
allowance	O
for	O
advertising	O
expenses	O
A	O
mathematical	B-KEY
model	I-KEY
of	O
the	O
functioning	O
of	O
an	O
insurance	O
company	O
with	O
allowance	O
for	O
advertising	O
expenses	O
is	O
suggested	O
.	O
The	O
basic	O
characteristics	O
of	O
the	O
capital	B-KEY
of	O
the	O
company	O
and	O
the	O
advertising	O
efficiency	O
are	O
examined	O
in	O
the	O
case	O
in	O
which	O
the	O
advertising	O
expenses	O
are	O
proportional	O
to	O
the	O
capital	B-KEY

Perspectives	O
on	O
academic	O
vs.	O
industry	B-KEY
environments	I-KEY
for	O
women	B-KEY
in	O
computer	B-KEY
science	I-KEY
The	O
authors	O
were	O
tenure	O
track	O
faculty	B-KEY
members	I-KEY
at	O
the	O
Colorado	O
School	O
of	O
Mines	O
and	O
later	O
moved	O
into	O
senior	O
positions	O
at	O
software	B-KEY
companies	I-KEY
.	O
Both	O
are	O
part	O
of	O
two-career	O
couples	O
as	O
well	O
,	O
and	O
both	O
have	O
two	O
children	B-KEY
.	O
In	O
this	O
article	O
,	O
they	O
discuss	O
their	O
impressions	O
and	O
share	O
anecdotes	O
regarding	O
the	O
differing	O
experiences	O
of	O
women	B-KEY
and	O
families	O
in	O
these	O
two	O
environments	O

Computer	B-KEY
aided	I-KEY
classification	I-KEY
of	O
masses	O
in	O
ultrasonic	B-KEY
mammography	I-KEY
Frequency	B-KEY
compounding	I-KEY
was	O
recently	O
investigated	O
for	O
computer	B-KEY
aided	I-KEY
classification	I-KEY
of	O
masses	O
in	O
ultrasonic	B-KEY
B-mode	I-KEY
images	I-KEY
as	O
benign	B-KEY
or	O
malignant	B-KEY
.	O
The	O
classification	O
was	O
performed	O
using	O
the	O
normalized	B-KEY
parameters	I-KEY
of	O
the	O
Nakagami	B-KEY
distribution	I-KEY
at	O
a	O
single	B-KEY
region	I-KEY
of	I-KEY
interest	I-KEY
at	O
the	O
site	O
of	O
the	O
mass	O
.	O
A	O
combination	O
of	O
normalized	B-KEY
Nakagami	I-KEY
parameters	I-KEY
from	O
two	O
different	O
images	O
of	O
a	O
mass	O
was	O
undertaken	O
to	O
improve	O
the	O
performance	O
of	O
classification	O
.	O
Receiver	B-KEY
operating	I-KEY
characteristic	I-KEY
-LRB-	O
ROC	O
-RRB-	O
analysis	O
showed	O
that	O
such	O
an	O
approach	O
resulted	O
in	O
an	O
area	O
of	O
0.83	O
under	O
the	O
ROC	B-KEY
curve	I-KEY
.	O
The	O
aim	O
of	O
the	O
work	O
described	O
in	O
this	O
paper	O
is	O
to	O
see	O
whether	O
a	O
feature	O
describing	O
the	O
characteristic	O
of	O
the	O
boundary	O
can	O
be	O
extracted	O
and	O
combined	O
with	O
the	O
Nakagami	O
parameter	O
to	O
further	O
improve	O
the	O
performance	O
of	O
classification	O
.	O
The	O
combination	O
of	O
the	O
features	O
has	O
been	O
performed	O
using	O
a	O
weighted	B-KEY
summation	I-KEY
.	O
Results	O
indicate	O
a	O
10	O
%	O
improvement	O
in	O
specificity	B-KEY
at	O
a	O
sensitivity	B-KEY
of	O
96	O
%	O
after	O
combining	O
the	O
information	O
at	O
the	O
site	O
and	O
at	O
the	O
boundary	O
.	O
Moreover	O
,	O
the	O
technique	O
requires	O
minimal	B-KEY
clinical	I-KEY
intervention	I-KEY
and	O
has	O
a	O
performance	O
that	O
reaches	O
that	O
of	O
the	O
trained	O
radiologist	O
.	O
It	O
is	O
hence	O
suggested	O
that	O
this	O
technique	O
may	O
be	O
utilized	O
in	O
practice	O
to	O
characterize	O
breast	B-KEY
masses	I-KEY

A	O
unifying	O
co-operative	B-KEY
Web	I-KEY
caching	I-KEY
architecture	I-KEY
Network	B-KEY
caching	I-KEY
of	O
objects	O
has	O
become	O
a	O
standard	O
way	O
of	O
reducing	O
network	O
traffic	O
and	O
latency	O
in	O
the	O
Web	O
.	O
However	O
,	O
Web	O
caches	O
exhibit	O
poor	O
performance	O
with	O
a	O
hit	O
rate	O
of	O
about	O
30	O
%	O
.	O
A	O
solution	O
to	O
improve	O
this	O
hit	O
rate	O
is	O
to	O
have	O
a	O
group	O
of	O
proxies	O
form	O
co-operation	O
where	O
objects	O
can	O
be	O
cached	O
for	O
later	O
retrieval	O
.	O
A	O
cooperative	B-KEY
cache	I-KEY
system	I-KEY
includes	O
protocols	B-KEY
for	O
hierarchical	O
and	O
transversal	B-KEY
caching	I-KEY
.	O
The	O
drawback	O
of	O
such	O
a	O
system	O
lies	O
in	O
the	O
resulting	O
network	B-KEY
load	I-KEY
due	O
to	O
the	O
number	O
of	O
messages	O
that	O
need	O
to	O
be	O
exchanged	O
to	O
locate	O
an	O
object	O
.	O
This	O
paper	O
proposes	O
a	O
new	O
co-operative	B-KEY
Web	I-KEY
caching	I-KEY
architecture	I-KEY
,	O
which	O
unifies	O
previous	O
methods	O
of	O
Web	O
caching	O
.	O
Performance	O
results	O
shows	O
that	O
the	O
architecture	O
achieve	O
up	O
to	O
70	O
%	O
co-operative	B-KEY
hit	I-KEY
rate	I-KEY
and	O
accesses	O
the	O
cached	O
object	O
in	O
at	O
most	O
two	O
hops	O
.	O
Moreover	O
,	O
the	O
architecture	O
is	O
scalable	O
with	O
low	O
traffic	O
and	O
database	O
overhead	O

Robust	B-KEY
stability	I-KEY
analysis	I-KEY
for	O
current-programmed	B-KEY
regulators	I-KEY
Uncertainty	B-KEY
models	I-KEY
for	O
the	O
three	O
basic	O
switch-mode	O
converters	O
:	O
buck	O
,	O
boost	O
,	O
and	O
buck-boost	O
are	O
given	O
in	O
this	O
paper	O
.	O
The	O
resulting	O
models	O
are	O
represented	O
by	O
linear	B-KEY
fractional	I-KEY
transformations	I-KEY
with	O
structured	B-KEY
dynamic	I-KEY
uncertainties	I-KEY
.	O
Uncertainties	O
are	O
assumed	O
for	O
the	O
load	B-KEY
resistance	I-KEY
R	O
=	O
R/sub	O
O	O
/	O
-LRB-	O
1	O
+	O
delta	O
/	O
sub	O
R	O
/	O
-RRB-	O
,	O
inductance	B-KEY
L	O
=	O
L/sub	O
O	O
/	O
-LRB-	O
1	O
+	O
delta	O
/	O
sub	O
L	O
/	O
-RRB-	O
,	O
and	O
capacitance	B-KEY
C	O
=	O
C/sub	O
O	O
/	O
-LRB-	O
1	O
+	O
delta	O
/	O
sub	O
C	O
/	O
-RRB-	O
.	O
The	O
interest	O
in	O
these	O
models	O
is	O
clearly	O
motivated	O
by	O
the	O
need	O
to	O
have	O
models	O
for	O
switch-mode	B-KEY
DC-DC	I-KEY
converters	I-KEY
that	O
are	O
compatible	O
with	O
robust	O
control	B-KEY
analysis	I-KEY
,	O
which	O
require	O
a	O
model	O
structure	O
consisting	O
of	O
a	O
nominal	B-KEY
model	I-KEY
and	O
a	O
norm-bounded	B-KEY
modeling	I-KEY
uncertainty	I-KEY
.	O
Therefore	O
,	O
robust	B-KEY
stability	I-KEY
analysis	I-KEY
can	O
be	O
realized	O
using	O
standard	O
mu	O
-	O
tools	O
.	O
At	O
the	O
end	O
of	O
the	O
paper	O
,	O
an	O
illustrative	O
example	O
is	O
given	O
which	O
shows	O
the	O
simplicity	O
of	O
the	O
procedure	O

Work	B-KEY
sequencing	I-KEY
in	O
a	O
manufacturing	B-KEY
cell	I-KEY
with	O
limited	B-KEY
labour	I-KEY
constraints	I-KEY
This	O
study	O
focuses	O
on	O
the	O
analysis	O
of	O
group	B-KEY
scheduling	I-KEY
heuristics	I-KEY
in	O
a	O
dual-constrained	O
,	O
automated	B-KEY
manufacturing	I-KEY
cell	I-KEY
,	O
where	O
labour	O
utilization	O
is	O
limited	O
to	O
setups	O
,	O
tear-downs	O
and	O
loads/unloads	O
.	O
This	O
scenario	O
is	O
realistic	O
in	O
today	O
's	O
automated	B-KEY
manufacturing	I-KEY
cells	I-KEY
.	O
The	O
results	O
indicate	O
that	O
policies	O
for	O
allocating	O
labour	O
to	O
tasks	O
have	O
very	O
little	O
impact	O
in	O
such	O
an	O
environment	O
.	O
Furthermore	O
,	O
the	O
performance	O
of	O
efficiency	O
oriented	O
,	O
exhaustive	O
,	O
group	B-KEY
scheduling	I-KEY
heuristics	I-KEY
deteriorated	O
while	O
the	O
performance	O
of	O
the	O
more	O
complex	O
,	O
non-exhaustive	O
heuristics	O
improved	O
.	O
Thus	O
,	O
it	O
is	O
recommended	O
that	O
production	O
managers	O
use	O
the	O
simplest	O
labour	O
scheduling	O
policy	O
,	O
and	O
instead	O
focus	O
their	O
efforts	O
to	O
activities	O
such	O
as	O
job	B-KEY
scheduling	I-KEY
and	O
production	B-KEY
planning	I-KEY
in	O
such	O
environments	O

Acts	O
to	O
facts	O
catalogue	O
The	O
paper	O
shows	O
a	O
way	O
to	O
satisfy	O
users	O
'	O
changing	O
and	O
specific	O
information	B-KEY
needs	I-KEY
by	O
providing	O
the	O
modified	O
format-author-collaborators-title-series-subject	O
-LRB-	O
FACTS	O
-RRB-	O
.	O
catalogue	O
instead	O
of	O
the	O
traditional	O
author-collaborator-title-series-subjects	O
-LRB-	O
ACTS	O
-RRB-	O
catalogue	O

TCRM	B-KEY
:	O
diagnosing	O
tuple	B-KEY
inconsistency	I-KEY
for	O
granulized	B-KEY
datasets	I-KEY
Many	O
approaches	O
to	O
granularization	B-KEY
have	O
been	O
presented	O
for	O
knowledge	B-KEY
discovery	I-KEY
.	O
However	O
,	O
the	O
inconsistent	O
tuples	O
that	O
exist	O
in	O
granulized	B-KEY
datasets	I-KEY
are	O
hardly	O
ever	O
revealed	O
.	O
We	O
developed	O
a	O
model	O
,	O
tuple	B-KEY
consistency	I-KEY
recognition	I-KEY
model	I-KEY
-LRB-	O
TCRM	B-KEY
-RRB-	O
to	O
help	O
efficiently	O
detect	O
inconsistent	O
tuples	O
for	O
datasets	O
that	O
are	O
granulized	O
.	O
The	O
main	O
outputs	O
of	O
the	O
developed	O
model	O
include	O
explored	O
inconsistent	O
tuples	O
and	O
consumed	O
processing	B-KEY
time	I-KEY
.	O
We	O
further	O
conducted	O
an	O
empirical	O
test	O
where	O
eighteen	O
continuous	O
real-life	O
datasets	O
granulized	O
by	O
the	O
equal	B-KEY
width	I-KEY
interval	I-KEY
technique	I-KEY
that	O
embedded	O
S-plus	B-KEY
histogram	I-KEY
binning	I-KEY
algorithm	I-KEY
-LRB-	O
SHBA	O
-RRB-	O
and	O
largest	B-KEY
binning	I-KEY
size	I-KEY
algorithm	I-KEY
-LRB-	O
LBSA	O
-RRB-	O
binning	O
algorithms	O
were	O
diagnosed	O
.	O
Remarkable	O
results	O
:	O
almost	O
40	O
%	O
of	O
the	O
granulized	B-KEY
datasets	I-KEY
contain	O
inconsistent	O
tuples	O
and	O
22	O
%	O
have	O
the	O
amount	O
of	O
inconsistent	O
tuples	O
more	O
than	O
20	O
%	O

New	O
thinking	O
on	O
rendering	B-KEY
Looks	O
at	O
how	O
graphics	B-KEY
hardware	I-KEY
solves	O
a	O
range	O
of	O
rendering	B-KEY
problems	O

Car-caravan	B-KEY
snaking	I-KEY
.	O
1	O
.	O
The	O
influence	O
of	O
pintle	B-KEY
pin	I-KEY
friction	I-KEY
A	O
brief	O
review	O
of	O
knowledge	O
of	O
car-caravan	B-KEY
snaking	I-KEY
is	O
carried	O
out	O
.	O
Against	O
the	O
background	O
described	O
,	O
a	O
fairly	O
detailed	O
mathematical	B-KEY
model	I-KEY
of	O
a	O
contemporary	O
car-trailer	B-KEY
system	I-KEY
is	O
constructed	O
and	O
a	O
baseline	O
set	O
of	O
parameter	O
values	O
is	O
given	O
.	O
In	O
reduced	O
form	O
,	O
the	O
model	O
is	O
shown	O
to	O
give	O
results	O
in	O
accordance	O
with	O
literature	O
.	O
The	O
properties	O
of	O
the	O
baseline	O
combination	O
are	O
explored	O
using	O
both	O
linear	O
and	O
non-linear	O
versions	O
of	O
the	O
model	O
.	O
The	O
influences	O
of	O
damping	O
at	O
the	O
pintle	O
joint	O
and	O
of	O
several	O
other	O
design	O
parameters	O
on	O
the	O
stability	O
of	O
the	O
linear	B-KEY
system	I-KEY
in	O
the	O
neighbourhood	O
of	O
the	O
critical	B-KEY
snaking	I-KEY
speed	I-KEY
are	O
calculated	O
and	O
discussed	O
.	O
Coulomb	B-KEY
friction	I-KEY
damping	I-KEY
at	O
the	O
pintle	O
pin	O
is	O
then	O
included	O
and	O
simulations	O
are	O
used	O
to	O
indicate	O
the	O
consequent	O
amplitude-dependent	B-KEY
behaviour	I-KEY
.	O
The	O
friction	O
damping	O
,	O
especially	O
when	O
its	O
level	O
has	O
to	O
be	O
chosen	O
by	O
the	O
user	O
,	O
is	O
shown	O
to	O
give	O
dangerous	O
characteristics	O
,	O
despite	O
having	O
some	O
capacity	O
for	O
stabilization	O
of	O
the	O
snaking	O
motions	O
.	O
It	O
is	O
concluded	O
that	O
pintle	B-KEY
pin	I-KEY
friction	I-KEY
damping	O
does	O
not	O
represent	O
a	O
satisfactory	O
solution	O
to	O
the	O
snaking	O
problem	O
.	O
The	O
paper	O
sets	O
the	O
scene	O
for	O
the	O
development	O
of	O
an	O
improved	O
solution	O

Pairwise	B-KEY
thermal	I-KEY
entanglement	I-KEY
in	O
the	O
n-qubit	O
-LRB-	O
n	O
<	O
or	O
=	O
5	O
-RRB-	O
Heisenberg	O
XX	O
chain	O
We	O
have	O
calculated	O
the	O
concurrence	O
of	O
the	O
pairwise	B-KEY
thermal	I-KEY
entanglement	I-KEY
for	O
the	O
four-qubit	O
and	O
five-qubit	B-KEY
Heisenberg	I-KEY
XX	I-KEY
chain	I-KEY
.	O
It	O
is	O
found	O
that	O
there	O
is	O
a	O
great	O
difference	O
between	O
the	O
even-qubit	O
and	O
the	O
odd-qubit	B-KEY
chain	I-KEY
in	O
the	O
aspect	O
of	O
the	O
critical	B-KEY
temperature	I-KEY
and	O
of	O
the	O
existence	O
of	O
the	O
entanglement	O
for	O
the	O
case	O
of	O
the	O
qubit	O
number	O
n	O
no	O
more	O
than	O
5	O

Will	O
CPXe	B-KEY
save	O
the	O
photofinishing	B-KEY
market	I-KEY
?	O
A	O
consortium	O
of	O
film	O
suppliers	O
and	O
electronics	O
firms	O
has	O
proposed	O
the	O
Common	B-KEY
Picture	I-KEY
Exchange	I-KEY
environment	I-KEY
.	O
It	O
will	O
let	O
diverse	O
providers	O
cooperate	O
via	O
the	O
Internet	O
to	O
sell	O
digital-photo	O
prints	O

Simulating	O
fermions	O
on	O
a	O
quantum	B-KEY
computer	I-KEY
The	O
real-time	B-KEY
probabilistic	I-KEY
simulation	I-KEY
of	O
quantum	O
systems	O
in	O
classical	B-KEY
computers	I-KEY
is	O
known	O
to	O
be	O
limited	O
by	O
the	O
so-called	O
dynamical	B-KEY
sign	I-KEY
problem	I-KEY
,	O
a	O
problem	O
leading	O
to	O
exponential	O
complexity	O
.	O
In	O
1981	O
Richard	O
Feynman	O
raised	O
some	O
provocative	O
questions	O
in	O
connection	O
to	O
the	O
``	O
exact	O
imitation	O
''	O
of	O
such	O
systems	O
using	O
a	O
special	O
device	O
named	O
a	O
``	O
quantum	B-KEY
computer	I-KEY
''	O
.	O
Feynman	O
hesitated	O
about	O
the	O
possibility	O
of	O
imitating	O
fermion	B-KEY
systems	I-KEY
using	O
such	O
a	O
device	O
.	O
Here	O
we	O
address	O
some	O
of	O
his	O
concerns	O
and	O
,	O
in	O
particular	O
,	O
investigate	O
the	O
simulation	O
of	O
fermionic	B-KEY
systems	I-KEY
.	O
We	O
show	O
how	O
quantum	B-KEY
computers	I-KEY
avoid	O
the	O
sign	B-KEY
problem	I-KEY
in	O
some	O
cases	O
by	O
reducing	O
the	O
complexity	O
from	O
exponential	O
to	O
polynomial	O
.	O
Our	O
demonstration	O
is	O
based	O
upon	O
the	O
use	O
of	O
isomorphisms	B-KEY
of	O
algebras	B-KEY
.	O
We	O
present	O
specific	O
quantum	O
algorithms	O
that	O
illustrate	O
the	O
main	O
points	O
of	O
our	O
algebraic	B-KEY
approach	O

A	O
brief	O
guide	O
to	O
competitive	B-KEY
intelligence	I-KEY
:	O
how	O
to	O
gather	O
and	O
use	O
information	O
on	O
competitors	O
The	O
author	O
outlines	O
the	O
processes	O
involved	O
in	O
competitive	B-KEY
intelligence	I-KEY
,	O
and	O
discusses	O
what	O
it	O
is	O
,	O
how	O
to	O
do	O
it	O
and	O
gives	O
examples	O
of	O
what	O
happens	O
when	O
companies	O
fail	O
to	O
monitor	O
their	O
competitive	O
environment	O
effectively	O
.	O
The	O
author	O
presents	O
a	O
case	O
study	O
,	O
showing	O
how	O
the	O
company	O
that	O
produced	O
the	O
pre-cursor	O
to	O
the	O
Barbie	B-KEY
doll	I-KEY
failed	O
to	O
look	O
at	O
their	O
business	B-KEY
environment	I-KEY
and	O
how	O
this	O
led	O
to	O
the	O
firm	O
's	O
failure	O
.	O
The	O
author	O
discusses	O
what	O
competitive	B-KEY
intelligence	I-KEY
is	O
,	O
and	O
what	O
it	O
is	O
not	O
,	O
and	O
why	O
it	O
is	O
important	O
for	O
businesses	O
,	O
and	O
presents	O
three	O
models	O
used	O
to	O
describe	O
the	O
competitive	B-KEY
intelligence	I-KEY
process	O
,	O
going	O
through	O
the	O
various	O
steps	O
involved	O
in	O
defining	O
intelligence	O
requirements	O
and	O
collecting	O
,	O
analyzing	O
,	O
communicating	O
and	O
utilizing	O
competitive	B-KEY
intelligence	I-KEY

Adaptive	B-KEY
image	I-KEY
enhancement	I-KEY
for	O
retinal	O
blood	O
vessel	O
segmentation	O
Retinal	B-KEY
blood	I-KEY
vessel	I-KEY
images	I-KEY
are	O
enhanced	O
by	O
removing	O
the	O
nonstationary	O
background	O
,	O
which	O
is	O
adaptively	O
estimated	O
based	O
on	O
local	B-KEY
neighbourhood	I-KEY
information	I-KEY
.	O
The	O
result	O
is	O
a	O
much	O
better	O
segmentation	O
of	O
the	O
blood	O
vessels	O
with	O
a	O
simple	O
algorithm	O
and	O
without	O
the	O
need	O
to	O
obtain	O
a	O
priori	O
illumination	O
knowledge	O
of	O
the	O
imaging	O
system	O

Evolving	O
robust	O
asynchronous	B-KEY
cellular	I-KEY
automata	I-KEY
for	O
the	O
density	O
task	O
In	O
this	O
paper	O
the	O
evolution	O
of	O
three	O
kinds	O
of	O
asynchronous	B-KEY
cellular	I-KEY
automata	I-KEY
are	O
studied	O
for	O
the	O
density	O
task	O
.	O
Results	O
are	O
compared	O
with	O
those	O
obtained	O
for	O
synchronous	B-KEY
automata	I-KEY
and	O
the	O
influence	O
of	O
various	O
asynchronous	O
update	O
policies	O
on	O
the	O
computational	O
strategy	O
is	O
described	O
.	O
How	O
synchronous	O
and	O
asynchronous	B-KEY
cellular	I-KEY
automata	I-KEY
behave	O
is	O
investigated	O
when	O
the	O
update	O
policy	O
is	O
gradually	O
changed	O
,	O
showing	O
that	O
asynchronous	O
cellular	O
automata	O
are	O
more	O
adaptable	O
.	O
The	O
behavior	O
of	O
synchronous	O
and	O
asynchronous	O
evolved	O
automata	O
are	O
studied	O
under	O
the	O
presence	O
of	O
random	B-KEY
noise	I-KEY
of	O
two	O
kinds	O
and	O
it	O
is	O
shown	O
that	O
asynchronous	B-KEY
cellular	I-KEY
automata	I-KEY
implicitly	O
offer	O
superior	O
fault	O
tolerance	O

Effect	O
of	O
insulation	B-KEY
layer	I-KEY
on	O
transcribability	B-KEY
and	O
birefringence	B-KEY
distribution	I-KEY
in	O
optical	B-KEY
disk	I-KEY
substrate	I-KEY
As	O
the	O
need	O
for	O
information	B-KEY
storage	I-KEY
media	I-KEY
with	O
high	B-KEY
storage	I-KEY
density	I-KEY
increases	O
,	O
digital	B-KEY
video	I-KEY
disks	I-KEY
-LRB-	O
DVDs	O
-RRB-	O
with	O
smaller	B-KEY
recording	I-KEY
marks	I-KEY
and	O
thinner	B-KEY
optical	I-KEY
disk	I-KEY
substrates	I-KEY
than	O
those	O
of	O
conventional	O
DVDs	O
are	O
being	O
required	O
.	O
Therefore	O
,	O
improving	O
the	O
replication	B-KEY
quality	I-KEY
of	O
land-groove	B-KEY
or	O
pit	B-KEY
structure	I-KEY
and	O
reducing	O
the	O
birefringence	B-KEY
distribution	I-KEY
are	O
emerging	O
as	O
important	O
criteria	O
in	O
the	O
fabrication	B-KEY
of	O
high-density	O
optical	B-KEY
disk	I-KEY
substrates	I-KEY
.	O
We	O
control	O
the	O
transcribability	B-KEY
and	O
distribution	O
of	O
birefringence	O
by	O
inserting	O
an	O
insulation	B-KEY
layer	I-KEY
under	O
the	O
stamper	B-KEY
during	O
injection-compression	B-KEY
molding	I-KEY
of	O
DVD	B-KEY
RAM	I-KEY
substrates	I-KEY
.	O
The	O
effects	O
of	O
the	O
insulation	B-KEY
layer	I-KEY
on	O
the	O
geometrical	O
and	O
optical	B-KEY
properties	I-KEY
,	O
such	O
as	O
transcribability	B-KEY
and	O
birefringence	B-KEY
distribution	I-KEY
,	O
are	O
examined	O
experimentally	O
.	O
The	O
inserted	O
insulation	B-KEY
layer	I-KEY
is	O
found	O
to	O
be	O
very	O
effective	O
in	O
improving	O
the	O
quality	O
of	O
replication	O
and	O
leveling	O
out	O
the	O
first	O
peak	O
of	O
the	O
gapwise	B-KEY
birefringence	I-KEY
distribution	I-KEY
near	O
the	O
mold	O
wall	O
and	O
reducing	O
the	O
average	O
birefringence	O
value	O
,	O
because	O
the	O
insulation	O
layer	O
retarded	O
the	O
growth	O
of	O
the	O
solidified	O
layer	O

Topology-reducing	B-KEY
surface	I-KEY
simplification	I-KEY
using	O
a	O
discrete	B-KEY
solid	I-KEY
representation	I-KEY
This	O
paper	O
presents	O
a	O
new	O
approach	O
for	O
generating	O
coarse-level	B-KEY
approximations	I-KEY
of	O
topologically	B-KEY
complex	I-KEY
models	I-KEY
.	O
Dramatic	O
topology	O
reduction	O
is	O
achieved	O
by	O
converting	O
a	O
3D	B-KEY
model	I-KEY
to	O
and	O
from	O
a	O
volumetric	B-KEY
representation	I-KEY
.	O
Our	O
approach	O
produces	O
valid	O
,	O
error-bounded	B-KEY
models	I-KEY
and	O
supports	O
the	O
creation	O
of	O
approximations	O
that	O
do	O
not	O
interpenetrate	O
the	O
original	O
model	O
,	O
either	O
being	O
completely	O
contained	O
in	O
the	O
input	O
solid	O
or	O
bounding	O
it	O
.	O
Several	O
simple	O
to	O
implement	O
versions	O
of	O
our	O
approach	O
are	O
presented	O
and	O
discussed	O
.	O
We	O
show	O
that	O
these	O
methods	O
perform	O
significantly	O
better	O
than	O
other	O
surface-based	O
approaches	O
when	O
simplifying	O
topologically-rich	O
models	O
such	O
as	O
scene	B-KEY
parts	I-KEY
and	O
complex	B-KEY
mechanical	I-KEY
assemblies	I-KEY

Understanding	O
Internet	B-KEY
traffic	I-KEY
streams	I-KEY
:	O
dragonflies	B-KEY
and	O
tortoises	B-KEY
We	O
present	O
the	O
concept	O
of	O
network	B-KEY
traffic	I-KEY
streams	I-KEY
and	O
the	O
ways	O
they	O
aggregate	O
into	O
flows	O
through	O
Internet	O
links	O
.	O
We	O
describe	O
a	O
method	O
of	O
measuring	O
the	O
size	O
and	O
lifetime	O
of	O
Internet	O
streams	O
,	O
and	O
use	O
this	O
method	O
to	O
characterize	O
traffic	B-KEY
distributions	I-KEY
at	O
two	O
different	O
sites	O
.	O
We	O
find	O
that	O
although	O
most	O
streams	O
-LRB-	O
about	O
45	O
percent	O
of	O
them	O
-RRB-	O
are	O
dragonflies	B-KEY
,	O
lasting	O
less	O
than	O
2	O
seconds	O
,	O
a	O
significant	O
number	O
of	O
streams	O
have	O
lifetimes	O
of	O
hours	O
to	O
days	O
,	O
and	O
can	O
carry	O
a	O
high	O
proportion	O
-LRB-	O
50-60	O
percent	O
-RRB-	O
of	O
the	O
total	O
bytes	O
on	O
a	O
given	O
link	O
.	O
We	O
define	O
tortoises	B-KEY
as	O
streams	O
that	O
last	O
longer	O
than	O
15	O
minutes	O
.	O
We	O
point	O
out	O
that	O
streams	O
can	O
be	O
classified	O
not	O
only	O
by	O
lifetime	O
-LRB-	O
dragonflies	B-KEY
and	O
tortoises	B-KEY
-RRB-	O
but	O
also	O
by	O
size	O
-LRB-	O
mice	B-KEY
and	O
elephants	B-KEY
-RRB-	O
,	O
and	O
note	O
that	O
stream	O
size	O
and	O
lifetime	O
are	O
independent	O
dimensions	O
.	O
We	O
submit	O
that	O
ISPs	B-KEY
need	O
to	O
be	O
aware	O
of	O
the	O
distribution	O
of	O
Internet	O
stream	O
sizes	O
,	O
and	O
the	O
impact	O
of	O
the	O
difference	O
in	O
behavior	O
between	O
short	O
and	O
long	O
streams	O
.	O
In	O
particular	O
,	O
any	O
forwarding	B-KEY
cache	I-KEY
mechanisms	I-KEY
in	O
Internet	B-KEY
routers	I-KEY
must	O
be	O
able	O
to	O
cope	O
with	O
a	O
high	O
volume	O
of	O
short	O
streams	O
.	O
In	O
addition	O
ISPs	B-KEY
should	O
realize	O
that	O
long-running	B-KEY
streams	I-KEY
can	O
contribute	O
a	O
significant	O
fraction	O
of	O
their	O
packet	O
and	O
byte	O
volumes-something	O
they	O
may	O
not	O
have	O
allowed	O
for	O
when	O
using	O
traditional	O
``	O
flat	O
rate	O
user	O
bandwidth	O
consumption	O
''	O
approaches	O
to	O
provisioning	O
and	O
engineering	O

FLID-DL	B-KEY
:	O
congestion	B-KEY
control	I-KEY
for	O
layered	O
multicast	O
We	O
describe	O
fair	O
layered	O
increase/decrease	O
with	O
dynamic	B-KEY
layering	I-KEY
-LRB-	O
FLID-DL	O
-RRB-	O
:	O
a	O
new	O
multirate	O
congestion	O
control	O
algorithm	O
for	O
layered	O
multicast	O
sessions	O
.	O
FLID-DL	B-KEY
generalizes	O
the	O
receiver-driven	O
layered	O
congestion	B-KEY
control	I-KEY
protocol	O
-LRB-	O
RLC	O
-RRB-	O
introduced	O
by	O
Vicisano	O
et	O
al.	O
-LRB-	O
Proc	O
.	O
IEEE	O
INFOCOM	O
,	O
San	O
Francisco	O
,	O
CA	O
,	O
,	O
p.996-1003	O
,	O
Mar.	O
1998	O
-RRB-	O
ameliorating	O
the	O
problems	O
associated	O
with	O
large	O
Internet	O
group	B-KEY
management	I-KEY
protocol	I-KEY
-LRB-	I-KEY
IGMP	O
-RRB-	B-KEY
leave	O
latencies	O
and	O
abrupt	O
rate	O
increases	O
.	O
Like	O
RLC	O
,	O
FLID-DL	B-KEY
,	O
is	O
a	O
scalable	O
,	O
receiver-driven	O
congestion	B-KEY
control	I-KEY
mechanism	O
in	O
which	O
receivers	O
add	O
layers	O
at	O
sender-initiated	B-KEY
synchronization	I-KEY
points	O
and	O
leave	O
layers	O
when	O
they	O
experience	O
congestion	O
.	O
FLID-DL	B-KEY
congestion	B-KEY
control	I-KEY
coexists	O
with	O
transmission	B-KEY
control	I-KEY
protocol	I-KEY
-LRB-	O
TCP	O
-RRB-	O
flows	O
as	O
well	O
as	O
other	O
FLID-DL	B-KEY
sessions	O
and	O
supports	O
general	O
rates	O
on	O
the	O
different	O
multicast	B-KEY
layers	I-KEY
.	O
We	O
demonstrate	O
via	O
simulations	B-KEY
that	O
our	O
congestion	B-KEY
control	I-KEY
scheme	O
exhibits	O
better	O
fairness	O
properties	O
and	O
provides	O
better	O
throughput	B-KEY
than	O
previous	O
methods	O
.	O
A	O
key	O
contribution	O
that	O
enables	O
FLID-DL	B-KEY
and	O
may	O
be	O
useful	O
elsewhere	O
is	O
dynamic	B-KEY
layering	I-KEY
-LRB-	O
DL	O
-RRB-	O
,	O
which	O
mitigates	O
the	O
negative	O
impact	O
of	O
long	O
IGMP	B-KEY
leave	O
latencies	O
and	O
eliminates	O
the	O
need	O
for	O
probe	O
intervals	O
present	O
in	O
RLC	O
.	O
We	O
use	O
DL	O
to	O
respond	O
to	O
congestion	O
much	O
faster	O
than	O
IGMP	B-KEY
leave	O
operations	O
,	O
which	O
have	O
proven	O
to	O
be	O
a	O
bottleneck	O
in	O
practice	O
for	O
prior	O
work	O

Application	O
of	O
XML	B-KEY
for	O
neural	B-KEY
network	I-KEY
exchange	I-KEY
This	O
article	O
introduces	O
a	O
framework	O
for	O
the	O
interchange	O
of	O
trained	O
neural	O
network	O
models	O
.	O
An	O
XML-based	O
language	O
-LRB-	O
Neural	O
Network	O
Markup	O
Language	O
-RRB-	O
for	O
the	O
neural	O
network	O
model	O
description	O
is	O
offered	O
.	O
It	O
allows	O
to	O
write	O
down	O
all	O
the	O
components	O
of	O
neural	O
network	O
model	O
which	O
are	O
necessary	O
for	O
its	O
reproduction	O
.	O
We	O
propose	O
to	O
use	O
XML	B-KEY
notation	O
for	O
the	O
full	O
description	O
of	O
neural	O
models	O
,	O
including	O
data	B-KEY
dictionary	I-KEY
,	O
properties	O
of	O
training	O
sample	O
,	O
preprocessing	B-KEY
methods	I-KEY
,	O
details	O
of	O
network	B-KEY
structure	I-KEY
and	O
parameters	O
and	O
methods	O
for	O
network	B-KEY
output	I-KEY
interpretation	I-KEY

Sharpening	O
the	O
estimate	O
of	O
the	O
stability	B-KEY
constant	I-KEY
in	O
the	O
maximum-norm	O
of	O
the	O
Crank-Nicolson	B-KEY
scheme	I-KEY
for	O
the	O
one-dimensional	B-KEY
heat	I-KEY
equation	I-KEY
This	O
paper	O
is	O
concerned	O
with	O
the	O
stability	B-KEY
constant	I-KEY
C/sub	O
infinity	O
/	O
in	O
the	O
maximum-norm	O
of	O
the	O
Crank-Nicolson	B-KEY
scheme	I-KEY
applied	O
.	O
to	O
the	O
one-dimensional	B-KEY
heat	I-KEY
equation	I-KEY
.	O
A	O
well	O
known	O
result	O
due	O
to	O
S.J.	O
Serdyukova	O
is	O
that	O
C/sub	O
infinity	O
/	O
<	O
23	O
.	O
In	O
the	O
present	O
paper	O
,	O
by	O
using	O
a	O
sharp	B-KEY
resolvent	I-KEY
estimate	I-KEY
for	O
the	O
discrete	B-KEY
Laplacian	I-KEY
together	O
with	O
the	O
Cauchy	B-KEY
formula	I-KEY
,	O
it	O
is	O
shown	O
that	O
3	O
<	O
or	O
=	O
C/sub	O
infinity	O
/	O
<	O
4.325	O
.	O
This	O
bound	O
also	O
holds	O
when	O
the	O
heat	O
equation	O
is	O
considered	O
on	O
a	O
bounded	O
interval	O
along	O
with	O
Dirichlet	O
or	O
Neumann	B-KEY
boundary	I-KEY
conditions	I-KEY

Influence	O
of	O
the	O
process	O
design	O
on	O
the	O
control	O
strategy	O
:	O
application	O
in	O
electropneumatic	O
field	O
This	O
article	O
proposes	O
an	O
example	O
of	O
electropneumatic	B-KEY
system	I-KEY
where	O
the	O
architecture	O
of	O
the	O
process	O
is	O
modified	O
with	O
respect	O
to	O
both	O
the	O
specifications	O
for	O
position	O
and	O
velocity	O
tracking	B-KEY
and	O
a	O
criterion	O
concerning	O
the	O
energy	B-KEY
consumption	I-KEY
.	O
Experimental	O
results	O
are	O
compared	O
and	O
analyzed	O
using	O
an	O
industrial	O
bench	O
test	O
.	O
For	O
this	O
,	O
a	O
complete	O
model	O
of	O
the	O
system	O
is	O
presented	O
,	O
and	O
two	O
kinds	O
of	O
nonlinear	B-KEY
control	I-KEY
laws	O
are	O
developed	O
,	O
a	O
monovariable	O
and	O
multivariable	O
type	O
based	O
on	O
the	O
flatness	B-KEY
theory	I-KEY

STEM	B-KEY
:	O
Secure	B-KEY
Telephony	I-KEY
Enabled	I-KEY
Middlebox	I-KEY
Dynamic	B-KEY
applications	I-KEY
,	O
including	O
IP	B-KEY
telephony	I-KEY
,	O
have	O
not	O
seen	O
wide	O
acceptance	O
within	O
enterprises	O
because	O
of	O
problems	O
caused	O
by	O
the	O
existing	O
network	B-KEY
infrastructure	I-KEY
.	O
Static	O
elements	O
,	O
including	O
firewalls	B-KEY
and	O
network	B-KEY
address	I-KEY
translation	I-KEY
devices	I-KEY
,	O
are	O
not	O
capable	O
of	O
allowing	O
dynamic	B-KEY
applications	I-KEY
to	O
operate	O
properly	O
.	O
The	O
Secure	B-KEY
Telephony	I-KEY
Enabled	I-KEY
Middlebox	I-KEY
-LRB-	O
STEM	B-KEY
-RRB-	O
architecture	O
is	O
an	O
enhancement	O
of	O
the	O
existing	O
network	B-KEY
design	I-KEY
to	O
remove	O
the	O
issues	O
surrounding	O
static	B-KEY
devices	I-KEY
.	O
The	O
architecture	O
incorporates	O
an	O
improved	O
firewall	B-KEY
that	O
can	O
interpret	O
and	O
utilize	O
information	O
in	O
the	O
application	B-KEY
layer	I-KEY
of	O
packets	O
to	O
ensure	O
proper	O
functionality	O
.	O
In	O
addition	O
to	O
allowing	O
dynamic	B-KEY
applications	I-KEY
to	O
function	O
normally	O
,	O
the	O
STEM	B-KEY
architecture	O
also	O
incorporates	O
several	O
detection	O
and	O
response	O
mechanisms	O
for	O
well-known	O
network-based	O
vulnerabilities	O
.	O
This	O
article	O
describes	O
the	O
key	O
components	O
of	O
the	O
architecture	O
with	O
respect	O
to	O
the	O
SIP	B-KEY
protocol	I-KEY

Banking	O
on	O
SMA	O
funds	O
-LSB-	O
separately	B-KEY
managed	I-KEY
accounts	I-KEY
-RSB-	O
From	O
investment	B-KEY
management	I-KEY
to	O
technology	B-KEY
to	O
back-office	B-KEY
services	I-KEY
,	O
outsourcers	B-KEY
are	O
elbowing	O
their	O
way	O
into	O
the	O
SMA	O
business	O
.	O
Small	B-KEY
banks	I-KEY
are	O
paying	O
attention-and	O
hoping	O
to	O
reap	O
the	O
rewards	O

Design	O
and	O
prototype	O
of	O
a	O
performance	B-KEY
tool	I-KEY
interface	I-KEY
for	O
OpenMP	O
This	O
paper	O
proposes	O
a	O
performance	B-KEY
tools	I-KEY
interface	I-KEY
for	O
OpenMP	O
,	O
similar	O
in	O
spirit	O
to	O
the	O
MPI	B-KEY
profiling	I-KEY
interface	I-KEY
in	O
its	O
intent	O
to	O
define	O
a	O
clear	O
and	O
portable	O
API	B-KEY
that	O
makes	O
OpenMP	O
execution	O
events	O
visible	O
to	O
runtime	O
performance	O
tools	O
.	O
We	O
present	O
our	O
design	O
using	O
a	O
source-level	B-KEY
instrumentation	I-KEY
approach	I-KEY
based	O
on	O
OpenMP	B-KEY
directive	I-KEY
rewriting	I-KEY
.	O
Rules	O
to	O
instrument	O
each	O
directive	O
and	O
their	O
combination	O
are	O
applied	O
to	O
generate	O
calls	O
to	O
the	O
interface	O
consistent	O
with	O
directive	B-KEY
semantics	I-KEY
and	O
to	O
pass	O
context	O
information	O
-LRB-	O
e.g.	O
,	O
source	O
code	O
locations	O
-RRB-	O
in	O
a	O
portable	O
and	O
efficient	O
way	O
.	O
Our	O
proposed	O
OpenMP	O
performance	O
API	B-KEY
further	O
allows	O
user	O
functions	O
and	O
arbitrary	B-KEY
code	I-KEY
regions	I-KEY
to	O
be	O
marked	O
and	O
performance	O
measurement	O
to	O
be	O
controlled	O
using	O
new	O
OpenMP	O
directives	O
.	O
To	O
prototype	O
the	O
proposed	O
OpenMP	O
performance	O
interface	O
,	O
we	O
have	O
developed	O
compatible	O
performance	B-KEY
libraries	I-KEY
for	O
the	O
EXPERT	B-KEY
automatic	I-KEY
event	I-KEY
trace	I-KEY
analyzer	I-KEY
-LSB-	O
17	O
,	O
18	O
-RSB-	O
and	O
the	O
TAU	B-KEY
performance	I-KEY
analysis	I-KEY
framework	I-KEY
-LSB-	O
13	O
-RSB-	O
.	O
The	O
directive	O
instrumentation	O
transformations	O
we	O
define	O
are	O
implemented	O
in	O
a	O
source-to-source	B-KEY
translation	I-KEY
tool	I-KEY
called	O
OPARI	B-KEY
.	O
Application	O
examples	O
are	O
presented	O
for	O
both	O
EXPERT	O
and	O
TAU	O
to	O
show	O
the	O
OpenMP	O
performance	O
interface	O
and	O
OPARI	B-KEY
instrumentation	O
tool	O
in	O
operation	O
.	O
When	O
used	O
together	O
with	O
the	O
MPI	B-KEY
profiling	I-KEY
interface	I-KEY
-LRB-	O
as	O
the	O
examples	O
also	O
demonstrate	O
-RRB-	O
,	O
our	O
proposed	O
approach	O
provides	O
a	O
portable	O
and	O
robust	O
solution	O
to	O
performance	O
analysis	O
of	O
OpenMP	O
and	O
mixed-mode	O
-LRB-	O
OpenMP	O
+	O
MPI	O
-RRB-	O
applications	O

All	O
change	O
-LSB-	O
agile	B-KEY
business	I-KEY
-RSB-	O
What	O
does	O
it	O
take	O
for	O
an	O
organisation	O
to	O
become	O
an	O
agile	B-KEY
business	I-KEY
?	O
Its	O
employees	O
probably	O
need	O
to	O
adhere	O
to	O
new	O
procurement	O
policies	O
,	O
work	O
more	O
closely	O
with	O
colleagues	O
in	O
other	O
departments	O
,	O
meet	O
more	O
exacting	O
sales	O
targets	O
,	O
and	O
offer	O
higher	O
standards	O
of	O
customer	O
service	O
and	O
support	O
.	O
In	O
short	O
,	O
they	O
need	O
to	O
change	O
the	O
way	O
they	O
work	O
.	O
Implementing	O
technologies	O
to	O
support	O
agile	B-KEY
business	I-KEY
models	O
and	O
underpin	O
new	O
practices	O
is	O
a	O
complex	O
task	O
in	O
itself	O
.	O
But	O
getting	O
employees	O
to	O
adopt	O
new	O
practices	O
is	O
far	O
harder	O
,	O
and	O
one	O
that	O
requires	O
careful	O
handling	O
,	O
says	O
Barry	O
O'Connell	O
,	O
general	O
manager	O
of	O
business-to-employee	O
-LRB-	O
B2E	O
-RRB-	O
solutions	O
at	O
systems	O
vendor	O
Hewlett-Packard	O
-LRB-	O
HP	O
-RRB-	O

Implementation	O
of	O
DIMSIMs	B-KEY
for	O
stiff	B-KEY
differential	I-KEY
systems	I-KEY
Some	O
issues	O
related	O
to	O
the	O
implementation	O
of	O
diagonally	B-KEY
implicit	I-KEY
multistage	I-KEY
integration	I-KEY
methods	I-KEY
for	O
stiff	B-KEY
differential	I-KEY
systems	I-KEY
are	O
discussed	O
.	O
They	O
include	O
reliable	B-KEY
estimation	I-KEY
of	O
the	O
local	B-KEY
discretization	I-KEY
error	I-KEY
,	O
construction	O
of	O
continuous	O
interpolants	B-KEY
,	O
solution	O
of	O
nonlinear	B-KEY
systems	I-KEY
of	I-KEY
equations	I-KEY
by	O
simplified	B-KEY
Newton	I-KEY
iterations	I-KEY
,	O
choice	O
of	O
initial	O
stepsize	O
and	O
order	O
,	O
and	O
step	O
and	O
order	O
changing	O
strategy	O
.	O
Numerical	O
results	O
are	O
presented	O
which	O
indicate	O
that	O
an	O
experimental	B-KEY
Matlab	I-KEY
code	I-KEY
based	O
on	O
type	O
2	O
methods	O
of	O
order	O
one	O
,	O
two	O
and	O
three	O
outperforms	O
ode15s	O
code	O
from	O
Matlab	O
ODE	O
suite	O
on	O
problems	O
whose	O
Jacobian	O
has	O
eigenvalues	O
which	O
are	O
close	O
to	O
the	O
imaginary	O
axis	O

Cool	O
and	O
green	O
-LSB-	O
air	B-KEY
conditioning	I-KEY
-RSB-	O
In	O
these	O
days	O
of	O
global	O
warming	O
,	O
air	B-KEY
conditioning	I-KEY
engineers	O
need	O
to	O
specify	O
not	O
just	O
for	O
the	O
needs	O
of	O
the	O
occupants	O
,	O
but	O
also	O
to	O
maximise	O
energy	B-KEY
efficiency	I-KEY
.	O
Julian	O
Brunnock	O
outlines	O
the	O
key	O
areas	O
to	O
consider	O
for	O
energy	B-KEY
efficient	I-KEY
air	B-KEY
conditioning	I-KEY
systems	O

Sensorless	B-KEY
control	I-KEY
of	O
induction	B-KEY
motor	I-KEY
drives	I-KEY
Controlled	O
induction	B-KEY
motor	I-KEY
drives	I-KEY
without	O
mechanical	O
speed	O
sensors	O
at	O
the	O
motor	O
shaft	O
have	O
the	O
attractions	O
of	O
low	O
cost	O
and	O
high	O
reliability	B-KEY
.	O
To	O
replace	O
the	O
sensor	O
the	O
information	O
on	O
the	O
rotor	O
speed	O
is	O
extracted	O
from	O
measured	O
stator	B-KEY
voltages	I-KEY
and	O
currents	O
at	O
the	O
motor	O
terminals	O
.	O
Vector-controlled	B-KEY
drives	I-KEY
require	O
estimating	O
the	O
magnitude	B-KEY
and	O
spatial	B-KEY
orientation	I-KEY
of	O
the	O
fundamental	B-KEY
magnetic	I-KEY
flux	I-KEY
waves	I-KEY
in	O
the	O
stator	O
or	O
in	O
the	O
rotor	O
.	O
Open-loop	B-KEY
estimators	I-KEY
or	O
closed-loop	B-KEY
observers	I-KEY
are	O
used	O
for	O
this	O
purpose	O
.	O
They	O
differ	O
with	O
respect	O
to	O
accuracy	O
,	O
robustness	B-KEY
,	O
and	O
sensitivity	B-KEY
against	O
model	B-KEY
parameter	I-KEY
variations	I-KEY
.	O
Dynamic	O
performance	O
and	O
steady-state	B-KEY
speed	I-KEY
accuracy	I-KEY
in	O
the	O
low-speed	O
range	O
can	O
be	O
achieved	O
by	O
exploiting	O
parasitic	B-KEY
effects	I-KEY
of	O
the	O
machine	O
.	O
The	O
overview	O
in	O
this	O
paper	O
uses	O
signal	B-KEY
flow	I-KEY
graphs	I-KEY
of	O
complex	O
space	B-KEY
vector	I-KEY
quantities	I-KEY
to	O
provide	O
an	O
insightful	O
description	O
of	O
the	O
systems	O
used	O
in	O
sensorless	B-KEY
control	I-KEY
of	O
induction	O
motors	O

Three-dimensional	B-KEY
particle	I-KEY
image	I-KEY
tracking	I-KEY
for	O
dilute	B-KEY
particle-liquid	I-KEY
flows	I-KEY
in	O
a	O
pipe	O
A	O
three-dimensional	O
-LRB-	O
3D	O
-RRB-	O
particle	O
image	O
tracking	O
technique	O
was	O
used	O
to	O
study	O
the	O
coarse	O
spherical	O
particle-liquid	O
flows	O
in	O
a	O
pipe	O
.	O
The	O
flow	O
images	O
from	O
both	O
the	O
front	O
view	O
and	O
the	O
normal	O
side	O
view	O
,	O
which	O
was	O
reflected	O
into	O
the	O
front	O
view	O
by	O
a	O
mirror	O
,	O
were	O
recorded	O
with	O
a	O
CCD	B-KEY
camera	I-KEY
and	O
digitized	O
by	O
a	O
PC	O
with	O
an	O
image	O
grabber	O
card	O
.	O
An	O
image	O
processing	O
program	O
was	O
developed	O
to	O
enhance	O
and	O
segment	O
the	O
flow	O
image	O
,	O
and	O
then	O
to	O
identify	O
the	O
particles	O
.	O
Over	O
90	O
%	O
of	O
all	O
the	O
particles	O
can	O
be	O
identified	O
and	O
located	O
from	O
the	O
partially	O
overlapped	O
particle	O
images	O
using	O
the	O
circular	O
Hough	B-KEY
transform	I-KEY
.	O
Then	O
the	O
3D	B-KEY
position	I-KEY
of	O
each	O
detected	O
particle	O
was	O
determined	O
by	O
matching	O
its	O
front	O
view	O
image	O
to	O
its	O
side	O
view	O
image	O
.	O
The	O
particle	O
velocity	O
was	O
then	O
obtained	O
by	O
pairing	O
its	O
images	O
in	O
successive	O
video	O
fields	O
.	O
The	O
measurements	O
for	O
the	O
spherical	O
expanded	O
polystyrene	O
particle-oil	O
flows	O
show	O
that	O
the	O
particles	O
,	O
like	O
the	O
spherical	B-KEY
bubbles	I-KEY
in	O
laminar	O
bubbly	O
flows	O
,	O
tend	O
to	O
conglomerate	O
near	O
the	O
pipe	O
wall	O
and	O
to	O
line	O
up	O
to	O
form	O
the	O
particle	B-KEY
clusters	I-KEY
.	O
As	O
liquid	O
velocity	O
decreases	O
,	O
the	O
particle	B-KEY
clusters	I-KEY
disperse	O
and	O
more	O
particles	O
are	O
distributed	O
in	O
the	O
pipe	O
centre	O
region	O

Quantum	B-KEY
phase	I-KEY
gate	I-KEY
for	O
photonic	B-KEY
qubits	I-KEY
using	O
only	O
beam	O
splitters	O
and	O
postselection	B-KEY
We	O
show	O
that	O
a	O
beam	O
splitter	O
of	O
reflectivity	B-KEY
one-third	O
can	O
be	O
used	O
to	O
realize	O
a	O
quantum	B-KEY
phase	I-KEY
gate	I-KEY
operation	O
if	O
only	O
the	O
outputs	O
conserving	O
the	O
number	O
of	O
photons	O
on	O
each	O
side	O
are	O
postselected	O

Closed	O
loop	O
finite-element	O
modeling	O
of	O
active	B-KEY
constrained	I-KEY
layer	I-KEY
damping	I-KEY
in	O
the	O
time	B-KEY
domain	I-KEY
analysis	I-KEY
A	O
three-dimensional	B-KEY
finite-element	I-KEY
closed-loop	I-KEY
model	I-KEY
has	O
been	O
developed	O
to	O
predict	O
the	O
effects	O
of	O
active-passive	O
damping	O
on	O
a	O
vibrating	O
structure	O
.	O
The	O
Golla-Hughes-McTavish	B-KEY
method	I-KEY
is	O
employed	O
to	O
capture	O
the	O
viscoelastic	B-KEY
material	I-KEY
behavior	O
in	O
a	O
time	B-KEY
domain	I-KEY
analysis	I-KEY
.	O
The	O
parametric	O
study	O
includes	O
the	O
different	O
control	O
gains	O
as	O
well	O
as	O
geometric	O
parameters	O
related	O
to	O
the	O
active	B-KEY
constrained	I-KEY
layer	I-KEY
damping	I-KEY
-LRB-	O
ACLD	O
-RRB-	O
treatment	O
.	O
Comparisons	O
are	O
made	O
among	O
several	O
ACLD	B-KEY
models	I-KEY
,	O
the	O
passive	B-KEY
constrained	I-KEY
model	I-KEY
and	O
the	O
active	B-KEY
damping	I-KEY
model	I-KEY
.	O
The	O
results	O
obtained	O
here	O
reiterate	O
that	O
ACLD	O
is	O
somewhat	O
better	O
for	O
vibration	B-KEY
suppression	I-KEY
than	O
either	O
the	O
purely	O
passive	O
or	O
the	O
active	O
system	O
and	O
provides	O
higher	O
structural	B-KEY
damping	I-KEY
with	O
less	O
control	O
gain	O
when	O
compared	O
to	O
the	O
purely	O
active	O
system	O
.	O
Since	O
the	O
ACLD	O
performance	O
can	O
be	O
reduced	O
by	O
the	O
viscoelastic	B-KEY
layer	I-KEY
,	O
the	O
design	O
of	O
the	O
ACLD	B-KEY
model	I-KEY
must	O
be	O
given	O
a	O
careful	O
consideration	O
in	O
order	O
to	O
optimize	O
the	O
effect	O
of	O
passive	B-KEY
damping	I-KEY

On	O
conflict-free	B-KEY
executions	I-KEY
of	O
elementary	O
nets	O
Deals	O
with	O
analysis	O
of	O
elementary	B-KEY
Petri	I-KEY
nets	I-KEY
with	O
respect	O
to	O
possibilities	O
of	O
avoiding	O
conflicts	O
during	O
their	O
executions	O
.	O
There	O
are	O
two	O
main	O
aims	O
of	O
the	O
paper	O
.	O
The	O
first	O
is	O
to	O
find	O
a	O
method	O
of	O
checking	O
if	O
a	O
net	O
is	O
conflict-avoidable	O
-LRB-	O
i.e.	O
,	O
if	O
it	O
possesses	O
a	O
conflict-free	B-KEY
fair	I-KEY
run	I-KEY
-RRB-	O
.	O
The	O
second	O
is	O
to	O
find	O
a	O
method	O
of	O
rebuilding	O
any	O
net	O
to	O
a	O
totally	B-KEY
conflict-avoidable	I-KEY
net	I-KEY
-LRB-	O
i.e.	O
,	O
a	O
net	O
possessing	O
a	O
conflict-free	B-KEY
fair	I-KEY
run	I-KEY
in	O
every	O
one	O
process	O
-RRB-	O
with	O
the	O
same	O
behaviour	O
.	O
The	O
main	O
results	O
are	O
the	O
following	O
:	O
1	O
.	O
The	O
proof	O
of	O
decidability	B-KEY
,	O
for	O
elementary	O
nets	O
,	O
of	O
the	O
problem	O
of	O
existence	O
of	O
a	O
conflict-avoidable	O
fair	O
process	O
-LRB-	O
and	O
an	O
algorithm	O
producing	O
all	O
fair	O
runs	O
-RRB-	O
.	O
2	O
.	O
Construction	O
,	O
for	O
an	O
arbitrary	O
given	O
elementary	O
net	O
,	O
of	O
a	O
totally	B-KEY
conflict-avoidable	I-KEY
net	I-KEY
with	O
the	O
same	O
behaviour	O
.	O
The	O
net	O
,	O
completed	O
this	O
way	O
,	O
has	O
the	O
same	O
behaviour	O
as	O
the	O
original	O
one	O
.	O
Moreover	O
,	O
it	O
is	O
totally	O
conflict-avoidable	O
,	O
and	O
its	O
execution	O
may	O
be	O
supervised	O
-LRB-	O
in	O
order	O
to	O
ensure	O
conflict-freeness	O
-RRB-	O
by	O
the	O
reduced	B-KEY
case	I-KEY
graph	I-KEY
built	O
by	O
the	O
algorithm	O
of	O
the	O
former	O
section	O

An	O
overview	O
of	O
modems	B-KEY
This	O
paper	O
describes	O
cursory	O
glance	O
of	O
different	O
types	O
of	O
modems	B-KEY
classified	O
for	O
application	O
,	O
range	O
,	O
line	B-KEY
type	I-KEY
,	O
operating	B-KEY
mode	I-KEY
,	O
synchronizing	B-KEY
mode	I-KEY
,	O
modulation	B-KEY
,	O
etc.	O
,	O
highly	O
useful	O
for	O
all	O
engineering	B-KEY
students	I-KEY
of	O
communication	O
,	O
electrical	O
,	O
computer	O
science	O
and	O
information	B-KEY
technology	I-KEY
students	I-KEY
.	O
This	O
paper	O
also	O
describes	O
the	O
standards	B-KEY
and	O
protocols	B-KEY
used	O
and	O
the	O
future	O
trend	O

Towards	O
an	O
ontology	B-KEY
of	O
approximate	B-KEY
reason	I-KEY
This	O
article	O
introduces	O
structural	O
aspects	O
in	O
an	O
ontology	B-KEY
of	O
approximate	B-KEY
reason	I-KEY
.	O
The	O
basic	O
assumption	O
in	O
this	O
ontology	B-KEY
is	O
that	O
approximate	B-KEY
reason	I-KEY
is	O
a	O
capability	O
of	O
an	O
agent	O
.	O
Agents	O
are	O
designed	O
to	O
classify	O
information	B-KEY
granules	I-KEY
derived	O
from	O
sensors	O
that	O
respond	O
to	O
stimuli	O
in	O
the	O
environment	O
of	O
an	O
agent	O
or	O
received	O
from	O
other	O
agents	O
.	O
Classification	O
of	O
information	B-KEY
granules	I-KEY
is	O
carried	O
out	O
in	O
the	O
context	O
of	O
parameterized	O
approximation	O
spaces	O
and	O
a	O
calculus	O
of	O
granules	O
.	O
Judgment	O
in	O
agents	O
is	O
a	O
faculty	O
of	O
thinking	O
about	O
-LRB-	O
classifying	O
-RRB-	O
the	O
particular	O
relative	O
to	O
decision	B-KEY
rules	I-KEY
derived	O
from	O
data	O
.	O
Judgment	O
in	O
agents	O
is	O
reflective	O
,	O
but	O
not	O
in	O
the	O
classical	O
philosophical	O
sense	O
-LRB-	O
e.g.	O
,	O
the	O
notion	O
of	O
judgment	O
in	O
Kant	O
-RRB-	O
.	O
In	O
an	O
agent	O
,	O
a	O
reflective	B-KEY
judgment	I-KEY
itself	O
is	O
an	O
assertion	O
that	O
a	O
particular	O
decision	B-KEY
rule	I-KEY
derived	O
from	O
data	O
is	O
applicable	O
to	O
an	O
object	O
-LRB-	O
input	O
-RRB-	O
.	O
That	O
is	O
,	O
a	O
reflective	B-KEY
judgment	I-KEY
by	O
an	O
agent	O
is	O
an	O
assertion	O
that	O
a	O
particular	O
vector	O
of	O
attribute	O
-LRB-	O
sensor	O
-RRB-	O
values	O
matches	O
to	O
some	O
degree	O
the	O
conditions	O
for	O
a	O
particular	O
rule	O
.	O
In	O
effect	O
,	O
this	O
form	O
of	O
judgment	O
is	O
an	O
assertion	O
that	O
a	O
vector	O
of	O
sensor	O
values	O
reflects	O
a	O
known	O
property	O
of	O
data	O
expressed	O
by	O
a	O
decision	B-KEY
rule	I-KEY
.	O
Since	O
the	O
reasoning	O
underlying	O
a	O
reflective	B-KEY
judgment	I-KEY
is	O
inductive	O
and	O
surjective	O
-LRB-	O
not	O
based	O
on	O
a	O
priori	O
conditions	O
or	O
universals	O
-RRB-	O
,	O
this	O
form	O
of	O
judgment	O
is	O
reflective	O
,	O
but	O
not	O
in	O
the	O
sense	O
of	O
Kant	O
.	O
Unlike	O
Kant	O
,	O
a	O
reflective	B-KEY
judgment	I-KEY
is	O
surjective	O
in	O
the	O
sense	O
that	O
it	O
maps	O
experimental	O
attribute	O
values	O
onto	O
the	O
most	O
closely	O
matching	O
descriptors	O
-LRB-	O
conditions	O
-RRB-	O
in	O
a	O
derived	O
rule	O
.	O
Again	O
,	O
unlike	O
Kant	O
's	O
notion	O
of	O
judgment	O
,	O
a	O
reflective	B-KEY
judgment	I-KEY
is	O
not	O
the	O
result	O
of	O
searching	O
for	O
a	O
universal	O
that	O
pertains	O
to	O
a	O
particular	O
set	O
of	O
values	O
of	O
descriptors	O
.	O
Rather	O
,	O
a	O
reflective	B-KEY
judgment	I-KEY
by	O
an	O
agent	O
is	O
a	O
form	O
of	O
recognition	O
that	O
a	O
particular	O
vector	O
of	O
sensor	O
values	O
pertains	O
to	O
a	O
particular	O
rule	O
in	O
some	O
degree	O
.	O
This	O
recognition	O
takes	O
the	O
form	O
of	O
an	O
assertion	O
that	O
a	O
particular	O
descriptor	O
vector	O
is	O
associated	O
with	O
a	O
particular	O
decision	B-KEY
rule	I-KEY
.	O
These	O
considerations	O
can	O
be	O
repeated	O
for	O
other	O
forms	O
of	O
classifiers	O
besides	O
those	O
defined	O
by	O
decision	B-KEY
rules	I-KEY

In	O
medias	O
res	O
-LSB-	O
DVD	O
formats	O
-RSB-	O
Four	O
years	O
in	O
the	O
making	O
,	O
the	O
DVD	B-KEY
format	I-KEY
war	I-KEY
rages	O
on	O
,	O
no	O
winner	O
insight	O
.	O
meanwhile	O
,	O
the	O
spoils	O
of	O
war	O
abound	O
,	O
and	O
DVD	B-KEY
media	I-KEY
manufacturers	I-KEY
stand	O
poised	O
to	O
profit	O

On	O
abelian	B-KEY
branched	I-KEY
coverings	I-KEY
of	O
the	O
sphere	O
We	O
obtain	O
an	O
enumeration	B-KEY
formula	I-KEY
for	O
the	O
number	O
of	O
weak	B-KEY
equivalence	I-KEY
classes	I-KEY
of	O
the	O
branched	O
-LRB-	O
A	O
*	O
B	O
-RRB-	O
-	O
covering	O
of	O
the	O
sphere	O
with	O
m-branch	O
points	O
,	O
when	O
A	O
and	O
B	O
are	O
finite	B-KEY
abelian	I-KEY
groups	I-KEY
with	O
-LRB-	O
|	O
A	O
|	O
,	O
|	O
B	O
|	O
-RRB-	O
=	O
1	O
.	O
From	O
this	O
,	O
we	O
can	O
deduce	O
an	O
explicit	B-KEY
formula	I-KEY
for	O
enumerating	O
the	O
weak	B-KEY
equivalence	I-KEY
classes	I-KEY
of	O
pseudofree	B-KEY
spherical	I-KEY
-LRB-	O
Zp	O
*	O
Zq	O
-RRB-	O
-	O
actions	O
on	O
a	O
given	O
surface	O
,	O
when	O
p	O
and	O
q	O
are	O
distinct	O
primes	O

Documentum	B-KEY
completes	O
CM	O
Trifecta	O
Daily	O
,	O
people	O
participating	O
in	O
clinical	B-KEY
trials	I-KEY
for	O
drug	B-KEY
companies	I-KEY
fill	O
out	O
forms	O
describing	O
how	O
they	O
feel	O
physically	O
and	O
emotionally	O
.	O
For	O
some	O
trials	O
,	O
there	O
are	O
hundreds	O
,	O
possibly	O
thousands	O
,	O
of	O
participants	O
.	O
The	O
drug	B-KEY
companies	I-KEY
must	O
compile	O
all	O
the	O
forms	O
and	O
submit	O
them	O
electronically	O
to	O
the	O
FDA	B-KEY
.	O
That	O
's	O
where	O
Documentum	B-KEY
comes	O
in	O
.	O
``	O
We	O
've	O
streamlined	O
the	O
whole	O
process	O
of	O
managing	O
clinical	B-KEY
trial	I-KEY
content	O
for	O
companies	O
,	O
such	O
as	O
Johnson	O
&	O
Johnson	O
,	O
Bristol	O
Myers	O
Squibb	O
,	O
and	O
Pfizer	O
,	O
''	O
notes	O
Documentum	O
's	O
president	O
and	O
CEO	O
Dave	O
De	O
Walt	O
.	O
``	O
And	O
by	O
the	O
way	O
,	O
the	O
FDA	B-KEY
also	O
is	O
one	O
of	O
our	O
customers	O
,	O
as	O
well	O
as	O
the	O
EPA	O
and	O
the	O
FAA	O
.	O
''	O
And	O
there	O
are	O
about	O
1,300	O
other	O
organizations	O
in	O
various	O
industries	O
worldwide	O
that	O
rely	O
on	O
Documentum	B-KEY
's	O
technologies	O
,	O
consulting	O
,	O
and	O
training	B-KEY
services	I-KEY
.	O
The	O
company	O
's	O
products	O
are	O
designed	O
to	O
manage	O
digital	O
content	O
and	O
facilitate	O
online	O
transactions	O
,	O
partner	O
and	O
supplier	O
relationships	O
,	O
and	O
ebusiness	O
interactions	O

An	O
improved	O
self-organizing	O
CPN-based	O
fuzzy	O
system	O
with	O
adaptive	O
back-propagation	O
algorithm	O
This	O
paper	O
describes	O
an	O
improved	O
self-organizing	O
CPN-based	O
-LRB-	O
Counter-Propagation	B-KEY
Network	I-KEY
-RRB-	O
fuzzy	O
system	O
.	O
Two	O
self-organizing	O
algorithms	O
IUSOCPN	O
and	O
ISSOCPN	O
,	O
being	O
unsupervised	O
and	O
supervised	O
respectively	O
,	O
are	O
introduced	O
.	O
The	O
idea	O
is	O
to	O
construct	O
the	O
neural-fuzzy	B-KEY
system	I-KEY
with	O
a	O
two-phase	O
hybrid	B-KEY
learning	I-KEY
algorithm	O
,	O
which	O
utilizes	O
a	O
CPN-based	O
nearest-neighbor	O
clustering	O
scheme	O
for	O
both	O
structure	B-KEY
learning	I-KEY
and	O
initial	B-KEY
parameters	I-KEY
setting	I-KEY
,	O
and	O
a	O
gradient	B-KEY
descent	I-KEY
method	O
with	O
adaptive	O
learning	O
rate	O
for	O
fine	O
tuning	O
the	O
parameters	O
.	O
The	O
obtained	O
network	O
can	O
be	O
used	O
in	O
the	O
same	O
way	O
as	O
a	O
CPN	O
to	O
model	O
and	O
control	O
dynamic	O
systems	O
,	O
while	O
it	O
has	O
a	O
faster	O
learning	O
speed	O
than	O
the	O
original	O
back-propagation	O
algorithm	O
.	O
The	O
comparative	O
results	O
on	O
the	O
examples	O
suggest	O
that	O
the	O
method	O
is	O
fairly	O
efficient	O
in	O
terms	O
of	O
simple	O
structure	O
,	O
fast	O
learning	O
speed	O
,	O
and	O
relatively	O
high	O
modeling	O
accuracy	O

Parallel	O
and	O
distributed	B-KEY
Haskells	I-KEY
Parallel	O
and	O
distributed	B-KEY
languages	I-KEY
specify	O
computations	O
on	O
multiple	B-KEY
processors	I-KEY
and	O
have	O
a	O
computation	O
language	O
to	O
describe	O
the	O
algorithm	O
,	O
i.e.	O
what	O
to	O
compute	O
,	O
and	O
a	O
coordination	B-KEY
language	I-KEY
to	O
describe	O
how	O
to	O
organise	O
the	O
computations	O
across	O
the	O
processors	O
.	O
Haskell	O
has	O
been	O
used	O
as	O
the	O
computation	O
language	O
for	O
a	O
wide	O
variety	O
of	O
parallel	O
and	O
distributed	B-KEY
languages	I-KEY
,	O
and	O
this	O
paper	O
is	O
a	O
comprehensive	O
survey	O
of	O
implemented	O
languages	O
.	O
It	O
outlines	O
parallel	O
and	O
distributed	B-KEY
language	I-KEY
concepts	O
and	O
classifies	O
Haskell	O
extensions	O
using	O
them	O
.	O
Similar	O
example	O
programs	O
are	O
used	O
to	O
illustrate	O
and	O
contrast	O
the	O
coordination	B-KEY
languages	I-KEY
,	O
and	O
the	O
comparison	O
is	O
facilitated	O
by	O
the	O
common	O
computation	O
language	O
.	O
A	O
lazy	B-KEY
language	I-KEY
is	O
not	O
an	O
obvious	O
choice	O
for	O
parallel	O
or	O
distributed	O
computation	O
,	O
and	O
we	O
address	O
the	O
question	O
of	O
why	O
Haskell	O
is	O
a	O
common	O
functional	B-KEY
computation	I-KEY
language	I-KEY

Modeling	O
cutting	O
temperatures	O
for	O
turning	B-KEY
inserts	I-KEY
with	O
various	O
tool	B-KEY
geometries	I-KEY
and	O
materials	O
Temperatures	O
are	O
of	O
interest	O
in	O
machining	B-KEY
because	O
cutting	O
tools	O
often	O
fail	O
by	O
thermal	O
softening	O
or	O
temperature-activated	O
wear	O
.	O
Many	O
models	O
for	O
cutting	O
temperatures	O
have	O
been	O
developed	O
,	O
but	O
these	O
models	O
consider	O
only	O
simple	O
tool	B-KEY
geometries	I-KEY
such	O
as	O
a	O
rectangular	O
slab	O
with	O
a	O
sharp	O
corner	O
.	O
This	O
report	O
describes	O
a	O
finite	O
element	O
study	O
of	O
tool	O
temperatures	O
in	O
cutting	O
that	O
accounts	O
for	O
tool	B-KEY
nose	I-KEY
radius	I-KEY
and	O
included	O
angle	O
effects	O
.	O
A	O
temperature	B-KEY
correction	I-KEY
factor	I-KEY
model	O
that	O
can	O
be	O
used	O
in	O
the	O
design	O
and	O
selection	O
of	O
inserts	O
is	O
developed	O
to	O
account	O
for	O
these	O
effects	O
.	O
A	O
parametric	B-KEY
mesh	I-KEY
generator	I-KEY
is	O
used	O
to	O
generate	O
the	O
finite	B-KEY
element	I-KEY
models	I-KEY
of	O
tool	O
and	O
inserts	O
of	O
varying	O
geometries	O
.	O
The	O
steady-state	O
temperature	O
response	O
is	O
calculated	O
using	O
NASTRAN	O
solver	O
.	O
Several	O
finite	O
element	O
analysis	O
-LRB-	O
FEA	O
-RRB-	O
runs	O
are	O
performed	O
to	O
quantify	O
the	O
effects	O
of	O
inserts	O
included	O
angle	O
,	O
nose	O
radius	O
,	O
and	O
materials	O
for	O
the	O
insert	O
and	O
the	O
tool	O
holder	O
on	O
the	O
cutting	O
temperature	O
at	O
the	O
insert	O
rake	O
face	O
.	O
The	O
FEA	O
results	O
are	O
then	O
utilized	O
to	O
develop	O
a	O
temperature	B-KEY
correction	I-KEY
factor	I-KEY
model	O
that	O
accounts	O
for	O
these	O
effects	O
.	O
The	O
temperature	B-KEY
correction	I-KEY
factor	I-KEY
model	O
is	O
integrated	O
with	O
an	O
analytical	O
temperature	O
model	O
for	O
rectangular	O
inserts	O
to	O
predict	O
cutting	O
temperatures	O
for	O
contour	O
turning	O
with	O
inserts	O
of	O
various	O
shapes	O
and	O
nose	O
radii	O
.	O
Finally	O
,	O
experimental	O
measurements	O
of	O
cutting	O
temperature	O
using	O
the	O
tool-work	O
thermocouple	O
technique	O
are	O
performed	O
and	O
compared	O
with	O
the	O
predictions	O
of	O
the	O
new	O
temperature	O
model	O
.	O
The	O
comparisons	O
show	O
good	O
agreement	O

Optical	O
actuation	O
of	O
a	O
bistable	B-KEY
MEMS	I-KEY
This	O
paper	O
presents	O
an	O
optical	B-KEY
actuation	I-KEY
scheme	I-KEY
for	O
MEMS	B-KEY
devices	I-KEY
based	O
on	O
the	O
well-established	O
fact	O
that	O
light	O
possesses	O
momentum	O
,	O
and	O
hence	O
,	O
imparts	O
a	O
force	O
equal	O
to	O
2	O
W/c	O
when	O
reflected	O
by	O
a	O
surface	O
.	O
Here	O
,	O
W	O
is	O
the	O
total	O
power	O
of	O
the	O
reflected	O
light	O
,	O
and	O
c	O
is	O
the	O
speed	O
of	O
light	O
.	O
Radiation	B-KEY
pressure	I-KEY
,	O
as	O
it	O
is	O
known	O
,	O
is	O
nearly	O
insignificant	O
for	O
most	O
macroscale	O
applications	O
,	O
but	O
it	O
can	O
be	O
quite	O
significant	O
for	O
MEMS	B-KEY
devices	I-KEY
.	O
In	O
addition	O
,	O
light	O
actuation	O
offers	O
a	O
new	O
paradigm	O
.	O
First	O
,	O
intersecting	B-KEY
light	I-KEY
beams	I-KEY
do	O
not	O
interfere	O
,	O
in	O
contrast	O
to	O
electrical	O
conductors	O
,	O
which	O
short	O
when	O
they	O
come	O
into	O
contact	O
.	O
Second	O
,	O
light	O
can	O
operate	O
in	O
high	O
temperature	O
and	O
high	B-KEY
radiation	I-KEY
environments	I-KEY
far	O
outside	O
the	O
capability	O
of	O
solid	O
state	O
electronic	O
components	O
.	O
This	O
actuation	O
method	O
is	O
demonstrated	O
,	O
both	O
in	O
air	O
and	O
in	O
vacuum	O
,	O
by	O
switching	O
the	O
state	O
of	O
a	O
bistable	B-KEY
MEMS	I-KEY
device	O
.	O
The	O
associated	O
heat	B-KEY
transfer	I-KEY
model	I-KEY
is	O
also	O
presented	O

Estimation	O
of	O
blocking	O
probabilities	O
in	O
cellular	O
networks	O
with	O
dynamic	B-KEY
channel	I-KEY
assignment	I-KEY
Blocking	O
probabilities	O
in	O
cellular	B-KEY
mobile	I-KEY
communication	I-KEY
networks	I-KEY
using	O
dynamic	B-KEY
channel	I-KEY
assignment	I-KEY
are	O
hard	O
to	O
compute	O
for	O
realistic	O
sized	O
systems	O
.	O
This	O
computational	O
difficulty	O
is	O
due	O
to	O
the	O
structure	O
of	O
the	O
state	O
space	O
,	O
which	O
imposes	O
strong	B-KEY
coupling	I-KEY
constraints	I-KEY
amongst	O
components	O
of	O
the	O
occupancy	B-KEY
vector	I-KEY
.	O
Approximate	B-KEY
tractable	I-KEY
models	I-KEY
have	O
been	O
proposed	O
,	O
which	O
have	O
product	B-KEY
form	I-KEY
stationary	I-KEY
state	I-KEY
distributions	I-KEY
.	O
However	O
,	O
for	O
real	O
channel	O
assignment	O
schemes	O
,	O
the	O
product	O
form	O
is	O
a	O
poor	O
approximation	O
and	O
it	O
is	O
necessary	O
to	O
simulate	B-KEY
the	O
actual	O
occupancy	O
process	O
in	O
order	O
to	O
estimate	O
the	O
blocking	O
probabilities	O
.	O
Meaningful	O
estimates	O
of	O
the	O
blocking	O
probability	O
typically	O
require	O
an	O
enormous	O
amount	O
of	O
CPU	B-KEY
time	I-KEY
for	O
simulation	B-KEY
,	O
since	O
blocking	O
events	O
are	O
usually	O
rare	O
.	O
Advanced	O
simulation	B-KEY
approaches	O
use	O
importance	B-KEY
sampling	I-KEY
-LRB-	O
IS	O
-RRB-	O
to	O
overcome	O
this	O
problem	O
.	O
We	O
study	O
two	O
regimes	O
under	O
which	O
blocking	O
is	O
a	O
rare	O
event	O
:	O
low-load	B-KEY
and	O
high	B-KEY
cell	I-KEY
capacity	I-KEY
.	O
Our	O
simulations	B-KEY
use	O
the	O
standard	O
clock	O
-LRB-	O
SC	O
-RRB-	O
method	O
.	O
For	O
low	O
load	O
,	O
we	O
propose	O
a	O
change	O
of	O
measure	O
that	O
we	O
call	O
static	O
ISSC	O
,	O
which	O
has	O
bounded	B-KEY
relative	I-KEY
error	I-KEY
.	O
For	O
high	O
capacity	O
,	O
we	O
use	O
a	O
change	O
of	O
measure	O
that	O
depends	O
on	O
the	O
current	O
state	O
of	O
the	O
network	O
occupancy	O
.	O
This	O
is	O
the	O
dynamic	B-KEY
ISSC	I-KEY
method	I-KEY
.	O
We	O
prove	O
that	O
this	O
method	O
yields	O
zero	B-KEY
variance	I-KEY
estimators	I-KEY
for	O
single	B-KEY
clique	I-KEY
models	I-KEY
,	O
and	O
we	O
empirically	O
show	O
the	O
advantages	O
of	O
this	O
method	O
over	O
naive	O
simulation	B-KEY
for	O
networks	O
of	O
moderate	O
size	O
and	O
traffic	O
loads	O

Perceptual	B-KEY
audio	I-KEY
coding	I-KEY
using	O
adaptive	O
pre	O
-	O
and	O
post-filters	O
and	O
lossless	B-KEY
compression	I-KEY
This	O
paper	O
proposes	O
a	O
versatile	O
perceptual	B-KEY
audio	I-KEY
coding	I-KEY
method	O
that	O
achieves	O
high	B-KEY
compression	I-KEY
ratios	I-KEY
and	O
is	O
capable	O
of	O
low	B-KEY
encoding/decoding	I-KEY
delay	I-KEY
.	O
It	O
accommodates	O
a	O
variety	O
of	O
source	B-KEY
signals	I-KEY
-LRB-	O
including	O
both	O
music	B-KEY
and	O
speech	O
-RRB-	O
with	O
different	O
sampling	B-KEY
rates	I-KEY
.	O
It	O
is	O
based	O
on	O
separating	O
irrelevance	O
and	O
redundancy	B-KEY
reductions	I-KEY
into	O
independent	O
functional	O
units	O
.	O
This	O
contrasts	O
traditional	O
audio	O
coding	O
where	O
both	O
are	O
integrated	O
within	O
the	O
same	O
subband	O
decomposition	O
.	O
The	O
separation	O
allows	O
for	O
the	O
independent	O
optimization	O
of	O
the	O
irrelevance	O
and	O
redundancy	B-KEY
reduction	I-KEY
units	O
.	O
For	O
both	O
reductions	O
,	O
we	O
rely	O
on	O
adaptive	B-KEY
filtering	I-KEY
and	O
predictive	B-KEY
coding	I-KEY
as	O
much	O
as	O
possible	O
to	O
minimize	O
the	O
delay	O
.	O
A	O
psycho-acoustically	B-KEY
controlled	I-KEY
adaptive	I-KEY
linear	I-KEY
filter	I-KEY
is	O
used	O
for	O
the	O
irrelevance	B-KEY
reduction	I-KEY
,	O
and	O
the	O
redundancy	B-KEY
reduction	I-KEY
is	O
carried	O
out	O
by	O
a	O
predictive	B-KEY
lossless	I-KEY
coding	I-KEY
scheme	O
,	O
which	O
is	O
termed	O
weighted	B-KEY
cascaded	I-KEY
least	I-KEY
mean	I-KEY
squared	I-KEY
-LRB-	O
WCLMS	O
-RRB-	O
method	O
.	O
Experiments	O
are	O
carried	O
out	O
on	O
a	O
database	O
of	O
moderate	O
size	O
which	O
contains	O
mono-signals	O
of	O
different	O
sampling	B-KEY
rates	I-KEY
and	O
varying	O
nature	O
-LRB-	O
music	B-KEY
,	O
speech	O
,	O
or	O
mixed	O
-RRB-	O
.	O
They	O
show	O
that	O
the	O
proposed	O
WCLMS	B-KEY
lossless	I-KEY
coder	I-KEY
outperforms	O
other	O
competing	O
lossless	O
coders	O
in	O
terms	O
of	O
compression	O
ratios	O
and	O
delay	O
,	O
as	O
applied	O
to	O
the	O
pre-filtered	O
signal	O
.	O
Moreover	O
,	O
a	O
subjective	B-KEY
listening	I-KEY
test	I-KEY
of	O
the	O
combined	O
pre-filter/lossless	B-KEY
coder	I-KEY
and	O
a	O
state-of-the-art	O
perceptual	O
audio	O
coder	O
-LRB-	O
PAC	O
-RRB-	O
shows	O
that	O
the	O
new	O
method	O
achieves	O
a	O
comparable	O
compression	O
ratio	O
and	O
audio	B-KEY
quality	I-KEY
with	O
a	O
lower	O
delay	O

Formalization	O
of	O
weighted	B-KEY
factors	I-KEY
analysis	I-KEY
Weighted	B-KEY
factors	I-KEY
analysis	I-KEY
-LRB-	O
WeFA	O
-RRB-	O
has	O
been	O
proposed	O
as	O
a	O
new	O
approach	O
for	O
elicitation	O
,	O
representation	O
,	O
and	O
manipulation	O
of	O
knowledge	O
about	O
a	O
given	O
problem	O
,	O
generally	O
at	O
a	O
high	O
and	O
strategic	O
level	O
.	O
Central	O
to	O
this	O
proposal	O
is	O
that	O
a	O
group	O
of	O
experts	O
in	O
the	O
area	O
of	O
the	O
problem	O
can	O
identify	O
a	O
hierarchy	O
of	O
factors	O
with	O
positive	O
or	O
negative	O
influences	O
on	O
the	O
problem	O
outcome	O
.	O
The	O
tangible	O
output	O
of	O
WeFA	O
is	O
a	O
directed	B-KEY
weighted	I-KEY
graph	I-KEY
called	O
a	O
WeFA	B-KEY
graph	I-KEY
.	O
This	O
is	O
a	O
set	O
of	O
nodes	O
denoting	O
factors	O
that	O
can	O
directly	O
or	O
indirectly	O
influence	O
an	O
overall	O
aim	O
of	O
the	O
graph	O
.	O
The	O
aim	O
is	O
also	O
represented	O
by	O
a	O
node	O
.	O
Each	O
directed	B-KEY
arc	I-KEY
is	O
a	O
direct	O
influence	O
of	O
one	O
factor	O
on	O
another	O
.	O
A	O
chain	O
of	O
directed	B-KEY
arcs	I-KEY
indicates	O
an	O
indirect	O
influence	O
.	O
The	O
influences	O
may	O
be	O
identified	O
as	O
either	O
positive	O
or	O
negative	O
.	O
For	O
example	O
,	O
sales	O
and	O
costs	O
are	O
two	O
factors	O
that	O
influence	O
the	O
aim	O
of	O
profitability	B-KEY
in	O
an	O
organization	B-KEY
.	O
Sales	O
has	O
a	O
positive	O
influence	O
on	O
profitability	B-KEY
and	O
costs	O
has	O
a	O
negative	O
influence	O
on	O
profitability	B-KEY
.	O
In	O
addition	O
,	O
the	O
relative	O
significance	B-KEY
of	O
each	O
influence	O
is	O
represented	O
by	O
a	O
weight	O
.	O
We	O
develop	O
Binary	B-KEY
WeFA	I-KEY
which	O
is	O
a	O
variant	O
of	O
WeFA	O
where	O
the	O
factors	O
in	O
the	O
graph	O
are	O
restricted	O
to	O
being	O
either	O
true	O
or	O
false	O
.	O
Imposing	O
this	O
restriction	O
on	O
a	O
WeFA	B-KEY
graph	I-KEY
allows	O
us	O
to	O
be	O
more	O
precise	O
about	O
the	O
meaning	O
of	O
the	O
graph	O
and	O
of	O
reasoning	B-KEY
in	O
it	O
.	O
Binary	B-KEY
WeFA	I-KEY
is	O
a	O
new	O
proposal	O
that	O
provides	O
a	O
formal	O
yet	O
sufficiently	O
simple	O
language	O
for	O
logic-based	B-KEY
argumentation	I-KEY
for	O
use	O
by	O
business	O
people	O
in	O
decision-support	B-KEY
and	O
knowledge	B-KEY
management	I-KEY
.	O
Whilst	O
Binary	B-KEY
WeFA	I-KEY
is	O
expressively	O
simpler	O
than	O
other	O
logic-based	B-KEY
argumentation	I-KEY
formalisms	O
,	O
it	O
does	O
incorporate	O
a	O
novel	O
formalization	O
of	O
the	O
notion	O
of	O
significance	B-KEY

Multiecho	B-KEY
segmented	I-KEY
EPI	I-KEY
with	O
z-shimmed	B-KEY
background	I-KEY
gradient	I-KEY
compensation	I-KEY
-LRB-	O
MESBAC	O
-RRB-	O
pulse	O
sequence	O
for	O
fMRI	B-KEY
A	O
MultiEcho	B-KEY
Segmented	I-KEY
EPI	I-KEY
with	O
z-shimmed	B-KEY
BAckground	I-KEY
gradient	I-KEY
Compensation	I-KEY
-LRB-	O
MESBAC	O
-RRB-	O
pulse	O
sequence	O
is	O
proposed	O
and	O
validated	O
for	O
functional	O
MRI	O
-LRB-	O
fMRI	B-KEY
-RRB-	O
study	O
in	O
regions	O
suffering	O
from	O
severe	B-KEY
susceptibility	I-KEY
artifacts	I-KEY
.	O
This	O
sequence	O
provides	O
an	O
effective	O
tradeoff	O
between	O
spatial	O
and	O
temporal	B-KEY
resolution	I-KEY
and	O
reduces	O
image	B-KEY
distortion	I-KEY
and	O
signal	B-KEY
dropout	I-KEY
.	O
The	O
blood	O
oxygenation	O
level-dependent	O
-LRB-	O
BOLD	O
-RRB-	O
-	O
weighted	O
fMRI	B-KEY
signal	O
can	O
be	O
reliably	O
obtained	O
in	O
the	O
region	O
of	O
the	O
orbitofrontal	B-KEY
cortex	I-KEY
-LRB-	O
OFC	O
-RRB-	O
.	O
To	O
overcome	O
physiological	O
motion	O
artifacts	O
during	O
prolonged	O
multisegment	O
EPI	O
acquisition	O
,	O
two	O
sets	O
of	O
navigator	B-KEY
echoes	I-KEY
were	O
acquired	O
in	O
both	O
the	O
readout	O
and	O
phase-encoding	O
directions	O
.	O
Ghost	B-KEY
artifacts	I-KEY
generally	O
produced	O
by	O
single-shot	O
EPI	O
acquisition	O
were	O
eliminated	O
by	O
separately	O
placing	O
the	O
even	O
and	O
odd	O
echoes	O
in	O
different	O
k-space	O
trajectories	O
.	O
Unlike	O
most	O
z-shim	O
methods	O
that	O
focus	O
on	O
increasing	O
temporal	B-KEY
resolution	I-KEY
for	O
event-related	B-KEY
functional	I-KEY
brain	I-KEY
mapping	I-KEY
,	O
the	O
MESBAC	O
sequence	O
simultaneously	O
addresses	O
problems	O
of	O
image	B-KEY
distortion	I-KEY
and	O
signal	B-KEY
dropout	I-KEY
while	O
maintaining	O
sufficient	O
temporal	B-KEY
resolution	I-KEY
.	O
The	O
MESBAC	O
sequence	O
will	O
be	O
particularly	O
useful	O
for	O
pharmacological	O
and	O
affective	O
fMRI	B-KEY
studies	O
in	O
brain	O
regions	O
such	O
as	O
the	O
OFC	O
,	O
nucleus	O
accumbens	O
,	O
amygdala	O
,	O
para-hippocampus	O
,	O
etc	O

Adaptive	B-KEY
scheduling	I-KEY
of	O
batch	B-KEY
servers	I-KEY
in	O
flow	B-KEY
shops	I-KEY
Batch	B-KEY
servicing	I-KEY
is	O
a	O
common	O
way	O
of	O
benefiting	O
from	O
economies	O
of	O
scale	O
in	O
manufacturing	B-KEY
operations	I-KEY
.	O
Good	O
examples	O
of	O
production	B-KEY
systems	I-KEY
that	O
allow	O
for	O
batch	O
processing	O
are	O
ovens	B-KEY
found	O
in	O
the	O
aircraft	B-KEY
industry	I-KEY
and	O
in	O
semiconductor	B-KEY
manufacturing	I-KEY
.	O
In	O
this	O
paper	O
we	O
study	O
the	O
issue	O
of	O
dynamic	B-KEY
scheduling	I-KEY
of	O
such	O
systems	O
within	O
the	O
context	O
of	O
multi-stage	O
flow	B-KEY
shops	I-KEY
.	O
So	O
far	O
,	O
a	O
great	O
deal	O
of	O
research	O
has	O
concentrated	O
on	O
the	O
development	O
of	O
control	B-KEY
strategies	I-KEY
,	O
which	O
only	O
address	O
the	O
batch	O
stage	O
.	O
This	O
paper	O
proposes	O
an	O
integral	B-KEY
scheduling	I-KEY
approach	I-KEY
that	O
also	O
includes	O
succeeding	O
stages	O
.	O
In	O
this	O
way	O
,	O
we	O
aim	O
for	O
shop	B-KEY
optimization	I-KEY
,	O
instead	O
of	O
optimizing	O
performance	O
for	O
a	O
single	O
stage	O
.	O
Our	O
so-called	O
look-ahead	B-KEY
strategy	I-KEY
adapts	O
its	O
scheduling	O
decision	O
to	O
shop	O
status	O
,	O
which	O
includes	O
information	O
on	O
a	O
limited	O
number	O
of	O
near-future	B-KEY
arrivals	I-KEY
.	O
In	O
particular	O
,	O
we	O
study	O
a	O
two-stage	O
flow	B-KEY
shop	I-KEY
,	O
in	O
which	O
the	O
batch	O
stage	O
is	O
succeeded	O
by	O
a	O
serial	O
stage	O
.	O
The	O
serial	O
stage	O
may	O
be	O
realized	O
by	O
a	O
single	B-KEY
machine	I-KEY
or	O
by	O
parallel	B-KEY
machines	I-KEY
.	O
Through	O
an	O
extensive	O
simulation	B-KEY
study	I-KEY
it	O
is	O
demonstrated	O
how	O
shop	O
performance	O
can	O
be	O
improved	O
by	O
the	O
proposed	O
strategy	O
relative	O
to	O
existing	O
strategies	O

Multicell	B-KEY
converters	I-KEY
:	O
active	B-KEY
control	I-KEY
and	O
observation	O
of	O
flying-capacitor	B-KEY
voltages	I-KEY
The	O
multicell	B-KEY
converters	I-KEY
introduced	O
more	O
than	O
ten	O
years	O
ago	O
make	O
it	O
possible	O
to	O
distribute	O
the	O
voltage	O
constraints	O
among	O
series-connected	B-KEY
switches	I-KEY
and	O
to	O
improve	O
the	O
output	O
waveforms	O
-LRB-	O
increased	O
number	O
of	O
levels	O
and	O
apparent	O
frequency	O
-RRB-	O
.	O
The	O
balance	O
of	O
the	O
constraints	O
requires	O
an	O
appropriate	O
distribution	O
of	O
the	O
flying	O
voltages	O
.	O
This	O
paper	O
presents	O
some	O
solutions	O
for	O
the	O
active	B-KEY
control	I-KEY
of	O
the	O
voltages	O
across	O
the	O
flying	O
capacitors	O
in	O
the	O
presence	O
of	O
rapid	O
variation	O
of	O
the	O
input	B-KEY
voltage	I-KEY
.	O
The	O
latter	O
part	O
of	O
this	O
paper	O
is	O
dedicated	O
to	O
the	O
observation	O
of	O
these	O
voltages	O
using	O
an	O
original	O
modeling	O
of	O
the	O
converter	O

Approximation	O
theory	O
of	O
fuzzy	B-KEY
systems	I-KEY
based	O
upon	O
genuine	O
many-valued	B-KEY
implications	I-KEY
-	O
MIMO	O
cases	O
It	O
is	O
constructively	O
proved	O
that	O
the	O
multi-input-multi-output	O
fuzzy	B-KEY
systems	I-KEY
based	O
upon	O
genuine	O
many-valued	O
implications	O
are	O
universal	O
approximators	O
-LRB-	O
they	O
are	O
called	O
Boolean	O
type	O
fuzzy	O
systems	O
in	O
this	O
paper	O
-RRB-	O
.	O
The	O
general	O
approach	O
to	O
construct	O
such	O
fuzzy	B-KEY
systems	I-KEY
is	O
given	O
,	O
that	O
is	O
,	O
through	O
the	O
partition	O
of	O
the	O
output	O
region	O
-LRB-	O
by	O
the	O
given	O
accuracy	O
-RRB-	O
.	O
Two	O
examples	O
are	O
provided	O
to	O
demonstrate	O
the	O
way	O
in	O
which	O
fuzzy	B-KEY
systems	I-KEY
are	O
designed	O
to	O
approximate	O
given	O
functions	O
with	O
a	O
given	O
required	O
approximation	O
accuracy	O

Stabilization	O
of	O
global	B-KEY
invariant	I-KEY
sets	I-KEY
for	O
chaotic	O
systems	O
:	O
an	O
energy	O
based	O
control	O
approach	O
This	O
paper	O
presents	O
a	O
new	O
control	O
approach	O
for	O
steering	O
trajectories	O
of	O
three-dimensional	B-KEY
nonlinear	I-KEY
chaotic	I-KEY
systems	I-KEY
towards	O
stable	B-KEY
stationary	I-KEY
states	I-KEY
or	O
time-periodic	B-KEY
orbits	I-KEY
.	O
The	O
proposed	O
method	O
mainly	O
consists	O
in	O
a	O
sliding	B-KEY
mode-based	I-KEY
control	I-KEY
design	O
that	O
is	O
extended	O
by	O
an	O
explicit	O
consideration	O
of	O
system	O
energy	O
as	O
basis	O
for	O
both	O
controller	O
design	O
and	O
system	O
stabilization	O
.	O
The	O
control	O
objective	O
is	O
then	O
to	O
regulate	O
the	O
energy	O
with	O
respect	O
to	O
a	O
shaped	O
nominal	O
representation	O
implicitly	O
related	O
to	O
system	O
trajectories	O
.	O
In	O
this	O
paper	O
,	O
we	O
establish	O
some	O
theoretical	O
results	O
to	O
introduce	O
the	O
control	O
design	O
approach	O
referred	O
to	O
as	O
energy	B-KEY
based	I-KEY
sliding	I-KEY
mode	I-KEY
control	I-KEY
.	O
Then	O
,	O
some	O
capabilities	O
of	O
the	O
proposed	O
approach	O
are	O
illustrated	O
through	O
examples	O
related	O
to	O
the	O
chaotic	O
circuit	O
of	O
Chua	O

Radianz	B-KEY
and	O
Savvis	B-KEY
look	O
to	O
expand	O
service	O
in	O
wake	O
of	O
telecom	O
scandals	O
-LSB-	O
finance	O
-RSB-	O
With	O
confidence	O
in	O
network	B-KEY
providers	I-KEY
waning	O
,	O
Radianz	B-KEY
and	O
Savvis	B-KEY
try	O
to	O
prove	O
their	O
stability	O
.	O
Savvis	B-KEY
and	O
Radianz	B-KEY
,	O
which	O
both	O
specialize	O
in	O
providing	O
the	O
data-extranet	B-KEY
components	O
of	O
telecommunication	B-KEY
infrastructures	I-KEY
,	O
may	O
see	O
more	O
networking	O
doors	O
open	O
at	O
investment	B-KEY
banks	I-KEY
,	O
brokerage	B-KEY
houses	I-KEY
,	O
exchanges	B-KEY
and	O
alternative-trading	B-KEY
systems	I-KEY

Reaching	O
strong	B-KEY
consensus	I-KEY
in	O
a	O
general	O
network	O
The	O
strong	B-KEY
consensus	I-KEY
-LRB-	O
SC	O
-RRB-	O
problem	O
is	O
a	O
variant	O
of	O
the	O
conventional	O
distributed	B-KEY
consensus	I-KEY
problem	I-KEY
-LRB-	O
also	O
known	O
as	O
the	O
Byzantine	B-KEY
agreement	I-KEY
problem	O
-RRB-	O
.	O
The	O
SC	O
problem	O
requires	O
that	O
the	O
agreed	O
value	O
among	O
fault-free	B-KEY
processors	I-KEY
be	O
one	O
of	O
the	O
fault-free	B-KEY
processor	I-KEY
's	O
initial	O
values	O
.	O
Originally	O
,	O
the	O
problem	O
was	O
studied	O
in	O
a	O
fully	B-KEY
connected	I-KEY
network	I-KEY
with	O
malicious	O
faulty	O
processors	O
.	O
In	O
this	O
paper	O
,	O
the	O
SC	O
problem	O
is	O
re-examined	O
in	O
a	O
general	O
network	O
,	O
in	O
which	O
the	O
components	O
-LRB-	O
processors	O
and	O
communication	O
links	O
-RRB-	O
may	O
be	O
subjected	O
to	O
different	O
faulty	O
types	O
simultaneously	O
-LRB-	O
also	O
called	O
the	O
hybrid	B-KEY
fault	I-KEY
model	I-KEY
or	O
mixed	O
faulty	O
types	O
-RRB-	O
and	O
the	O
network	O
topology	O
does	O
not	O
have	O
to	O
be	O
fully	O
connected	O
.	O
The	O
proposed	O
protocol	O
can	O
tolerate	O
the	O
maximum	O
number	O
of	O
tolerable	O
faulty	O
components	O
such	O
that	O
each	O
fault-free	B-KEY
processor	I-KEY
obtains	O
a	O
common	O
value	O
for	O
the	O
SC	O
problem	O
in	O
a	O
general	O
network	O

Proof	O
that	O
the	O
election	B-KEY
problem	I-KEY
belongs	O
to	O
NF-completeness	B-KEY
problems	I-KEY
in	O
asynchronous	B-KEY
distributed	I-KEY
systems	I-KEY
This	O
paper	O
is	O
about	O
the	O
hardness	O
of	O
the	O
election	B-KEY
problem	I-KEY
in	O
asynchronous	B-KEY
distributed	I-KEY
systems	I-KEY
in	O
which	O
processes	O
can	O
crash	O
but	O
links	O
are	O
reliable	O
.	O
The	O
hardness	O
of	O
the	O
problem	O
is	O
defined	O
with	O
respect	O
to	O
the	O
difficulty	O
to	O
solve	O
it	O
despite	O
failures	O
.	O
It	O
is	O
shown	O
that	O
problems	O
encountered	O
in	O
the	O
system	O
are	O
classified	O
as	O
three	O
classes	O
of	O
problems	O
:	O
F	O
-LRB-	O
fault-tolerant	O
-RRB-	O
,	O
NF	O
-LRB-	O
Not	O
fault-tolerant	O
-RRB-	O
and	O
NFC	O
-LRB-	O
NF-completeness	O
-RRB-	O
.	O
Among	O
those	O
,	O
the	O
class	O
NFC	O
are	O
the	O
hardest	O
problems	O
to	O
solve	O
.	O
In	O
this	O
paper	O
,	O
we	O
prove	O
that	O
the	O
Election	B-KEY
problem	I-KEY
is	O
the	O
most	O
difficult	O
problem	O
which	O
belongs	O
to	O
the	O
class	O
NFC	O

Swamped	O
by	O
data	O
-LSB-	O
storage	O
-RSB-	O
While	O
the	O
cost	B-KEY
of	I-KEY
storage	I-KEY
has	O
plummeted	O
,	O
the	O
demand	O
continued	O
to	O
climb	O
and	O
there	O
are	O
plenty	O
of	O
players	O
out	O
there	O
offering	O
solutions	O
to	O
a	O
company	O
's	O
burgeoning	O
storage	O
needs	O

The	O
web	B-KEY
services	I-KEY
agenda	O
Even	O
the	O
most	O
battle-scarred	O
of	O
CIOs	O
have	O
become	O
excited	O
at	O
the	O
prospect	O
of	O
what	O
web	B-KEY
services	I-KEY
can	O
do	O
for	O
their	O
businesses	O
.	O
But	O
there	O
are	O
still	O
some	O
shortcomings	O
to	O
be	O
addressed	O

Techniques	O
for	O
compiling	B-KEY
and	O
implementing	O
all	O
NAS	B-KEY
parallel	I-KEY
benchmarks	I-KEY
in	O
HPF	O
The	O
NAS	B-KEY
parallel	I-KEY
benchmarks	I-KEY
-LRB-	O
NPB	O
-RRB-	O
are	O
a	O
well-known	O
benchmark	O
set	O
for	O
high-performance	B-KEY
machines	I-KEY
.	O
Much	O
effort	O
has	O
been	O
made	O
to	O
implement	O
them	O
in	O
High-Performance	O
Fortran	O
-LRB-	O
HPF	O
-RRB-	O
.	O
In	O
previous	O
attempts	O
,	O
however	O
,	O
the	O
HPF	O
versions	O
did	O
not	O
include	O
the	O
complete	O
set	O
of	O
benchmarks	O
,	O
and	O
the	O
performance	O
was	O
not	O
always	O
good	O
.	O
In	O
this	O
study	O
,	O
we	O
implement	O
all	O
eight	O
benchmarks	O
of	O
the	O
NPB	O
in	O
HPF	O
,	O
and	O
parallelize	O
them	O
using	O
an	O
HPF	B-KEY
compiler	I-KEY
that	O
we	O
have	O
developed	O
.	O
This	O
report	O
describes	O
the	O
implementation	O
techniques	O
and	O
compiler	B-KEY
features	O
necessary	O
to	O
achieve	O
good	O
performance	O
.	O
We	O
evaluate	O
the	O
HPF	O
version	O
on	O
the	O
Hitachi	O
SR2201	O
,	O
a	O
distributed-memory	O
parallel	O
machine	O
.	O
With	O
16	O
processors	O
,	O
the	O
execution	O
time	O
of	O
the	O
HPF	O
version	O
is	O
within	O
a	O
factor	O
of	O
1.5	O
of	O
the	O
hand-parallelized	O
version	O
of	O
the	O
NPB	O
2.3	O
beta	O

Waiting-time	B-KEY
distribution	I-KEY
of	O
a	O
discrete-time	B-KEY
multiserver	I-KEY
queue	I-KEY
with	O
correlated	B-KEY
arrivals	I-KEY
and	O
deterministic	B-KEY
service	I-KEY
times	I-KEY
:	O
D-MAP/D/k	B-KEY
system	I-KEY
We	O
derive	O
the	O
waiting-time	B-KEY
distribution	I-KEY
of	O
a	O
discrete-time	B-KEY
multiserver	I-KEY
queue	I-KEY
with	O
correlated	B-KEY
arrivals	I-KEY
and	O
deterministic	O
-LRB-	O
or	O
constant	O
-RRB-	O
service	O
times	O
.	O
We	O
show	O
that	O
the	O
procedure	O
for	O
obtaining	O
the	O
waiting-time	B-KEY
distribution	I-KEY
of	O
a	O
multiserver	O
queue	O
is	O
reduced	O
to	O
that	O
of	O
a	O
single-server	O
queue	O
.	O
We	O
present	O
a	O
complete	O
solution	O
to	O
the	O
waiting-time	B-KEY
distribution	I-KEY
of	O
D-MAP/D/k	O
queue	O
together	O
with	O
some	O
computational	O
results	O

Open	B-KEY
hypermedia	I-KEY
for	O
product	B-KEY
support	I-KEY
As	O
industrial	O
systems	O
become	O
increasingly	O
more	O
complex	O
,	O
the	O
maintenance	B-KEY
and	O
operating	B-KEY
information	I-KEY
increases	O
both	O
in	O
volume	O
and	O
complexity	O
.	O
With	O
the	O
current	O
pressures	O
on	O
manufacturing	O
,	O
the	O
management	O
of	O
information	B-KEY
resources	I-KEY
has	O
become	O
a	O
critical	O
issue	O
.	O
In	O
particular	O
,	O
ensuring	O
that	O
personnel	O
can	O
access	O
current	O
information	O
quickly	O
and	O
effectively	O
when	O
undertaking	O
a	O
specific	O
task	O
.	O
This	O
paper	O
discusses	O
some	O
of	O
the	O
issues	O
involved	O
in	O
,	O
and	O
the	O
benefits	O
of	O
using	O
,	O
open	B-KEY
hypermedia	I-KEY
to	O
manage	O
and	O
deliver	O
a	O
diverse	O
range	O
of	O
information	O
.	O
While	O
the	O
paper	O
concentrates	O
on	O
the	O
problems	O
specifically	O
associated	O
with	O
manufacturing	O
organizations	O
,	O
the	O
problems	O
are	O
generic	O
across	O
other	O
business	O
sectors	O
such	O
as	O
healthcare	O
,	O
defence	O
and	O
finance	O
.	O
The	O
open	B-KEY
hypermedia	I-KEY
approach	O
to	O
information	O
management	O
and	O
delivery	O
allows	O
a	O
multimedia	O
resource	O
base	O
to	O
be	O
used	O
for	O
a	O
range	O
of	O
applications	O
and	O
it	O
permits	O
a	O
user	O
to	O
have	O
controlled	O
access	O
to	O
the	O
required	O
information	O
in	O
an	O
easily	O
accessible	O
and	O
structured	O
manner	O
.	O
Recent	O
advancement	O
in	O
hypermedia	O
also	O
permits	O
just-in-time	B-KEY
support	I-KEY
in	O
the	O
most	O
appropriate	O
format	O
for	O
all	O
users	O
.	O
Our	O
approach	O
is	O
illustrated	O
by	O
the	O
discussion	O
of	O
a	O
case	O
study	O
in	O
which	O
an	O
open	B-KEY
hypermedia	I-KEY
system	O
delivers	O
maintenance	B-KEY
and	O
process	O
information	O
to	O
factory-floor	O
users	O
to	O
support	O
the	O
maintenance	B-KEY
and	O
operation	O
of	O
a	O
very	O
large	O
manufacturing	O
cell	O

Algorithms	O
for	O
improving	O
the	O
quality	O
of	O
R-trees	B-KEY
A	O
novel	O
approach	O
to	O
operation	O
with	O
a	O
structure	O
for	O
spatial	B-KEY
indexing	I-KEY
of	O
extended	B-KEY
objects	I-KEY
shaped	O
as	O
R-trees	B-KEY
is	O
considered	O
.	O
It	O
consists	O
of	O
the	O
initial	O
global	O
construction	O
of	O
an	O
efficient	O
R-tree	B-KEY
structure	O
and	O
the	O
subsequent	O
operation	O
with	O
it	O
using	O
conventional	O
dynamic	B-KEY
algorithms	I-KEY
.	O
A	O
global	O
strategy	O
for	O
constructing	O
an	O
R-tree	B-KEY
reduced	O
to	O
a	O
problem	O
of	O
dividing	O
a	O
set	O
of	O
rectangular	B-KEY
objects	I-KEY
into	O
K	O
parts	O
with	O
minimum	B-KEY
mutual	I-KEY
overlay	I-KEY
is	O
suggested	O
.	O
Base	O
,	O
box	O
,	O
and	O
``	O
Divide	O
and	O
Conquer	O
''	O
algorithms	O
are	O
suggested	O
.	O
The	O
results	O
of	O
experimental	O
modeling	O
of	O
the	O
execution	O
of	O
various	O
algorithms	O
are	O
discussed	O

Wave	B-KEY
propagation	I-KEY
related	O
to	O
high-speed	O
train	O
.	O
A	O
scaled	O
boundary	O
FE-approach	O
for	O
unbounded	O
domains	O
Analysis	O
of	O
wave	B-KEY
propagation	I-KEY
in	O
solid	B-KEY
materials	I-KEY
under	O
moving	O
loads	O
is	O
a	O
topic	O
of	O
great	O
interest	O
in	O
railway	B-KEY
engineering	I-KEY
.	O
The	O
objective	O
of	O
the	O
paper	O
is	O
three-dimensional	O
modelling	B-KEY
of	O
high-speed	B-KEY
train	I-KEY
related	I-KEY
ground	I-KEY
vibrations	I-KEY
;	O
in	O
particular	O
the	O
question	O
of	O
how	O
to	O
account	O
for	O
the	O
unbounded	B-KEY
media	I-KEY
is	O
addressed	O
.	O
For	O
efficient	O
and	O
accurate	O
modelling	B-KEY
of	O
railway	B-KEY
structural	I-KEY
components	I-KEY
taking	O
the	O
unbounded	B-KEY
media	I-KEY
into	O
account	O
,	O
a	O
hybrid	O
method	O
based	O
on	O
a	O
combination	O
of	O
the	O
conventional	O
finite	O
element	O
method	O
and	O
scaled	B-KEY
boundary	I-KEY
finite	I-KEY
element	I-KEY
method	I-KEY
is	O
established	O
.	O
In	O
the	O
paper	O
,	O
element	B-KEY
matrices	I-KEY
and	O
solution	B-KEY
procedures	I-KEY
for	O
the	O
scaled	B-KEY
boundary	I-KEY
finite	I-KEY
element	I-KEY
method	I-KEY
-LRB-	O
SBFEM	O
-RRB-	O
are	O
derived	O
.	O
A	O
non-linear	O
finite	O
element	O
iteration	O
scheme	O
using	O
Lagrange	B-KEY
multipliers	I-KEY
and	O
coupling	O
between	O
the	O
unbounded	O
domain	O
and	O
the	O
finite	O
element	O
domain	O
are	O
also	O
discussed	O
.	O
Two	O
numerical	O
examples	O
including	O
one	O
example	O
demonstrating	O
the	O
dynamical	B-KEY
response	I-KEY
of	O
a	O
railroad	B-KEY
section	I-KEY
are	O
presented	O
to	O
demonstrate	O
the	O
performance	O
of	O
the	O
proposed	O
method	O

Robust	B-KEY
output	I-KEY
feedback	I-KEY
model	I-KEY
predictive	I-KEY
control	I-KEY
using	O
off-line	B-KEY
linear	I-KEY
matrix	I-KEY
inequalities	I-KEY
A	O
fundamental	O
question	O
about	O
model	O
predictive	O
control	O
-LRB-	O
MPC	O
-RRB-	O
is	O
its	O
robustness	O
to	O
model	O
uncertainty	O
.	O
In	O
this	O
paper	O
,	O
we	O
present	O
a	O
robust	B-KEY
constrained	I-KEY
output	I-KEY
feedback	I-KEY
MPC	I-KEY
algorithm	I-KEY
that	O
can	O
stabilize	O
plants	O
with	O
both	O
polytopic	B-KEY
uncertainty	I-KEY
and	O
norm-bound	B-KEY
uncertainty	I-KEY
.	O
The	O
design	O
procedure	O
involves	O
off-line	O
design	O
of	O
a	O
robust	B-KEY
constrained	I-KEY
state	I-KEY
feedback	I-KEY
MPC	I-KEY
law	I-KEY
and	O
a	O
state	B-KEY
estimator	I-KEY
using	O
linear	O
matrix	O
inequalities	O
-LRB-	O
LMIs	O
-RRB-	O
.	O
Since	O
we	O
employ	O
an	O
off-line	O
approach	O
for	O
the	O
controller	O
design	O
which	O
gives	O
a	O
sequence	O
of	O
explicit	O
control	O
laws	O
,	O
we	O
are	O
able	O
to	O
analyze	O
the	O
robust	O
stabilizability	O
of	O
the	O
combined	O
control	O
laws	O
and	O
estimator	O
,	O
and	O
by	O
adjusting	O
the	O
design	O
parameters	O
,	O
guarantee	O
robust	O
stability	O
of	O
the	O
closed-loop	B-KEY
system	I-KEY
in	O
the	O
presence	O
of	O
constraints	O
.	O
The	O
algorithm	O
is	O
illustrated	O
with	O
two	O
examples	O

Optimization	O
of	O
requantization	O
parameter	O
for	O
MPEG	O
transcoding	O
This	O
paper	O
considers	O
transcoding	O
in	O
which	O
an	O
MPEG	O
stream	O
is	O
converted	O
to	O
a	O
low-bit-rate	B-KEY
MPEG	I-KEY
stream	I-KEY
,	O
and	O
proposes	O
a	O
method	O
in	O
which	O
the	O
transcoding	B-KEY
error	I-KEY
can	O
be	O
reduced	O
by	O
optimally	O
selecting	O
the	O
quantization	O
parameter	O
for	O
each	O
macroblock	B-KEY
.	O
In	O
transcoding	O
with	O
a	O
low	O
compression	B-KEY
ratio	I-KEY
,	O
it	O
is	O
crucial	O
to	O
prohibit	O
transcoding	O
with	O
a	O
requantization	O
parameter	O
which	O
is	O
1	O
to	O
2	O
times	O
the	O
quantization	O
parameter	O
of	O
the	O
input	O
stream	O
.	O
Consequently	O
,	O
as	O
the	O
first	O
step	O
,	O
an	O
optimization	O
method	O
for	O
the	O
requantization	O
parameter	O
is	O
proposed	O
which	O
cares	O
for	O
the	O
error	B-KEY
propagation	I-KEY
effect	I-KEY
by	O
interframe	B-KEY
prediction	I-KEY
.	O
Then	O
,	O
the	O
proposed	O
optimization	O
method	O
is	O
extended	O
so	O
that	O
the	O
method	O
can	O
also	O
be	O
applied	O
to	O
the	O
case	O
of	O
a	O
high	O
compression	B-KEY
ratio	I-KEY
in	O
which	O
the	O
rate-distortion	B-KEY
curve	I-KEY
is	O
approximated	O
for	O
each	O
macroblock	B-KEY
in	O
the	O
range	O
of	O
requantization	O
parameters	O
larger	O
than	O
2	O
times	O
the	O
quantization	O
parameter	O
.	O
It	O
is	O
verified	O
by	O
a	O
simulation	B-KEY
experiment	O
that	O
the	O
PSNR	B-KEY
is	O
improved	O
by	O
0.5	O
to	O
0.8	O
dB	O
compared	O
to	O
the	O
case	O
in	O
which	O
a	O
6	O
Mbit/s	O
MPEG	O
stream	O
is	O
not	O
optimized	O
by	O
twofold	O
recompression	O

Development	O
of	O
railway	O
VR	O
safety	O
simulation	O
system	O
Abnormal	O
conditions	O
occur	O
in	O
railway	B-KEY
transportation	I-KEY
due	O
to	O
trouble	O
or	O
accidents	B-KEY
and	O
it	O
affects	O
a	O
number	O
of	O
passengers	O
.	O
It	O
is	O
very	O
important	O
,	O
therefore	O
,	O
to	O
quickly	O
recover	O
and	O
return	O
to	O
normal	B-KEY
train	I-KEY
operation	I-KEY
.	O
For	O
this	O
purpose	O
we	O
developed	O
a	O
system	O
,	O
``	O
Computer	B-KEY
VR	I-KEY
Simulation	I-KEY
System	I-KEY
for	O
the	O
Safety	O
of	O
Railway	B-KEY
Transportation	I-KEY
.	O
''	O
It	O
is	O
a	O
new	O
type	O
simulation	O
system	O
to	O
evaluate	O
measures	O
to	O
be	O
taken	O
under	O
abnormal	O
conditions	O
.	O
Users	O
of	O
this	O
simulation	O
system	O
cooperate	O
with	O
one	O
another	O
to	O
correct	O
the	O
abnormal	O
conditions	O
that	O
have	O
occurred	O
in	O
virtual	O
reality	O
.	O
This	O
paper	O
reports	O
the	O
newly	O
developed	O
simulation	O
system	O

Securing	O
the	O
Internet	O
routing	B-KEY
infrastructure	I-KEY
The	O
unprecedented	O
growth	O
of	O
the	O
Internet	O
over	O
the	O
last	O
years	O
,	O
and	O
the	O
expectation	O
of	O
an	O
even	O
faster	O
increase	O
in	O
the	O
numbers	O
of	O
users	O
and	O
networked	B-KEY
systems	I-KEY
,	O
resulted	O
in	O
the	O
Internet	O
assuming	O
its	O
position	O
as	O
a	O
mass	O
communication	O
medium	O
.	O
At	O
the	O
same	O
time	O
,	O
the	O
emergence	O
of	O
an	O
increasingly	O
large	O
number	O
of	O
application	O
areas	O
and	O
the	O
evolution	O
of	O
the	O
networking	B-KEY
technology	I-KEY
suggest	O
that	O
in	O
the	O
near	O
future	O
the	O
Internet	O
may	O
become	O
the	O
single	O
integrated	B-KEY
communication	I-KEY
infrastructure	I-KEY
.	O
However	O
,	O
as	O
the	O
dependence	O
on	O
the	O
networking	B-KEY
infrastructure	I-KEY
grows	O
,	O
its	O
security	O
becomes	O
a	O
major	O
concern	O
,	O
in	O
light	O
of	O
the	O
increased	O
attempt	O
to	O
compromise	O
the	O
infrastructure	O
.	O
In	O
particular	O
,	O
the	O
routing	O
operation	O
is	O
a	O
highly	O
visible	O
target	O
that	O
must	O
be	O
shielded	O
against	O
a	O
wide	O
range	O
of	O
attacks	O
.	O
The	O
injection	O
of	O
false	B-KEY
routing	I-KEY
information	I-KEY
can	O
easily	O
degrade	O
network	B-KEY
performance	I-KEY
,	O
or	O
even	O
cause	O
denial	O
of	O
service	O
for	O
a	O
large	O
number	O
of	O
hosts	O
and	O
networks	O
over	O
a	O
long	O
period	O
of	O
time	O
.	O
Different	O
approaches	O
have	O
been	O
proposed	O
to	O
secure	O
the	O
routing	B-KEY
protocols	I-KEY
,	O
with	O
a	O
variety	O
of	O
countermeasures	B-KEY
,	O
which	O
,	O
nonetheless	O
,	O
have	O
not	O
eradicated	O
the	O
vulnerability	O
of	O
the	O
routing	B-KEY
infrastructure	I-KEY
.	O
In	O
this	O
article	O
,	O
we	O
survey	O
the	O
up-to-date	O
secure	B-KEY
routing	I-KEY
schemes	I-KEY
.	O
that	O
appeared	O
over	O
the	O
last	O
few	O
years	O
.	O
Our	O
critical	O
point	O
of	O
view	O
and	O
thorough	O
review	O
of	O
the	O
literature	O
are	O
an	O
attempt	O
to	O
identify	O
directions	O
for	O
future	O
research	B-KEY
on	O
an	O
indeed	O
difficult	O
and	O
still	O
largely	O
open	O
problem	O

Intensity	B-KEY
based	I-KEY
affine	I-KEY
registration	I-KEY
including	O
feature	B-KEY
similarity	I-KEY
for	O
spatial	B-KEY
normalization	I-KEY
This	O
paper	O
presents	O
a	O
new	O
spatial	B-KEY
normalization	I-KEY
with	O
affine	B-KEY
transformation	I-KEY
.	O
The	O
quantitative	O
comparison	O
of	O
brain	B-KEY
architecture	I-KEY
across	O
different	O
subjects	O
requires	O
a	O
common	B-KEY
coordinate	I-KEY
system	I-KEY
.	O
For	O
the	O
analysis	O
of	O
a	O
specific	O
brain	O
area	O
,	O
it	O
is	O
necessary	O
to	O
normalize	O
and	O
compare	O
a	O
region	B-KEY
of	I-KEY
interest	I-KEY
and	O
the	O
global	B-KEY
brain	I-KEY
.	O
The	O
intensity	O
based	O
registration	O
method	O
matches	O
the	O
global	B-KEY
brain	I-KEY
well	O
,	O
but	O
a	O
region	B-KEY
of	I-KEY
interest	I-KEY
may	O
not	O
be	O
locally	O
normalized	O
compared	O
to	O
the	O
feature	O
based	O
method	O
.	O
The	O
method	O
in	O
this	O
paper	O
uses	O
feature	B-KEY
similarities	I-KEY
of	O
local	O
regions	O
as	O
well	O
as	O
intensity	O
similarities	O
.	O
The	O
lateral	B-KEY
ventricle	I-KEY
and	O
central	B-KEY
gray	I-KEY
nuclei	I-KEY
of	O
the	O
brain	O
,	O
including	O
the	O
corpus	B-KEY
callosum	I-KEY
,	O
which	O
is	O
used	O
for	O
features	O
in	O
schizophrenia	B-KEY
detection	I-KEY
,	O
is	O
appropriately	O
normalized	O
.	O
Our	O
method	O
reduces	O
the	O
difference	O
of	O
feature	O
areas	O
such	O
as	O
the	O
corpus	B-KEY
callosum	I-KEY
-LRB-	O
7.7	O
%	O
,	O
2.4	O
%	O
-RRB-	O
and	O
lateral	B-KEY
ventricle	I-KEY
-LRB-	O
8.2	O
%	O
,	O
13.5	O
%	O
-RRB-	O
compared	O
with	O
mutual	O
information	O
and	O
Talairach	B-KEY
methods	I-KEY

Anatomy	O
of	O
the	O
coupling	B-KEY
query	I-KEY
in	O
a	O
Web	B-KEY
warehouse	I-KEY
To	O
populate	O
a	O
data	B-KEY
warehouse	I-KEY
specifically	O
designed	O
for	O
Web	O
data	O
,	O
i.e.	O
Web	B-KEY
warehouse	I-KEY
,	O
it	O
is	O
imperative	O
to	O
harness	O
relevant	O
documents	O
from	O
the	O
Web	O
.	O
In	O
this	O
paper	O
,	O
we	O
describe	O
a	O
query	O
mechanism	O
called	O
coupling	B-KEY
query	I-KEY
to	O
glean	O
relevant	O
Web	O
data	O
in	O
the	O
context	O
of	O
our	O
Web	O
warehousing	O
system	O
called	O
Warehouse	B-KEY
Of	I-KEY
Web	I-KEY
Data	I-KEY
-LRB-	O
WHOWEDA	O
-RRB-	O
.	O
A	O
coupling	B-KEY
query	I-KEY
may	O
be	O
used	O
for	O
querying	O
both	O
HTML	O
and	O
XML	B-KEY
documents	I-KEY
.	O
Important	O
features	O
of	O
our	O
query	O
mechanism	O
are	O
the	O
ability	O
to	O
query	O
metadata	B-KEY
,	O
content	B-KEY
,	O
internal	O
and	O
external	O
-LRB-	O
hyperlink	O
-RRB-	O
structure	O
of	O
Web	B-KEY
documents	I-KEY
based	O
on	O
partial	B-KEY
knowledge	I-KEY
,	O
ability	O
to	O
express	O
constraints	O
on	O
tag	B-KEY
attributes	I-KEY
and	O
tagless	B-KEY
segment	I-KEY
of	O
data	O
,	O
ability	O
to	O
express	O
conjunctive	O
as	O
well	O
as	O
disjunctive	B-KEY
query	I-KEY
conditions	I-KEY
compactly	O
,	O
ability	O
to	O
control	O
execution	O
of	O
a	O
Web	O
query	O
and	O
preservation	O
of	O
the	O
topological	B-KEY
structure	I-KEY
of	O
hyperlinked	B-KEY
documents	I-KEY
in	O
the	O
query	O
results	O
.	O
We	O
also	O
discuss	O
how	O
to	O
formulate	O
a	O
query	O
graphically	O
and	O
in	O
textual	O
form	O
using	O
a	O
coupling	O
graph	O
and	O
coupling	B-KEY
text	I-KEY
,	O
respectively	O

On	O
deciding	O
stability	O
of	O
constrained	B-KEY
homogeneous	I-KEY
random	I-KEY
walks	I-KEY
and	O
queueing	B-KEY
systems	I-KEY
We	O
investigate	O
stability	O
of	O
scheduling	O
policies	O
in	O
queueing	B-KEY
systems	I-KEY
.	O
To	O
this	O
day	O
no	O
algorithmic	O
characterization	O
exists	O
for	O
checking	O
stability	O
of	O
a	O
given	O
policy	O
in	O
a	O
given	O
queueing	B-KEY
system	I-KEY
.	O
In	O
this	O
paper	O
we	O
introduce	O
a	O
certain	O
generalized	B-KEY
priority	I-KEY
policy	I-KEY
and	O
prove	O
that	O
the	O
stability	O
of	O
this	O
policy	O
is	O
algorithmically	O
undecidable	O
.	O
We	O
also	O
prove	O
that	O
stability	O
of	O
a	O
homogeneous	O
random	O
walk	O
in	O
L/sub	O
+	O
/	O
/	O
sup	O
d	O
/	O
is	O
undecidable	O
.	O
Finally	O
,	O
we	O
show	O
that	O
the	O
problem	O
of	O
computing	O
a	O
fluid	O
limit	O
of	O
a	O
queueing	B-KEY
system	I-KEY
or	O
of	O
a	O
constrained	B-KEY
homogeneous	I-KEY
random	I-KEY
walk	I-KEY
is	O
undecidable	O
.	O
To	O
the	O
best	O
of	O
our	O
knowledge	O
these	O
are	O
the	O
first	O
undecidability	B-KEY
results	I-KEY
in	O
the	O
area	O
of	O
stability	O
of	O
queueing	B-KEY
systems	I-KEY
and	O
random	O
walks	O
in	O
L/sub	O
+	O
/	O
/	O
sup	O
d	O
/	O
.	O
We	O
conjecture	O
that	O
stability	O
of	O
common	O
policies	O
like	O
First-In-First-Out	O
and	O
priority	B-KEY
policy	I-KEY
is	O
also	O
an	O
undecidable	B-KEY
problem	I-KEY

Harmless	B-KEY
delays	I-KEY
in	O
Cohen-Grossberg	B-KEY
neural	I-KEY
networks	I-KEY
Without	O
assuming	O
monotonicity	B-KEY
and	O
differentiability	B-KEY
of	O
the	O
activation	B-KEY
functions	I-KEY
and	O
any	O
symmetry	O
of	O
interconnections	B-KEY
,	O
we	O
establish	O
some	O
sufficient	O
conditions	O
for	O
the	O
globally	B-KEY
asymptotic	I-KEY
stability	I-KEY
of	O
a	O
unique	O
equilibrium	O
for	O
the	O
Cohen-Grossberg	O
-LRB-	O
1983	O
-RRB-	O
neural	O
network	O
with	O
multiple	B-KEY
delays	I-KEY
.	O
Lyapunov	B-KEY
functionals	I-KEY
and	O
functions	O
combined	O
with	O
the	O
Razumikhin	B-KEY
technique	I-KEY
are	O
employed	O
.	O
The	O
criteria	O
are	O
all	O
independent	O
of	O
the	O
magnitudes	O
of	O
the	O
delays	O
,	O
and	O
thus	O
the	O
delays	O
under	O
these	O
conditions	O
are	O
harmless	O

International	B-KEY
customers	I-KEY
,	O
suppliers	O
,	O
and	O
document	B-KEY
delivery	I-KEY
in	O
a	O
fee-based	O
information	O
service	O
The	O
Purdue	O
University	O
Libraries	O
library	O
fee-based	O
information	O
service	O
,	O
the	O
Technical	B-KEY
Information	I-KEY
Service	I-KEY
-LRB-	O
TIS	O
-RRB-	O
,	O
works	O
with	O
both	O
international	B-KEY
customers	I-KEY
and	O
international	B-KEY
suppliers	I-KEY
to	O
meet	O
its	O
customers	O
'	O
needs	O
for	O
difficult	O
and	O
esoteric	O
document	B-KEY
requests	I-KEY
.	O
Successful	O
completion	O
of	O
these	O
orders	O
requires	O
the	O
ability	O
to	O
verify	O
fragmentary	O
citations	O
;	O
ascertain	O
documents	O
'	O
availability	O
;	O
obtain	O
pricing	B-KEY
information	I-KEY
;	O
calculate	O
inclusive	B-KEY
cost	I-KEY
quotes	I-KEY
;	O
meet	O
customers	O
'	O
deadlines	O
;	O
accept	O
international	B-KEY
payments	I-KEY
;	O
and	O
ship	O
across	O
borders	O
.	O
While	O
international	O
orders	O
make	O
tip	O
a	O
small	O
percent	O
of	O
the	O
total	O
workload	O
,	O
these	O
challenging	O
and	O
rewarding	O
orders	O
meet	O
customers	O
'	O
needs	O
and	O
offer	O
continuous	O
improvement	O
opportunities	O
to	O
the	O
staff	O

Conditions	O
for	O
the	O
local	B-KEY
manipulation	I-KEY
of	O
Gaussian	B-KEY
states	I-KEY
We	O
present	O
a	O
general	O
necessary	O
and	O
sufficient	O
criterion	O
for	O
the	O
possibility	O
of	O
a	O
state	B-KEY
transformation	I-KEY
from	O
one	O
mixed	O
Gaussian	B-KEY
state	I-KEY
to	O
another	O
of	O
a	O
bipartite	B-KEY
continuous-variable	I-KEY
system	I-KEY
with	O
two	O
modes	O
.	O
The	O
class	O
of	O
operations	O
that	O
will	O
be	O
considered	O
is	O
the	O
set	O
of	O
local	O
Gaussian	O
completely	O
positive	O
trace-preserving	B-KEY
maps	I-KEY

Taxonomy	O
's	O
role	O
in	O
content	B-KEY
management	I-KEY
A	O
taxonomy	O
is	O
simply	O
a	O
way	O
of	O
classifying	O
things	O
.	O
Still	O
,	O
there	O
is	O
a	O
rapidly	O
growing	O
list	O
of	O
vendors	O
offering	O
taxonomy	B-KEY
software	I-KEY
and	O
related	O
applications	O
.	O
They	O
promise	O
many	O
benefits	O
,	O
especially	O
to	O
enterprise	B-KEY
customers	I-KEY
:	O
Content	B-KEY
management	I-KEY
will	O
be	O
more	O
efficient	O
.	O
Corporate	B-KEY
portals	I-KEY
will	O
be	O
enhanced	O
by	O
easily	O
created	O
Yahoo!	O
-	O
like	O
directories	O
of	O
internal	B-KEY
information	I-KEY
.	O
And	O
the	O
end-user	O
experience	O
will	O
be	O
dramatically	O
improved	O
by	O
more	O
successful	O
content	O
retrieval	O
and	O
more	O
effective	B-KEY
knowledge	I-KEY
discovery	I-KEY
.	O
But	O
today	O
's	O
taxonomy	O
products	O
represent	O
emerging	O
technologies	O
.	O
They	O
are	O
not	O
out-of-the-box	O
solutions	O
.	O
And	O
even	O
the	O
most	O
automated	O
systems	O
require	O
some	O
manual	O
assistance	O
from	O
people	O
who	O
know	O
how	O
to	O
classify	O
content	O

Quantum-state	B-KEY
information	I-KEY
retrieval	I-KEY
in	O
a	O
Rydberg-atom	B-KEY
data	I-KEY
register	I-KEY
We	O
analyze	O
a	O
quantum	B-KEY
search	I-KEY
protocol	I-KEY
to	O
retrieve	O
phase	B-KEY
information	I-KEY
from	O
a	O
Rydberg-atom	B-KEY
data	I-KEY
register	I-KEY
using	O
a	O
subpicosecond	B-KEY
half-cycle	I-KEY
electric	I-KEY
field	I-KEY
pulse	I-KEY
.	O
Calculations	O
show	O
that	O
the	O
half-cycle	B-KEY
pulse	I-KEY
can	O
perform	O
the	O
phase	B-KEY
retrieval	I-KEY
only	O
within	O
a	O
range	O
of	O
peak	B-KEY
field	I-KEY
values	I-KEY
.	O
By	O
varying	O
the	O
phases	O
of	O
the	O
constituent	B-KEY
orbitals	I-KEY
of	O
the	O
Rydberg	B-KEY
wave	I-KEY
packet	I-KEY
register	I-KEY
,	O
we	O
demonstrate	O
coherent	B-KEY
control	I-KEY
of	O
the	O
phase	B-KEY
retrieval	I-KEY
process	O
.	O
By	O
specially	O
programming	O
the	O
phases	O
of	O
the	O
orbitals	O
comprising	O
the	O
initial	B-KEY
wave	I-KEY
packet	I-KEY
,	O
we	O
show	O
that	O
it	O
is	O
possible	O
to	O
use	O
the	O
search	B-KEY
method	I-KEY
as	O
a	O
way	O
to	O
synthesize	O
single	B-KEY
energy	I-KEY
eigenstates	I-KEY

A	O
novel	O
genetic	B-KEY
algorithm	I-KEY
for	O
the	O
design	O
of	O
a	O
signed	O
power-of-two	O
coefficient	O
quadrature	B-KEY
mirror	I-KEY
filter	I-KEY
lattice	B-KEY
filter	I-KEY
bank	I-KEY
A	O
novel	O
genetic	B-KEY
algorithm	I-KEY
-LRB-	O
GA	O
-RRB-	O
for	O
the	O
design	O
of	O
a	O
canonical	O
signed	O
power-of-two	O
-LRB-	O
SPT	O
-RRB-	O
coefficient	O
lattice	O
structure	O
quadrature	B-KEY
mirror	I-KEY
filter	I-KEY
bank	O
is	O
presented	O
.	O
Genetic	O
operations	O
may	O
render	O
the	O
SPT	O
representation	O
of	O
a	O
value	O
noncanonical	O
.	O
A	O
new	O
encoding	B-KEY
scheme	I-KEY
is	O
introduced	O
to	O
encode	O
the	O
SPT	O
values	O
.	O
In	O
this	O
new	O
scheme	O
,	O
the	O
canonical	O
property	O
of	O
the	O
SPT	O
values	O
is	O
preserved	O
under	O
genetic	O
operations	O
.	O
Additionally	O
,	O
two	O
new	O
features	O
that	O
drastically	O
improve	O
the	O
performance	O
of	O
our	O
GA	O
are	O
introduced	O
.	O
-LRB-	O
1	O
-RRB-	O
An	O
additional	O
level	O
of	O
natural	B-KEY
selection	I-KEY
is	O
introduced	O
to	O
simulate	O
the	O
effect	O
of	O
natural	B-KEY
selection	I-KEY
when	O
sperm	O
cells	O
compete	O
to	O
fertilize	O
an	O
ovule	O
;	O
this	O
dramatically	O
improves	O
the	O
offspring	B-KEY
survival	I-KEY
rate	I-KEY
.	O
A	O
conventional	O
GA	O
is	O
analogous	O
to	O
intracytoplasmic	O
sperm	O
injection	O
and	O
has	O
an	O
extremely	O
low	O
offspring	B-KEY
survival	I-KEY
rate	I-KEY
,	O
resulting	O
in	O
very	O
slow	O
convergence	O
.	O
-LRB-	O
2	O
-RRB-	O
The	O
probability	O
of	O
mutation	O
for	O
each	O
codon	O
of	O
a	O
chromosome	O
is	O
weighted	O
by	O
the	O
reciprocal	O
of	O
its	O
effect	O
.	O
Because	O
of	O
these	O
new	O
features	O
,	O
the	O
performance	O
of	O
our	O
new	O
GA	O
outperforms	O
conventional	O
GAs	O

People	O
who	O
make	O
a	O
difference	O
:	O
mentors	B-KEY
and	O
role	B-KEY
models	I-KEY
The	O
literature	O
of	O
gender	B-KEY
issues	I-KEY
in	O
computing	B-KEY
steadfastly	O
and	O
uniformly	O
has	O
advocated	O
the	O
use	O
of	O
mentors	B-KEY
and	O
role	B-KEY
models	I-KEY
-LRB-	O
M&RM	O
-RRB-	O
for	O
recruiting	O
and	O
retaining	O
women	O
in	O
computer	B-KEY
science	O
.	O
This	O
paper	O
,	O
therefore	O
,	O
accepts	O
the	O
results	O
of	O
research	O
studies	O
and	O
avoids	O
reiterating	O
details	O
of	O
the	O
projects	O
but	O
offers	O
instead	O
a	O
practical	O
guide	O
for	O
using	O
M&RM	O
to	O
recruit	O
and	O
retain	O
women	O
in	O
computer	B-KEY
science	O
.	O
The	O
guide	O
provides	O
pragmatic	O
advice	O
,	O
describing	O
several	O
different	O
facets	O
of	O
the	O
M&RM	O
concept	O

Monoids	B-KEY
all	I-KEY
polygons	I-KEY
over	O
which	O
are	O
omega	O
-	O
stable	O
:	O
proof	O
of	O
the	O
Mustafin-Poizat	B-KEY
conjecture	I-KEY
A	O
monoid	O
S	O
is	O
called	O
an	O
omega	O
-	O
stabilizer	O
-LRB-	O
superstabilizer	O
,	O
or	O
stabilizer	O
-RRB-	O
if	O
every	O
S-polygon	B-KEY
has	O
an	O
omega	O
-	O
stable	O
-LRB-	O
superstable	O
,	O
or	O
stable	O
-RRB-	O
theory	O
.	O
It	O
is	O
proved	O
that	O
every	O
omega	O
-	O
stabilizer	O
is	O
a	O
regular	B-KEY
monoid	I-KEY
.	O
This	O
confirms	O
the	O
Mustafin-Poizat	B-KEY
conjecture	I-KEY
and	O
allows	O
us	O
to	O
end	O
up	O
the	O
description	O
of	O
omega	O
-	O
stabilizers	O

Watermarking	B-KEY
techniques	I-KEY
for	O
electronic	B-KEY
delivery	I-KEY
of	O
remote	B-KEY
sensing	I-KEY
images	I-KEY
Earth	B-KEY
observation	I-KEY
missions	I-KEY
have	O
recently	O
attracted	O
a	O
growing	O
interest	O
,	O
mainly	O
due	O
to	O
the	O
large	O
number	O
of	O
possible	O
applications	O
capable	O
of	O
exploiting	O
remotely	O
sensed	O
data	O
and	O
images	O
.	O
Along	O
with	O
the	O
increase	O
of	O
market	O
potential	O
,	O
the	O
need	O
arises	O
for	O
the	O
protection	O
of	O
the	O
image	O
products	O
.	O
Such	O
a	O
need	O
is	O
a	O
very	O
crucial	O
one	O
,	O
because	O
the	O
Internet	O
and	O
other	O
public/private	O
networks	O
have	O
become	O
preferred	O
means	O
of	O
data	O
exchange	O
.	O
A	O
critical	O
issue	O
arising	O
when	O
dealing	O
with	O
digital	B-KEY
image	I-KEY
distribution	I-KEY
is	O
copyright	B-KEY
protection	I-KEY
.	O
Such	O
a	O
problem	O
has	O
been	O
largely	O
addressed	O
by	O
resorting	O
to	O
watermarking	O
technology	O
.	O
A	O
question	O
that	O
obviously	O
arises	O
is	O
whether	O
the	O
requirements	O
imposed	O
by	O
remote	O
sensing	O
imagery	O
are	O
compatible	O
with	O
existing	O
watermarking	B-KEY
techniques	I-KEY
.	O
On	O
the	O
basis	O
of	O
these	O
motivations	O
,	O
the	O
contribution	O
of	O
this	O
work	O
is	O
twofold	O
:	O
assessment	O
of	O
the	O
requirements	O
imposed	O
by	O
remote	O
sensing	O
applications	O
on	O
watermark-based	O
copyright	B-KEY
protection	I-KEY
,	O
and	O
modification	O
of	O
two	O
well-established	O
digital	B-KEY
watermarking	I-KEY
techniques	O
to	O
meet	O
such	O
constraints	O
.	O
More	O
specifically	O
,	O
the	O
concept	O
of	O
near-lossless	B-KEY
watermarking	I-KEY
is	O
introduced	O
and	O
two	O
possible	O
algorithms	O
matching	O
such	O
a	O
requirement	O
are	O
presented	O
.	O
Experimental	O
results	O
are	O
shown	O
to	O
measure	O
the	O
impact	O
of	O
watermark	O
introduction	O
on	O
a	O
typical	O
remote	O
sensing	O
application	O
,	O
i.e.	O
,	O
unsupervised	B-KEY
image	I-KEY
classification	I-KEY

Control	O
of	O
a	O
coupled	B-KEY
map	I-KEY
lattice	I-KEY
model	O
for	O
vortex	B-KEY
shedding	I-KEY
in	O
the	O
wake	B-KEY
of	O
a	O
cylinder	B-KEY
The	O
flow	O
behind	O
a	O
vibrating	B-KEY
flexible	I-KEY
cable	I-KEY
at	O
low	B-KEY
Reynolds	I-KEY
numbers	I-KEY
can	O
exhibit	O
complex	O
wake	B-KEY
structures	O
such	O
as	O
lace-like	O
patterns	O
,	O
vortex	B-KEY
dislocations	I-KEY
and	O
frequency	O
cells	O
.	O
These	O
structures	O
have	O
been	O
observed	O
in	O
experiments	O
and	O
numerical	O
simulations	O
,	O
and	O
are	O
predicted	O
by	O
a	O
previously	O
developed	O
low-order	O
coupled	B-KEY
map	I-KEY
lattice	I-KEY
-LRB-	O
CML	O
-RRB-	O
.	O
The	O
discrete	O
-LRB-	O
in	O
time	O
and	O
space	O
-RRB-	O
CML	O
models	O
consist	O
of	O
a	O
series	O
of	O
diffusively	O
coupled	B-KEY
circle	I-KEY
map	I-KEY
oscillators	I-KEY
along	O
the	O
cable	O
span	O
.	O
Motivated	O
by	O
a	O
desire	O
to	O
modify	O
the	O
complex	O
wake	B-KEY
patterns	O
behind	O
flexible	O
vibrating	O
cables	O
,	O
we	O
have	O
studied	O
the	O
addition	O
of	O
control	O
terms	O
into	O
the	O
highly	O
efficient	O
CML	O
models	O
and	O
explored	O
the	O
resulting	O
dynamics	O
.	O
Proportional	O
,	O
adaptive	O
proportional	O
and	O
discontinuous	O
non-linear	O
-LRB-	O
DNL	O
-RRB-	O
control	O
methods	O
were	O
used	O
to	O
derive	O
the	O
control	O
laws	O
.	O
The	O
first	O
method	O
employed	O
occasional	O
proportional	B-KEY
feedback	I-KEY
.	O
The	O
adaptive	O
method	O
used	O
spatio-temporal	B-KEY
feedback	I-KEY
control	I-KEY
.	O
The	O
DNL	O
method	O
used	O
a	O
discontinuous	O
feedback	O
linearization	O
procedure	O
,	O
and	O
the	O
controller	O
was	O
designed	O
for	O
the	O
resulting	O
linearized	O
system	O
using	O
eigenvalue	O
assignment	O
.	O
These	O
techniques	O
were	O
applied	O
to	O
a	O
modeled	O
vortex	B-KEY
dislocation	I-KEY
structure	O
in	O
the	O
wake	B-KEY
of	O
a	O
vibrating	O
cable	O
in	O
uniform	O
freestream	O
flow	O
.	O
Parallel	O
shedding	O
patterns	O
were	O
achieved	O
for	O
a	O
range	O
of	O
forcing	O
frequency-forcing	O
amplitude	O
combinations	O
studied	O
to	O
validate	O
the	O
control	O
theory	O
.	O
The	O
adaptive	O
proportional	O
and	O
DNL	O
methods	O
were	O
found	O
to	O
be	O
more	O
effective	O
than	O
the	O
proportional	O
control	O
method	O
due	O
to	O
the	O
incorporation	O
of	O
a	O
spatially	O
varying	O
feedback	O
gain	O
across	O
the	O
cylinder	B-KEY
span	O
.	O
The	O
DNL	O
method	O
was	O
found	O
to	O
be	O
the	O
most	O
efficient	O
controller	O
of	O
the	O
low-order	O
CML	O
model	O
.	O
The	O
required	O
control	O
level	O
across	O
the	O
cable	O
span	O
was	O
correlated	O
to	O
the	O
1/1	B-KEY
lock-on	I-KEY
behavior	O
of	O
the	O
temporal	B-KEY
circle	I-KEY
map	I-KEY

Permission	B-KEY
grids	I-KEY
:	O
practical	O
,	O
error-bounded	B-KEY
simplification	I-KEY
We	O
introduce	O
the	O
permission	B-KEY
grid	I-KEY
,	O
a	O
spatial	B-KEY
occupancy	I-KEY
grid	I-KEY
which	O
can	O
be	O
used	O
to	O
guide	O
almost	O
any	O
standard	O
polygonal	B-KEY
surface	I-KEY
simplification	I-KEY
algorithm	I-KEY
into	O
generating	O
an	O
approximation	B-KEY
with	O
a	O
guaranteed	B-KEY
geometric	I-KEY
error	I-KEY
bound	I-KEY
.	O
In	O
particular	O
,	O
all	O
points	O
on	O
the	O
approximation	B-KEY
are	O
guaranteed	O
to	O
be	O
within	O
some	O
user-specified	B-KEY
distance	I-KEY
from	O
the	O
original	O
surface	O
.	O
Such	O
bounds	O
are	O
notably	O
absent	O
from	O
many	O
current	O
simplification	O
methods	O
,	O
and	O
are	O
becoming	O
increasingly	O
important	O
for	O
applications	O
in	O
scientific	B-KEY
computing	I-KEY
and	O
adaptive	B-KEY
level	I-KEY
of	I-KEY
detail	I-KEY
control	I-KEY
.	O
Conceptually	O
simple	O
,	O
the	O
permission	B-KEY
grid	I-KEY
defines	O
a	O
volume	O
in	O
which	O
the	O
approximation	B-KEY
must	O
lie	O
,	O
and	O
does	O
not	O
permit	O
the	O
underlying	O
simplification	O
algorithm	O
to	O
generate	O
approximations	B-KEY
outside	O
the	O
volume	O
.	O
The	O
permission	B-KEY
grid	I-KEY
makes	O
three	O
important	O
,	O
practical	O
improvements	O
over	O
current	O
error-bounded	B-KEY
simplification	I-KEY
methods	O
.	O
First	O
,	O
it	O
works	O
on	O
arbitrary	B-KEY
triangular	I-KEY
models	I-KEY
,	O
handling	O
all	O
manners	O
of	O
mesh	B-KEY
degeneracies	I-KEY
gracefully	O
.	O
Further	O
,	O
the	O
error	B-KEY
tolerance	I-KEY
may	O
be	O
easily	O
expanded	O
as	O
simplification	O
proceeds	O
,	O
allowing	O
the	O
construction	O
of	O
an	O
error-bounded	O
level	O
of	O
detail	O
hierarchy	O
with	O
vertex	B-KEY
correspondences	I-KEY
among	O
all	O
levels	O
of	O
detail	O
.	O
And	O
finally	O
,	O
the	O
permission	B-KEY
grid	I-KEY
has	O
a	O
representation	B-KEY
complexity	I-KEY
independent	O
of	O
the	O
size	O
of	O
the	O
input	O
model	O
,	O
and	O
a	O
small	O
running	B-KEY
time	I-KEY
overhead	I-KEY
,	O
making	O
it	O
more	O
practical	O
and	O
efficient	O
than	O
current	O
methods	O
with	O
similar	O
guarantees	O

Aggregators	O
versus	O
disintermediators	B-KEY
:	O
battling	O
it	O
out	O
in	O
the	O
information	O
superhighstreet	O
Perhaps	O
the	O
future	O
of	O
large-scale	B-KEY
content	I-KEY
aggregators	I-KEY
is	O
now	O
no	O
longer	O
in	O
doubt	O
but	O
this	O
was	O
not	O
the	O
case	O
10	O
years	O
ago	O
,	O
when	O
many	O
leading	O
industry	O
experts	O
were	O
much	O
more	O
pessimistic	O
in	O
their	O
predictions	O
.	O
In	O
the	O
year	O
that	O
Dialog	O
celebrates	O
its	O
thirtieth	O
anniversary	O
as	O
the	O
world	O
's	O
oldest	O
and	O
largest	O
professional	O
online	B-KEY
information	I-KEY
service	I-KEY
,	O
it	O
is	O
appropriate	O
to	O
look	O
back	O
at	O
these	O
changing	O
perceptions	O
,	O
the	O
reasons	O
for	O
these	O
changes	O
,	O
and	O
why	O
the	O
experts	O
got	O
it	O
wrong	O
.	O
We	O
also	O
look	O
at	O
the	O
present	O
day	O
;	O
the	O
value	O
that	O
large-scale	B-KEY
content	I-KEY
aggregators	I-KEY
bring	O
to	O
the	O
information	B-KEY
supply	I-KEY
chain	I-KEY
;	O
and	O
we	O
discuss	O
why	O
users	O
would	O
choose	O
to	O
use	O
aggregators	O
as	O
opposed	O
to	O
going	O
directly	O
to	O
the	O
publishers	O

A	O
summary	O
of	O
methods	O
applied	O
to	O
tool	B-KEY
condition	I-KEY
monitoring	I-KEY
in	O
drilling	B-KEY
Presents	O
a	O
summary	O
of	O
the	O
monitoring	B-KEY
methods	I-KEY
,	O
signal	B-KEY
analysis	I-KEY
and	O
diagnostic	B-KEY
techniques	I-KEY
for	O
tool	B-KEY
wear	I-KEY
and	O
failure	B-KEY
monitoring	I-KEY
in	O
drilling	B-KEY
that	O
have	O
been	O
tested	O
and	O
reported	O
in	O
the	O
literature	O
.	O
The	O
paper	O
covers	O
only	O
indirect	B-KEY
monitoring	I-KEY
methods	I-KEY
such	O
as	O
force	O
,	O
vibration	O
and	O
current	O
measurements	O
.	O
Signal	B-KEY
analysis	I-KEY
techniques	O
cover	O
all	O
the	O
methods	O
that	O
have	O
been	O
used	O
with	O
indirect	O
measurements	O
including	O
e.g.	O
statistical	B-KEY
parameters	I-KEY
and	O
Fast	O
Fourier	O
and	O
Wavelet	B-KEY
Transform	I-KEY
.	O
Only	O
a	O
limited	O
number	O
of	O
automatic	B-KEY
diagnostic	I-KEY
tools	I-KEY
have	O
been	O
developed	O
for	O
diagnosis	O
of	O
the	O
condition	O
of	O
the	O
tool	O
in	O
drilling	B-KEY
.	O
All	O
of	O
these	O
rather	O
diverse	O
approaches	O
that	O
have	O
been	O
available	O
are	O
covered	O
in	O
this	O
study	O
.	O
Only	O
in	O
a	O
few	O
of	O
the	O
papers	O
have	O
attempts	O
been	O
made	O
to	O
compare	O
the	O
chosen	O
approach	O
with	O
other	O
methods	O
.	O
Many	O
of	O
the	O
papers	O
only	O
present	O
one	O
approach	O
and	O
unfortunately	O
quite	O
often	O
the	O
test	O
material	O
of	O
the	O
study	O
is	O
limited	O
especially	O
in	O
what	O
comes	O
to	O
the	O
cutting	O
process	O
parameter	O
variation	O
and	O
also	O
workpiece	O
material	O

An	O
automated	B-KEY
irradiation	I-KEY
device	I-KEY
for	O
use	O
in	O
cyclotrons	B-KEY
Two	O
cyclotrons	B-KEY
are	O
being	O
operated	O
at	O
IPEN-CNEN/SP	O
:	O
one	O
model	O
CV-28	B-KEY
,	O
capable	O
of	O
accelerating	O
protons	B-KEY
with	O
energies	O
up	O
to	O
24	O
MeV	O
and	O
beam	O
currents	O
up	O
to	O
30	O
mu	O
A	O
,	O
and	O
three	O
other	O
particles	O
;	O
the	O
other	O
one	O
,	O
model	O
Cyclone	B-KEY
30	I-KEY
,	O
accelerates	O
protons	B-KEY
with	O
energy	O
of	O
30	O
MeV	O
and	O
currents	O
up	O
to	O
350	O
mu	O
A.	O
Both	O
have	O
the	O
objective	O
of	O
irradiating	O
targets	O
both	O
for	O
radioisotope	B-KEY
production	I-KEY
for	O
use	O
in	O
nuclear	B-KEY
medicine	I-KEY
and	O
general	B-KEY
research	I-KEY
.	O
The	O
development	O
of	O
irradiating	O
systems	O
completely	O
automatized	O
was	O
the	O
objective	O
of	O
this	O
work	O
,	O
always	O
aiming	O
to	O
reduce	O
the	O
radiation	B-KEY
exposition	I-KEY
dose	I-KEY
to	O
the	O
workers	O
and	O
to	O
increase	O
the	O
reliability	O
of	O
use	O
of	O
these	O
systems	O

Gender	O
benders	O
-LSB-	O
women	B-KEY
in	O
computing	B-KEY
profession	I-KEY
-RSB-	O
As	O
a	O
minority	O
in	O
the	O
upper	O
levels	O
of	O
the	O
computing	B-KEY
profession	I-KEY
,	O
women	B-KEY
are	O
sometimes	O
mistreated	O
through	O
ignorance	O
or	O
malice	O
.	O
Some	O
women	B-KEY
have	O
learned	O
to	O
respond	O
with	O
wit	O
and	O
panache	O

Ultra-high	B-KEY
speed	I-KEY
positioning	I-KEY
control	I-KEY
of	O
a	O
gravure	B-KEY
engraving	I-KEY
unit	I-KEY
using	O
a	O
discrete-time	B-KEY
two-degree-of-freedom	I-KEY
H/sub	I-KEY
infinity	I-KEY
/	I-KEY
control	I-KEY
The	O
piezoelectric	B-KEY
actuator	I-KEY
has	O
high-speed	O
response	O
in	O
comparison	O
with	O
the	O
electro-magnetic	O
actuator	O
.	O
However	O
,	O
it	O
is	O
not	O
easy	O
to	O
achieve	O
both	O
high-speed	O
and	O
high-precision	B-KEY
response	I-KEY
by	O
feedforward	B-KEY
control	O
only	O
because	O
the	O
piezoelectric	O
element	O
has	O
nonlinear	B-KEY
properties	I-KEY
such	O
as	O
the	O
hysteresis	B-KEY
effect	O
.	O
Thus	O
,	O
feedback	B-KEY
control	I-KEY
is	O
required	O
to	O
achieve	O
good	O
performance	O
.	O
We	O
develop	O
a	O
control	B-KEY
design	I-KEY
method	I-KEY
to	O
achieve	O
both	O
high-speed	O
and	O
high-precision	B-KEY
response	I-KEY
for	O
piezoelectric	B-KEY
actuators	I-KEY
using	O
the	O
discrete-time	O
H/sub	O
infinity	O
/	O
control	O
method	O
and	O
the	O
two-degree-of-freedom	O
control	O
scheme	O
.	O
The	O
effectiveness	O
of	O
our	O
proposed	O
method	O
has	O
been	O
shown	O
by	O
simulation	O
and	O
experimental	O
results	O
.	O
The	O
most	O
important	O
contribution	O
of	O
our	O
study	O
is	O
that	O
our	O
method	O
can	O
be	O
directly	O
applied	O
to	O
commercial	O
machines	O

Quasi	B-KEY
stage	I-KEY
order	I-KEY
conditions	I-KEY
for	O
SDIRK	B-KEY
methods	I-KEY
The	O
stage	O
order	O
condition	O
is	O
a	O
simplifying	O
assumption	O
that	O
reduces	O
the	O
number	O
of	O
order	O
conditions	O
to	O
be	O
fulfilled	O
when	O
designing	O
a	O
Runge-Kutta	O
-LRB-	O
RK	O
-RRB-	O
method	O
.	O
Because	O
a	O
DIRK	O
-LRB-	O
diagonally	O
implicit	O
RK	O
-RRB-	O
method	O
can	O
not	O
have	O
stage	O
order	O
greater	O
than	O
1	O
,	O
we	O
introduce	O
quasi	B-KEY
stage	I-KEY
order	I-KEY
conditions	I-KEY
and	O
derive	O
some	O
of	O
their	O
properties	O
for	O
DIRKs	O
.	O
We	O
use	O
these	O
conditions	O
to	O
derive	O
a	O
low-order	O
DIRK	O
method	O
with	O
embedded	B-KEY
error	I-KEY
estimator	I-KEY
.	O
Numerical	B-KEY
tests	I-KEY
with	O
stiff	O
ODEs	O
and	O
DAEs	O
of	O
index	O
1	O
and	O
2	O
indicate	O
that	O
the	O
method	O
is	O
competitive	O
with	O
other	O
RK	O
methods	O
for	O
low	O
accuracy	O
tolerances	O

Optical	B-KEY
two-step	I-KEY
modified	I-KEY
signed-digit	I-KEY
addition	I-KEY
based	O
on	O
binary	B-KEY
logic	I-KEY
gates	I-KEY
A	O
new	O
modified	O
signed-digit	O
-LRB-	O
MSD	O
-RRB-	O
addition	O
algorithm	O
based	O
on	O
binary	B-KEY
logic	I-KEY
gates	I-KEY
is	O
proposed	O
for	O
parallel	B-KEY
computing	I-KEY
.	O
It	O
is	O
shown	O
that	O
by	O
encoding	O
each	O
of	O
the	O
input	B-KEY
MSD	I-KEY
digits	I-KEY
and	O
flag	B-KEY
digits	I-KEY
into	O
a	O
pair	O
of	O
binary	B-KEY
bits	I-KEY
,	O
the	O
number	O
of	O
addition	B-KEY
steps	I-KEY
can	O
be	O
reduced	O
to	O
two	O
.	O
The	O
flag	B-KEY
digit	I-KEY
is	O
introduced	O
to	O
characterize	O
the	O
next	O
low	B-KEY
order	I-KEY
pair	I-KEY
-LRB-	O
NLOP	O
-RRB-	O
of	O
the	O
input	O
digits	O
in	O
order	O
to	O
suppress	O
carry	O
propagation	O
.	O
The	O
rules	O
for	O
two-step	B-KEY
addition	I-KEY
of	O
binary	B-KEY
coded	I-KEY
MSD	I-KEY
-LRB-	O
BCMSD	O
-RRB-	O
numbers	O
are	O
formulated	O
that	O
can	O
be	O
implemented	O
using	O
optical	B-KEY
shadow-casting	I-KEY
logic	I-KEY
system	I-KEY

Data	B-KEY
assimilation	I-KEY
of	O
local	B-KEY
model	I-KEY
error	I-KEY
forecasts	I-KEY
in	O
a	O
deterministic	B-KEY
model	I-KEY
One	O
of	O
the	O
most	O
popular	O
data	B-KEY
assimilation	I-KEY
techniques	O
in	O
use	O
today	O
are	O
of	O
the	O
Kalman	B-KEY
filter	I-KEY
type	O
,	O
which	O
provide	O
an	O
improved	O
estimate	O
of	O
the	O
state	O
of	O
a	O
system	O
up	O
to	O
the	O
current	O
time	O
level	O
,	O
based	O
on	O
actual	O
measurements	O
.	O
From	O
a	O
forecasting	O
viewpoint	O
,	O
this	O
corresponds	O
to	O
an	O
updating	O
of	O
the	O
initial	O
conditions	O
.	O
The	O
standard	O
forecasting	O
procedure	O
is	O
to	O
then	O
run	O
the	O
model	O
uncorrected	O
into	O
the	O
future	O
,	O
driven	O
by	O
predicted	O
boundary	O
and	O
forcing	B-KEY
conditions	I-KEY
.	O
The	O
problem	O
with	O
this	O
methodology	O
is	O
that	O
the	O
updated	O
initial	O
conditions	O
quickly	O
`	O
wash-out	O
'	O
,	O
thus	O
,	O
after	O
a	O
certain	O
forecast	B-KEY
horizon	I-KEY
the	O
model	O
predictions	O
are	O
no	O
better	O
than	O
from	O
an	O
initially	O
uncorrected	O
model	O
.	O
This	O
study	O
demonstrates	O
that	O
through	O
the	O
assimilation	O
of	O
error	O
forecasts	O
-LRB-	O
in	O
the	O
present	O
case	O
made	O
using	O
so-called	O
local	O
models	O
-RRB-	O
entire	O
model	O
domains	O
can	O
be	O
corrected	O
for	O
extended	O
forecast	B-KEY
horizons	I-KEY
-LRB-	O
i.e.	O
long	O
after	O
updated	O
initial	O
conditions	O
have	O
become	O
washed-out	O
-RRB-	O
,	O
thus	O
demonstrating	O
significant	O
improvements	O
over	O
the	O
conventional	O
methodology	O
.	O
Some	O
alternate	O
uses	O
of	O
local	O
models	O
are	O
also	O
explored	O
for	O
the	O
re-distribution	O
of	O
error	O
forecasts	O
over	O
the	O
entire	O
model	O
domain	O
,	O
which	O
are	O
then	O
compared	O
with	O
more	O
conventional	O
Kalman	B-KEY
filter	I-KEY
type	O
schemes	O

Summarization	O
beyond	O
sentence	O
extraction	O
:	O
A	O
probabilistic	O
approach	O
to	O
sentence	B-KEY
compression	I-KEY
When	O
humans	O
produce	O
summaries	O
of	O
documents	O
,	O
they	O
do	O
not	O
simply	O
extract	O
sentences	O
and	O
concatenate	O
them	O
.	O
Rather	O
,	O
they	O
create	O
new	O
sentences	O
that	O
are	O
grammatical	B-KEY
,	O
that	O
cohere	O
with	O
one	O
another	O
,	O
and	O
that	O
capture	O
the	O
most	O
salient	O
pieces	O
of	O
information	O
in	O
the	O
original	O
document	O
.	O
Given	O
that	O
large	O
collections	O
of	O
text/abstract	O
pairs	O
are	O
available	O
online	O
,	O
it	O
is	O
now	O
possible	O
to	O
envision	O
algorithms	O
that	O
are	O
trained	O
to	O
mimic	O
this	O
process	O
.	O
In	O
this	O
paper	O
,	O
we	O
focus	O
on	O
sentence	B-KEY
compression	I-KEY
,	O
a	O
simpler	O
version	O
of	O
this	O
larger	O
challenge	O
.	O
We	O
aim	O
to	O
achieve	O
two	O
goals	O
simultaneously	O
:	O
our	O
compressions	O
should	O
be	O
grammatical	B-KEY
,	O
and	O
they	O
should	O
retain	O
the	O
most	O
important	O
pieces	O
of	O
information	O
.	O
These	O
two	O
goals	O
can	O
conflict	O
.	O
We	O
devise	O
both	O
a	O
noisy-channel	B-KEY
and	O
a	O
decision-tree	B-KEY
approach	O
to	O
the	O
problem	O
,	O
and	O
we	O
evaluate	O
results	O
against	O
manual	O
compressions	O
and	O
a	O
simple	O
baseline	O

Cooperative	O
three	O
-	O
and	O
four-player	O
quantum	O
games	O
A	O
cooperative	O
multi-player	O
quantum	O
game	O
played	O
by	O
3	O
and	O
4	O
players	O
has	O
been	O
studied	O
.	O
A	O
quantum	B-KEY
superposed	I-KEY
operator	I-KEY
is	O
introduced	O
in	O
this	O
work	O
which	O
solves	O
the	O
non-zero	O
sum	O
difficulty	O
in	O
previous	O
treatments	O
.	O
The	O
role	O
of	O
quantum	B-KEY
entanglement	I-KEY
of	O
the	O
initial	B-KEY
state	I-KEY
is	O
discussed	O
in	O
detail	O

Optimal	O
bandwidth	O
utilization	O
of	O
all-optical	B-KEY
ring	I-KEY
with	O
a	O
converter	O
of	O
degree	O
4	O
In	O
many	O
models	O
of	O
all-optical	B-KEY
routing	I-KEY
,	O
a	O
set	O
of	O
communication	B-KEY
paths	I-KEY
in	O
a	O
network	O
is	O
given	O
,	O
and	O
a	O
wavelength	O
is	O
to	O
be	O
assigned	O
to	O
each	O
path	O
so	O
that	O
paths	O
sharing	O
an	O
edge	O
receive	O
different	O
wavelengths	O
.	O
The	O
goal	O
is	O
to	O
assign	O
as	O
few	O
wavelengths	O
as	O
possible	O
,	O
in	O
order	O
to	O
use	O
the	O
optical	O
bandwidth	O
efficiently	O
.	O
If	O
a	O
node	O
of	O
a	O
network	O
contains	O
a	O
wavelength	B-KEY
converter	I-KEY
,	O
any	O
path	O
that	O
passes	O
through	O
this	O
node	O
may	O
change	O
its	O
wavelength	O
.	O
Having	O
converters	O
at	O
some	O
of	O
the	O
nodes	O
can	O
reduce	O
the	O
number	O
of	O
wavelengths	O
required	O
for	O
routing	O
.	O
This	O
paper	O
presents	O
a	O
wavelength	B-KEY
converter	I-KEY
with	O
degree	O
4	O
and	O
gives	O
a	O
routing	O
algorithm	O
which	O
shows	O
that	O
any	O
routing	O
with	O
load	O
L	O
can	O
be	O
realized	O
with	O
L	O
wavelengths	O
when	O
a	O
node	O
of	O
an	O
all-optical	B-KEY
ring	I-KEY
hosts	O
such	O
a	O
wavelength	B-KEY
converter	I-KEY
.	O
It	O
is	O
also	O
proved	O
that	O
4	O
is	O
the	O
minimum	O
degree	O
of	O
the	O
converter	O
to	O
reach	O
the	O
full	O
utilization	O
of	O
the	O
available	O
wavelengths	O
if	O
only	O
one	O
node	O
of	O
an	O
all-optical	B-KEY
ring	I-KEY
hosts	O
a	O
converter	O

Virtual	O
reality	O
treatment	O
of	O
flying	B-KEY
phobia	I-KEY
Flying	B-KEY
phobia	I-KEY
-LRB-	O
FP	O
-RRB-	O
might	O
become	O
a	O
very	O
incapacitating	O
and	O
disturbing	O
problem	O
in	O
a	O
person	O
's	O
social	O
,	O
working	O
,	O
and	O
private	O
areas	O
.	O
Psychological	B-KEY
interventions	O
based	O
on	O
exposure	O
therapy	O
have	O
proved	O
to	O
be	O
effective	O
,	O
but	O
given	O
the	O
particular	O
nature	O
of	O
this	O
disorder	O
they	O
bear	O
important	O
limitations	O
.	O
Exposure	B-KEY
therapy	I-KEY
for	O
FP	O
might	O
be	O
excessively	O
costly	O
in	O
terms	O
of	O
time	O
,	O
money	O
,	O
and	O
efforts	O
.	O
Virtual	O
reality	O
-LRB-	O
VR	O
-RRB-	O
overcomes	O
these	O
difficulties	O
as	O
different	O
significant	O
environments	O
might	O
be	O
created	O
,	O
where	O
the	O
patient	O
can	O
interact	O
with	O
what	O
he	O
or	O
she	O
fears	O
while	O
in	O
a	O
totally	O
safe	O
and	O
protected	O
environment	O
,	O
the	O
therapist	O
's	O
consulting	O
room	O
.	O
This	O
paper	O
intends	O
,	O
on	O
one	O
hand	O
,	O
to	O
show	O
the	O
different	O
scenarios	O
designed	O
by	O
our	O
team	O
for	O
the	O
VR	O
treatment	O
of	O
FP	O
,	O
and	O
on	O
the	O
other	O
,	O
to	O
present	O
the	O
first	O
results	O
supporting	O
the	O
effectiveness	O
of	O
this	O
new	O
tool	O
for	O
the	O
treatment	O
of	O
FP	O
in	O
a	O
multiple	O
baseline	O
study	O

Quantum-controlled	B-KEY
measurement	I-KEY
device	I-KEY
for	O
quantum-state	B-KEY
discrimination	I-KEY
We	O
propose	O
a	O
``	O
programmable	O
''	O
quantum	O
device	O
that	O
is	O
able	O
to	O
perform	O
a	O
specific	O
generalized	O
measurement	O
from	O
a	O
certain	O
set	O
of	O
measurements	O
depending	O
on	O
a	O
quantum	B-KEY
state	I-KEY
of	O
a	O
``	O
program	B-KEY
register	I-KEY
.	O
''	O
In	O
particular	O
,	O
we	O
study	O
a	O
situation	O
when	O
the	O
programmable	O
measurement	O
device	O
serves	O
for	O
the	O
unambiguous	O
discrimination	O
between	O
nonorthogonal	B-KEY
states	I-KEY
.	O
The	O
particular	O
pair	O
of	O
states	O
that	O
can	O
be	O
unambiguously	O
discriminated	O
is	O
specified	O
by	O
the	O
state	O
of	O
a	O
program	B-KEY
qubit	I-KEY
.	O
The	O
probability	O
of	O
successful	O
discrimination	O
is	O
not	O
optimal	O
for	O
all	O
admissible	O
pairs	O
.	O
However	O
,	O
for	O
some	O
subsets	O
it	O
can	O
be	O
very	O
close	O
to	O
the	O
optimal	O
value	O

A	O
simple	O
etalon-stabilized	O
visible	B-KEY
laser	I-KEY
diode	I-KEY
Visible	B-KEY
laser	I-KEY
diodes	I-KEY
-LRB-	O
LDs	O
-RRB-	O
are	O
inexpensively	O
available	O
with	O
single-transverse-mode	B-KEY
,	O
single-longitudinal-mode	B-KEY
operation	O
with	O
a	O
coherence	O
length	O
in	O
the	O
metre	O
range	O
.	O
With	O
constant	B-KEY
current	I-KEY
bias	I-KEY
and	O
constant	B-KEY
operating	I-KEY
temperature	I-KEY
,	O
the	O
optical	O
output	O
power	O
and	O
operating	O
wavelength	O
are	O
stable	O
.	O
A	O
simple	O
and	O
inexpensive	O
way	O
is	O
developed	O
to	O
maintain	O
a	O
constant	O
LD	O
temperature	O
as	O
the	O
temperature	O
of	O
the	O
local	O
environment	O
varies	O
,	O
by	O
monitoring	O
the	O
initially	O
changing	O
wavelength	O
with	O
an	O
external	O
etalon	O
and	O
using	O
this	O
information	O
to	O
apply	O
a	O
heating	B-KEY
correction	I-KEY
to	O
the	O
monitor	B-KEY
photodiode	I-KEY
commonly	O
integral	O
to	O
the	O
LD	O
package	O
.	O
The	O
fractional	B-KEY
wavelength	I-KEY
stability	I-KEY
achieved	O
is	O
limited	O
by	O
the	O
solid	O
etalon	O
to	O
7	O
*	O
10/sup	O
-6	O
/	O
degrees	O
C/sup	O
-1	O
/	O

Questioning	O
the	O
RFP	O
process	O
-LSB-	O
telecom	O
-RSB-	O
In	O
the	O
current	O
climate	O
,	O
the	O
most	O
serious	O
concern	O
about	O
the	O
purchasing	O
habits	O
of	O
telecom	B-KEY
carriers	I-KEY
is	O
obviously	O
the	O
lack	O
of	O
spending	O
.	O
Even	O
against	O
a	O
backdrop	O
of	O
economic	O
constraints	O
and	O
financial	O
struggles	O
,	O
however	O
,	O
genuine	O
concerns	O
about	O
the	O
purchasing	B-KEY
process	I-KEY
itself	O
are	O
being	O
raised	O
by	O
some	O
of	O
those	O
closest	O
to	O
it	O

Information	B-KEY
and	O
information	B-KEY
technology	O
This	O
paper	O
reveals	O
the	O
concepts	O
of	O
information	B-KEY
and	O
information	B-KEY
technology	O
.	O
It	O
also	O
describes	O
the	O
close	O
relationship	O
between	O
information	B-KEY
and	O
information	B-KEY
technology	O
.	O
It	O
explains	O
a	O
basic	O
mechanism	O
of	O
different	O
devices	O
of	O
information	B-KEY
technology	O
and	O
connotes	O
how	O
they	O
are	O
useful	O
to	O
store	O
,	O
process	O
and	O
retrieve	O
the	O
information	O
.	O
In	O
addition	O
of	O
this	O
,	O
the	O
paper	O
shows	O
the	O
present	O
status	O
of	O
information	B-KEY
technology	O
and	O
Indian	O
universities	O

A	O
Virtual	B-KEY
Test	I-KEY
Facility	I-KEY
for	O
the	O
simulation	O
of	O
dynamic	O
response	O
in	O
materials	O
The	O
Center	O
for	O
Simulating	O
Dynamic	O
Response	O
of	O
Materials	O
at	O
the	O
California	O
Institute	O
of	O
Technology	O
is	O
constructing	O
a	O
virtual	B-KEY
shock	I-KEY
physics	I-KEY
facility	I-KEY
for	O
studying	O
the	O
response	O
of	O
various	O
target	O
materials	O
to	O
very	O
strong	O
shocks	O
.	O
The	O
Virtual	B-KEY
Test	I-KEY
Facility	I-KEY
-LRB-	O
VTF	O
-RRB-	O
is	O
an	O
end-to-end	O
,	O
fully	O
three-dimensional	O
simulation	O
of	O
the	O
detonation	O
of	O
high	B-KEY
explosives	I-KEY
-LRB-	O
HE	O
-RRB-	O
,	O
shock	B-KEY
wave	I-KEY
propagation	I-KEY
,	O
solid	B-KEY
material	I-KEY
response	I-KEY
to	O
pressure	B-KEY
loading	I-KEY
,	O
and	O
compressible	B-KEY
turbulence	I-KEY
.	O
The	O
VTF	O
largely	O
consists	O
of	O
a	O
parallel	B-KEY
fluid	I-KEY
solver	I-KEY
and	O
a	O
parallel	B-KEY
solid	I-KEY
mechanics	I-KEY
package	O
that	O
are	O
coupled	O
together	O
by	O
the	O
exchange	O
of	O
boundary	O
data	O
.	O
The	O
Eulerian	O
fluid	O
code	O
and	O
Lagrangian	O
solid	O
mechanics	O
model	O
interact	O
via	O
a	O
novel	O
approach	O
based	O
on	O
level	O
sets	O
.	O
The	O
two	O
main	O
computational	O
packages	O
are	O
integrated	O
through	O
the	O
use	O
of	O
Pyre	B-KEY
,	O
a	O
problem	B-KEY
solving	I-KEY
environment	I-KEY
written	O
in	O
the	O
Python	B-KEY
scripting	I-KEY
language	I-KEY
.	O
Pyre	B-KEY
allows	O
application	O
developers	O
to	O
interchange	O
various	O
computational	O
models	O
and	O
solver	O
packages	O
without	O
recompiling	O
code	O
,	O
and	O
it	O
provides	O
standardized	O
access	O
to	O
several	O
data	B-KEY
visualization	I-KEY
engines	O
and	O
data	O
input	O
mechanisms	O
.	O
In	O
this	O
paper	O
,	O
we	O
outline	O
the	O
main	O
components	O
of	O
the	O
VTF	O
,	O
discuss	O
their	O
integration	O
via	O
Pyre	B-KEY
,	O
and	O
describe	O
some	O
recent	O
accomplishments	O
in	O
large-scale	O
simulation	O
using	O
the	O
VTF	O

Car-caravan	B-KEY
snaking	I-KEY
.	O
2	O
Active	B-KEY
caravan	I-KEY
braking	I-KEY
For	O
part	O
1	O
,	O
see	O
ibid.	O
,	O
p.707-22	O
.	O
Founded	O
on	O
the	O
review	O
and	O
results	O
of	O
Part	O
1	O
,	O
Part	O
2	O
contains	O
a	O
description	O
of	O
the	O
virtual	B-KEY
design	I-KEY
of	O
an	O
active	O
braking	O
system	O
for	O
caravans	O
or	O
other	O
types	O
of	O
trailer	B-KEY
,	O
to	O
suppress	O
snaking	O
vibrations	O
,	O
while	O
being	O
simple	O
from	O
a	O
practical	O
viewpoint	O
.	O
The	O
design	O
process	O
and	O
the	O
design	O
itself	O
are	O
explained	O
.	O
The	O
performance	O
is	O
examined	O
by	O
simulations	O
and	O
it	O
is	O
concluded	O
that	O
the	O
system	O
is	O
effective	O
,	O
robust	O
and	O
realizable	O
with	O
modest	O
and	O
available	O
components	O

Silicon	O
debug	O
of	O
a	O
PowerPC	O
TM	O
microprocessor	O
using	O
model	B-KEY
checking	I-KEY
When	O
silicon	O
is	O
available	O
,	O
newly	O
designed	O
microprocessors	O
are	O
tested	O
in	O
specially	O
equipped	O
hardware	O
laboratories	O
,	O
where	O
real	O
applications	O
can	O
be	O
run	O
at	O
hardware	O
speeds	O
.	O
However	O
,	O
the	O
large	O
volumes	O
of	O
code	O
being	O
run	O
,	O
plus	O
the	O
limited	O
access	O
to	O
the	O
internal	O
nodes	O
of	O
the	O
chip	O
,	O
make	O
it	O
very	O
difficult	O
to	O
characterize	O
the	O
nature	O
of	O
any	O
failures	O
that	O
occur	O
.	O
We	O
describe	O
how	O
temporal	B-KEY
logic	I-KEY
model	B-KEY
checking	I-KEY
was	O
used	O
to	O
quickly	O
characterize	O
a	O
design	O
error	O
exhibited	O
during	O
hardware	B-KEY
testing	I-KEY
of	O
a	O
PowerPC	B-KEY
microprocessor	I-KEY
.	O
We	O
outline	O
the	O
conditions	O
under	O
which	O
model	B-KEY
checking	I-KEY
can	O
efficiently	O
characterize	O
such	O
failures	O
,	O
and	O
show	O
how	O
the	O
particular	O
error	O
we	O
detected	O
could	O
have	O
been	O
revealed	O
early	O
in	O
the	O
design	O
cycle	O
,	O
by	O
model	B-KEY
checking	I-KEY
a	O
short	O
and	O
simple	O
correctness	B-KEY
specification	I-KEY
.	O
We	O
discuss	O
the	O
implications	O
of	O
this	O
for	O
verification	B-KEY
methodologies	I-KEY
over	O
the	O
full	O
design	O
cycle	O

Continuous-time	O
linear	B-KEY
systems	I-KEY
:	O
folklore	O
and	O
fact	O
We	O
consider	O
a	O
family	O
of	O
continuous	B-KEY
input-output	I-KEY
maps	I-KEY
representing	O
linear	O
time-invariant	B-KEY
systems	I-KEY
that	O
take	O
a	O
set	O
of	O
signals	O
into	O
itself	O
.	O
It	O
is	O
shown	O
that	O
this	O
family	O
contains	O
maps	O
whose	O
impulse	B-KEY
response	I-KEY
is	O
the	O
zero	B-KEY
function	I-KEY
,	O
but	O
which	O
take	O
certain	O
inputs	O
into	O
nonzero	O
outputs	O
.	O
It	O
is	O
shown	O
also	O
that	O
this	O
family	O
contains	O
members	O
whose	O
input-output	O
properties	O
are	O
not	O
described	O
by	O
their	O
frequency	B-KEY
domain	I-KEY
response	I-KEY
functions	O
,	O
and	O
that	O
the	O
maps	O
considered	O
need	O
not	O
even	O
commute	B-KEY

Automatic	B-KEY
multilevel	I-KEY
thresholding	I-KEY
for	O
image	B-KEY
segmentation	I-KEY
by	O
the	O
growing	B-KEY
time	I-KEY
adaptive	I-KEY
self-organizing	I-KEY
map	I-KEY
In	O
this	O
paper	O
,	O
a	O
Growing	B-KEY
TASOM	I-KEY
-LRB-	O
Time	O
Adaptive	O
Self-Organizing	O
Map	O
-RRB-	O
network	O
called	O
``	O
GTASOM	B-KEY
''	O
along	O
with	O
a	O
peak	B-KEY
finding	I-KEY
process	I-KEY
is	O
proposed	O
for	O
automatic	B-KEY
multilevel	I-KEY
thresholding	I-KEY
.	O
The	O
proposed	O
GTASOM	B-KEY
is	O
tested	O
for	O
image	B-KEY
segmentation	I-KEY
.	O
Experimental	O
results	O
demonstrate	O
that	O
the	O
GTASOM	B-KEY
is	O
a	O
reliable	O
and	O
accurate	O
tool	O
for	O
image	B-KEY
segmentation	I-KEY
and	O
its	O
results	O
outperform	O
other	O
thresholding	O
methods	O

Simulation	O
of	O
physicochemical	O
processes	O
of	O
erosion-corrosion	O
of	O
metals	O
in	O
two-phase	B-KEY
flows	I-KEY
A	O
computational	O
model	O
for	O
the	O
erosion-corrosion	O
of	O
the	O
metals	O
used	O
in	O
power	O
equipment	O
in	O
two-phase	B-KEY
flows	I-KEY
-LRB-	O
RAMEK-2	B-KEY
-RRB-	O
was	O
developed	O
.	O
The	O
results	O
of	O
calculations	O
of	O
the	O
dependency	O
of	O
the	O
intensity	O
of	O
the	O
erosion-corrosion	O
of	O
structural	B-KEY
steels	I-KEY
as	O
a	O
function	O
of	O
the	O
thermodynamic	O
,	O
hydrodynamic	O
and	O
water	O
chemistry	O
parameters	O
of	O
these	O
flows	O
in	O
the	O
working	O
paths	O
of	O
thermal	O
power	O
stations	O
and	O
nuclear	O
power	O
stations	O
are	O
presented	O
in	O
a	O
three-dimensional	B-KEY
space	I-KEY
.	O
On	O
the	O
basis	O
of	O
mathematical	O
models	O
,	O
application	B-KEY
software	I-KEY
was	O
created	O
for	O
forecasting	O
the	O
erosion-corrosion	O
resource	O
and	O
for	O
optimizing	O
the	O
rules	O
on	O
diagnosis	O
and	O
protective	B-KEY
maintenance	I-KEY
of	O
erosion-corrosion	O
of	O
the	O
elements	O
of	O
the	O
wet-steam	B-KEY
path	I-KEY
in	O
power	O
stations	O

Technology	O
decisions	O
2002	O
The	O
paper	O
looks	O
at	O
the	O
critical	O
hardware	O
,	O
software	O
,	O
and	O
services	B-KEY
choices	I-KEY
manufacturers	O
are	O
making	O
as	O
they	O
begin	O
to	O
emerge	O
from	O
the	O
recession	O
and	O
position	O
themselves	O
for	O
the	O
future	O

The	O
analysis	O
and	O
control	O
of	O
longitudinal	O
vibrations	O
from	O
wave	O
viewpoint	O
The	O
analysis	O
and	O
control	O
of	O
longitudinal	O
vibrations	O
in	O
a	O
rod	O
from	O
feedback	B-KEY
wave	I-KEY
viewpoint	O
are	O
synthesized	O
.	O
Both	O
collocated	O
and	O
noncollocated	O
feedback	B-KEY
wave	I-KEY
control	O
strategies	O
are	O
explored	O
.	O
The	O
control	B-KEY
design	I-KEY
is	O
based	O
on	O
the	O
local	O
properties	O
of	O
wave	B-KEY
transmission	I-KEY
and	O
reflection	O
in	O
the	O
vicinity	O
of	O
the	O
control	B-KEY
force	I-KEY
applied	O
area	O
,	O
hence	O
there	O
is	O
no	O
complex	B-KEY
closed	I-KEY
form	I-KEY
solution	I-KEY
involved	O
.	O
The	O
controller	O
is	O
designed	O
to	O
achieve	O
various	O
goals	O
,	O
such	O
as	O
absorbing	O
the	O
incoming	O
vibration	B-KEY
energy	I-KEY
,	O
creating	O
a	O
vibration	B-KEY
free	I-KEY
zone	I-KEY
and	O
eliminating	O
standing	B-KEY
waves	I-KEY
in	O
the	O
structure	O
.	O
The	O
findings	O
appear	O
to	O
be	O
very	O
useful	O
in	O
practice	O
due	O
to	O
the	O
simplicity	O
in	O
the	O
implementation	O
of	O
the	O
controllers	O

Construction	O
of	O
two-sided	B-KEY
bounds	I-KEY
for	O
initial-boundary	B-KEY
value	I-KEY
problems	I-KEY
This	O
paper	O
extends	O
the	O
bounding	B-KEY
operator	I-KEY
approach	O
developed	O
for	O
boundary	O
value	O
problems	O
to	O
the	O
case	O
of	O
initial-boundary	O
value	O
problems	O
-LRB-	O
IBVPs	O
-RRB-	O
.	O
Following	O
the	O
general	O
principle	O
of	O
bounding	B-KEY
operators	I-KEY
enclosing	O
methods	O
for	O
the	O
case	O
of	O
partial	B-KEY
differential	I-KEY
equations	I-KEY
are	O
discussed	O
.	O
In	O
particular	O
,	O
continuous	O
discretization	O
methods	O
with	O
an	O
appropriate	O
error	O
bound	O
controlled	O
shift	O
and	O
monotone	O
extensions	O
of	O
Rothe	O
's	O
method	O
for	O
parabolic	B-KEY
problems	I-KEY
are	O
investigated	O

Keeping	O
Web	O
accessibility	O
in	O
mind	O
:	O
I&R	O
services	O
for	O
all	O
After	O
presenting	O
three	O
compelling	O
reasons	O
for	O
making	O
Web	B-KEY
sites	I-KEY
accessible	I-KEY
to	O
persons	O
with	O
a	O
broad	O
range	O
of	O
disabilities	B-KEY
-LRB-	O
it	O
's	O
the	O
morally	O
right	O
thing	O
to	O
do	O
,	O
it	O
's	O
the	O
smart	O
thing	O
to	O
do	O
from	O
an	O
economic	O
perspective	O
,	O
and	O
it	O
's	O
required	O
by	O
law	O
-RRB-	O
,	O
the	O
author	O
discusses	O
design	O
issues	O
that	O
impact	O
persons	O
with	O
particular	O
types	O
of	O
disabilities	B-KEY
.	O
She	O
presents	O
practical	O
advice	O
for	O
assessing	O
and	O
addressing	O
accessibility	O
problems	O
.	O
An	O
extensive	O
list	O
of	O
resources	O
for	O
further	O
information	O
is	O
appended	O
,	O
as	O
is	O
a	O
list	O
of	O
sites	O
which	O
simulate	O
the	O
impact	O
of	O
specific	O
accessibility	O
problems	O
on	O
persons	O
with	O
disabilities	B-KEY

Blended	B-KEY
implementation	I-KEY
of	O
block	B-KEY
implicit	I-KEY
methods	I-KEY
for	O
ODEs	B-KEY
In	O
this	O
paper	O
we	O
further	O
develop	O
a	O
new	O
approach	O
for	O
naturally	O
defining	O
the	O
nonlinear	B-KEY
splittings	I-KEY
needed	O
for	O
the	O
implementation	O
of	O
block	B-KEY
implicit	I-KEY
methods	I-KEY
for	O
ODEs	B-KEY
,	O
which	O
has	O
been	O
considered	O
by	O
Brugnano	O
-LSB-	O
J.	O
Comput	O
.	O
Appl	O
.	O
Math	O
.	O
116	O
-LRB-	O
2000	O
-RRB-	O
41	O
-RSB-	O
and	O
by	O
Brugnano	O
and	O
Trigiante	O
-LSB-	O
in	O
:	O
Recent	O
Trends	O
in	O
Numerical	O
Analysis	O
,	O
Nova	O
Science	O
,	O
New	O
York	O
,	O
2000	O
,	O
pp.	O
81-105	O
-RSB-	O
.	O
The	O
basic	O
idea	O
is	O
that	O
of	O
defining	O
the	O
numerical	B-KEY
method	I-KEY
as	O
the	O
combination	O
-LRB-	O
blending	O
-RRB-	O
of	O
two	O
suitable	O
component	O
methods	O
.	O
By	O
carefully	O
choosing	O
such	O
methods	O
,	O
it	O
is	O
shown	O
that	O
very	O
efficient	O
implementations	O
can	O
be	O
obtained	O
.	O
Moreover	O
,	O
some	O
of	O
them	O
,	O
characterized	O
by	O
a	O
diagonal	B-KEY
splitting	I-KEY
,	O
are	O
well	O
suited	O
for	O
parallel	B-KEY
computers	I-KEY
.	O
Numerical	B-KEY
tests	I-KEY
comparing	O
the	O
performances	O
of	O
the	O
proposed	O
implementation	O
with	O
existing	O
ones	O
are	O
also	O
presented	O
,	O
in	O
order	O
to	O
make	O
evident	O
the	O
potential	O
of	O
the	O
approach	O

A	O
modified	B-KEY
Fieller	I-KEY
interval	I-KEY
for	O
the	O
interval	B-KEY
estimation	I-KEY
of	O
effective	B-KEY
doses	I-KEY
for	O
a	O
logistic	B-KEY
dose-response	I-KEY
curve	I-KEY
Interval	B-KEY
estimation	I-KEY
of	O
the	O
gamma	O
%	O
effective	B-KEY
dose	I-KEY
-LRB-	O
mu	O
/	O
sub	O
gamma	O
/	O
say	O
-RRB-	O
is	O
often	O
based	O
on	O
the	O
asymptotic	B-KEY
variance	I-KEY
of	O
the	O
maximum	B-KEY
likelihood	I-KEY
estimator	I-KEY
-LRB-	O
delta	B-KEY
interval	I-KEY
-RRB-	O
or	O
Fieller	O
's	O
theorem	O
-LRB-	O
Fieller	O
interval	O
-RRB-	O
.	O
Sitter	O
and	O
Wu	O
-LRB-	O
1993	O
-RRB-	O
compared	O
the	O
delta	O
and	O
Fieller	O
intervals	O
for	O
the	O
median	B-KEY
effective	I-KEY
dose	I-KEY
-LRB-	O
mu	O
/	O
sub	O
50	O
/	O
-RRB-	O
assuming	O
a	O
logistic	O
dose-response	O
curve	O
.	O
Their	O
results	O
indicated	O
that	O
although	O
Fieller	O
intervals	O
are	O
generally	O
superior	O
to	O
delta	B-KEY
intervals	I-KEY
,	O
they	O
appear	O
to	O
be	O
conservative	O
.	O
Here	O
an	O
adjusted	O
form	O
of	O
the	O
Fieller	O
interval	O
for	O
mu	O
/	O
sub	O
gamma	O
/	O
termed	O
an	O
adjusted	O
Fieller	O
-LRB-	O
AF	O
-RRB-	O
interval	O
is	O
introduced	O
.	O
A	O
comparison	O
of	O
the	O
AF	O
interval	O
with	O
the	O
delta	O
and	O
Fieller	O
intervals	O
is	O
provided	O
and	O
the	O
properties	O
of	O
these	O
three	O
interval	B-KEY
estimation	I-KEY
methods	O
are	O
investigated	O

Supporting	O
unified	B-KEY
interface	I-KEY
to	O
wrapper	B-KEY
generator	I-KEY
in	O
integrated	B-KEY
information	I-KEY
retrieval	I-KEY
Given	O
the	O
ever-increasing	O
scale	O
and	O
diversity	O
of	O
information	O
and	O
applications	O
on	O
the	O
Internet	B-KEY
,	O
improving	O
the	O
technology	O
of	O
information	O
retrieval	O
is	O
an	O
urgent	O
research	O
objective	O
.	O
Retrieved	O
information	O
is	O
either	O
semi-structured	O
or	O
unstructured	O
in	O
format	O
and	O
its	O
sources	O
are	O
extremely	O
heterogeneous	O
.	O
In	O
consequence	O
,	O
the	O
task	O
of	O
efficiently	O
gathering	O
and	O
extracting	O
information	O
from	O
documents	O
can	O
be	O
both	O
difficult	O
and	O
tedious	O
.	O
Given	O
this	O
variety	O
of	O
sources	O
and	O
formats	O
,	O
many	O
choose	O
to	O
use	O
mediator/wrapper	O
architecture	O
,	O
but	O
its	O
use	O
demands	O
a	O
fast	O
means	O
of	O
generating	O
efficient	O
wrappers	O
.	O
In	O
this	O
paper	O
,	O
we	O
present	O
a	O
design	O
for	O
an	O
automatic	B-KEY
eXtensible	I-KEY
Markup	I-KEY
Language	I-KEY
-LRB-	O
XML	O
-RRB-	O
-	O
based	O
framework	O
with	O
which	O
to	O
generate	O
wrappers	O
rapidly	O
.	O
Wrappers	O
created	O
with	O
this	O
framework	O
support	O
a	O
unified	B-KEY
interface	I-KEY
for	O
a	O
meta-search	O
information	O
retrieval	O
system	O
based	O
on	O
the	O
Internet	B-KEY
Search	O
Service	O
using	O
the	O
Common	O
Object	O
Request	O
Broker	O
Architecture	O
-LRB-	O
CORBA	B-KEY
-RRB-	O
standard	O
.	O
Greatly	O
advantaged	O
by	O
the	O
compatibility	O
of	O
CORBA	B-KEY
and	O
XML	O
,	O
a	O
user	O
can	O
quickly	O
and	O
easily	O
develop	O
information-gathering	O
applications	O
,	O
such	O
as	O
a	O
meta-search	B-KEY
engine	I-KEY
or	O
any	O
other	O
information	O
source	O
retrieval	O
method	O
.	O
The	O
two	O
main	O
things	O
our	O
design	O
provides	O
are	O
a	O
method	O
of	O
wrapper	B-KEY
generation	I-KEY
that	O
is	O
fast	O
,	O
simple	O
,	O
and	O
efficient	O
,	O
and	O
a	O
wrapper	B-KEY
generator	I-KEY
that	O
is	O
CORBA	B-KEY
and	O
XML-compliant	O
and	O
that	O
supports	O
a	O
unified	B-KEY
interface	I-KEY

Measuring	O
return	O
:	O
revealing	O
ROI	B-KEY
The	O
most	O
critical	O
part	O
of	O
the	O
return-on-investment	B-KEY
odyssey	O
is	O
to	O
develop	O
metrics	O
that	O
matter	O
to	O
the	O
business	O
and	O
to	O
measure	O
systems	O
in	O
terms	O
of	O
their	O
ability	O
to	O
help	O
achieve	O
those	O
business	O
goals	O
.	O
Everything	O
must	O
flow	O
from	O
those	O
key	B-KEY
metrics	I-KEY
.	O
And	O
do	O
n't	O
forget	O
to	O
revisit	O
those	O
every	O
now	O
and	O
then	O
,	O
too	O
.	O
Since	O
all	O
systems	O
wind	O
down	O
over	O
time	O
,	O
it	O
's	O
important	O
to	O
keep	O
tabs	O
on	O
how	O
well	O
your	O
automation	B-KEY
investment	I-KEY
is	O
meeting	O
the	O
metrics	O
established	O
by	O
your	O
company	O
.	O
Manufacturers	O
are	O
clamoring	O
for	O
a	O
tool	O
to	O
help	O
quantify	O
returns	O
and	O
analyze	O
the	O
results	O

Schema	B-KEY
evolution	I-KEY
in	O
data	B-KEY
warehouses	I-KEY
We	O
address	O
the	O
issues	O
related	O
to	O
the	O
evolution	O
and	O
maintenance	O
of	O
data	O
warehousing	O
systems	O
,	O
when	O
underlying	O
data	B-KEY
sources	I-KEY
change	O
their	O
schema	O
capabilities	O
.	O
These	O
changes	O
can	O
invalidate	O
views	O
at	O
the	O
data	O
warehousing	O
system	O
.	O
We	O
present	O
an	O
approach	O
for	O
dynamically	O
adapting	O
views	O
according	O
to	O
schema	O
changes	O
arising	O
on	O
source	B-KEY
relations	I-KEY
.	O
This	O
type	O
of	O
maintenance	O
concerns	O
both	O
the	O
schema	O
and	O
the	O
data	O
of	O
the	O
data	B-KEY
warehouse	I-KEY
.	O
The	O
main	O
issue	O
is	O
to	O
avoid	O
the	O
view	O
recomputation	O
from	O
scratch	O
especially	O
when	O
views	O
are	O
defined	O
from	O
multiple	O
sources	O
.	O
The	O
data	O
of	O
the	O
data	B-KEY
warehouse	I-KEY
is	O
used	O
primarily	O
in	O
organizational	B-KEY
decision-making	I-KEY
and	O
may	O
be	O
strategic	O
.	O
Therefore	O
,	O
the	O
schema	O
of	O
the	O
data	B-KEY
warehouse	I-KEY
can	O
evolve	O
for	O
modeling	O
new	O
requirements	O
resulting	O
from	O
analysis	O
or	O
data-mining	O
processing	O
.	O
Our	O
approach	O
provides	O
means	O
to	O
support	O
schema	B-KEY
evolution	I-KEY
of	O
the	O
data	B-KEY
warehouse	I-KEY
independently	O
of	O
the	O
data	B-KEY
sources	I-KEY

Law	O
librarians	O
'	O
survey	B-KEY
:	O
are	O
academic	B-KEY
law	I-KEY
librarians	I-KEY
in	O
decline	O
?	O
The	O
author	O
reports	O
on	O
the	O
results	O
of	O
one	O
extra	O
element	O
in	O
the	O
BIALL/SPTL	B-KEY
survey	B-KEY
,	O
designed	O
to	O
acquire	O
further	O
information	O
about	O
academic	B-KEY
law	I-KEY
librarians	I-KEY
.	O
The	O
survey	B-KEY
has	O
fulfilled	O
the	O
aim	O
of	O
providing	O
a	O
snapshot	O
of	O
the	O
academic	B-KEY
law	I-KEY
library	I-KEY
profession	O
and	O
has	O
examined	O
the	O
concerns	O
that	O
have	O
been	O
raised	O
.	O
Perhaps	O
most	O
importantly	O
,	O
it	O
has	O
shown	O
that	O
more	O
long-term	O
work	O
needs	O
to	O
be	O
done	O
to	O
monitor	O
the	O
situation	O
effectively	O
.	O
We	O
hope	O
that	O
BIALL	O
will	O
take	O
on	O
this	O
challenge	O
and	O
help	O
to	O
maintain	O
the	O
status	O
of	O
academic	B-KEY
law	I-KEY
librarians	I-KEY
and	O
aid	O
them	O
in	O
their	O
work	O

Quantum	B-KEY
computing	I-KEY
with	O
spin	B-KEY
qubits	I-KEY
in	O
semiconductor	B-KEY
structures	I-KEY
We	O
survey	O
recent	O
work	O
on	O
designing	O
and	O
evaluating	O
quantum	B-KEY
computing	I-KEY
implementations	O
based	O
on	O
nuclear	O
or	O
bound-electron	O
spins	O
in	O
semiconductor	B-KEY
heterostructures	I-KEY
at	O
low	B-KEY
temperatures	I-KEY
and	O
in	O
high	B-KEY
magnetic	I-KEY
fields	I-KEY
.	O
General	O
overview	O
is	O
followed	O
by	O
a	O
summary	O
of	O
results	O
of	O
our	O
theoretical	O
calculations	O
of	O
decoherence	O
time	O
scales	O
and	O
spin-spin	B-KEY
interactions	I-KEY
.	O
The	O
latter	O
were	O
carried	O
out	O
for	O
systems	O
for	O
which	O
the	O
two-dimensional	O
electron	O
gas	O
provides	O
the	O
dominant	B-KEY
carrier	I-KEY
for	O
spin	B-KEY
dynamics	I-KEY
via	O
exchange	O
of	O
spin-excitons	O
in	O
the	O
integer	B-KEY
quantum	I-KEY
Hall	I-KEY
regime	I-KEY

VPP	B-KEY
Fortran	I-KEY
and	O
the	O
design	O
of	O
HPF/JA	O
extensions	O
VPP	B-KEY
Fortran	I-KEY
is	O
a	O
data	B-KEY
parallel	I-KEY
language	O
that	O
has	O
been	O
designed	O
for	O
the	O
VPP	O
series	O
of	O
supercomputers	O
.	O
In	O
addition	O
to	O
pure	O
data	B-KEY
parallelism	I-KEY
,	O
it	O
contains	O
certain	O
low-level	O
features	O
that	O
were	O
designed	O
to	O
extract	O
high	B-KEY
performance	I-KEY
from	O
user	O
programs	O
.	O
A	O
comparison	O
of	O
VPP	B-KEY
Fortran	I-KEY
and	O
High-Performance	O
Fortran	O
-LRB-	O
HPF	O
-RRB-	O
2.0	O
shows	O
that	O
these	O
low-level	O
features	O
are	O
not	O
available	O
in	O
HPF	O
2.0	O
.	O
The	O
features	O
include	O
asynchronous	B-KEY
interprocessor	I-KEY
communication	I-KEY
,	O
explicit	B-KEY
shadow	I-KEY
,	O
and	O
the	O
LOCAL	O
directive	O
.	O
They	O
were	O
shown	O
in	O
VPP	B-KEY
Fortran	I-KEY
to	O
be	O
very	O
useful	O
in	O
handling	O
real-world	O
applications	O
,	O
and	O
they	O
have	O
been	O
included	O
in	O
the	O
HPF/JA	O
extensions	O
.	O
They	O
are	O
described	O
in	O
the	O
paper	O
.	O
The	O
HPF/JA	O
Language	O
Specification	O
Version	O
1.0	O
is	O
an	O
extension	O
of	O
HPF	O
2.0	O
to	O
achieve	O
practical	O
performance	O
for	O
real-world	O
applications	O
and	O
is	O
a	O
result	O
of	O
collaboration	O
in	O
the	O
Japan	O
Association	O
for	O
HPF	O
-LRB-	O
JAHPF	O
-RRB-	O
.	O
Some	O
practical	O
programming	O
and	O
tuning	O
procedures	O
with	O
the	O
HPF/JA	O
Language	O
Specification	O
are	O
described	O
,	O
using	O
the	O
NAS	O
Parallel	O
Benchmark	B-KEY
BT	O
as	O
an	O
example	O

Comparison	O
of	O
non-stationary	O
time	O
series	O
in	O
the	O
frequency	O
domain	O
In	O
this	O
paper	O
we	O
compare	O
two	O
nonstationary	B-KEY
time	I-KEY
series	I-KEY
using	O
nonparametric	B-KEY
procedures	I-KEY
.	O
Evolutionary	O
spectra	O
are	O
estimated	O
for	O
the	O
two	O
series	O
.	O
Randomization	B-KEY
tests	I-KEY
are	O
performed	O
on	O
groups	O
of	O
spectral	B-KEY
estimates	I-KEY
for	O
both	O
related	O
and	O
independent	B-KEY
time	I-KEY
series	I-KEY
.	O
Simulation	B-KEY
studies	O
show	O
that	O
in	O
certain	O
cases	O
the	O
tests	O
perform	O
reasonably	O
well	O
.	O
The	O
tests	O
are	O
applied	O
to	O
observed	O
geological	O
and	O
financial	B-KEY
time	I-KEY
series	I-KEY

An	O
entanglement	B-KEY
measure	I-KEY
based	O
on	O
the	O
capacity	O
of	O
dense	O
coding	O
An	O
asymptotic	B-KEY
entanglement	I-KEY
measure	I-KEY
for	O
any	O
bipartite	O
states	O
is	O
derived	O
in	O
the	O
light	O
of	O
the	O
dense	O
coding	O
capacity	O
optimized	O
with	O
respect	O
to	O
local	O
quantum	O
operations	O
and	O
classical	O
communications	O
.	O
General	O
properties	O
and	O
some	O
examples	O
with	O
explicit	O
forms	O
of	O
this	O
entanglement	B-KEY
measure	I-KEY
are	O
investigated	O

Evaluating	O
alternative	B-KEY
manufacturing	I-KEY
control	I-KEY
strategies	I-KEY
using	O
a	O
benchmark	B-KEY
system	I-KEY
This	O
paper	O
describes	O
an	O
investigation	O
of	O
the	O
effects	O
of	O
dynamic	B-KEY
job	I-KEY
routing	I-KEY
and	O
job	B-KEY
sequencing	I-KEY
decisions	I-KEY
on	O
the	O
performance	O
of	O
a	O
distributed	B-KEY
control	I-KEY
system	I-KEY
and	O
its	O
adaptability	O
against	O
disturbances	O
.	O
This	O
experimental	O
work	O
was	O
carried	O
out	O
to	O
compare	O
the	O
performance	O
of	O
alternative	O
control	O
strategies	O
in	O
various	O
manufacturing	O
environments	O
and	O
to	O
investigate	O
the	O
relationship	O
between	O
the	O
`	O
control	O
'	O
and	O
`	O
controlled	O
'	O
systems	O
.	O
The	O
experimental	B-KEY
test-bed	I-KEY
presented	O
in	O
this	O
paper	O
consists	O
of	O
an	O
agent-based	B-KEY
control	I-KEY
system	I-KEY
-LRB-	O
implemented	O
in	O
C++	O
-RRB-	O
and	O
a	O
discrete-event	B-KEY
simulation	I-KEY
model	I-KEY
.	O
Using	O
this	O
test-bed	O
,	O
various	O
control	O
strategies	O
were	O
tested	O
on	O
a	O
benchmark	B-KEY
manufacturing	I-KEY
system	I-KEY
by	O
varying	O
production	B-KEY
volumes	I-KEY
-LRB-	O
to	O
model	O
the	O
production	O
system	O
with	O
looser/tighter	O
schedules	O
-RRB-	O
and	O
disturbance	B-KEY
frequencies	I-KEY
.	O
It	O
was	O
found	O
that	O
hybrid	B-KEY
strategies	I-KEY
that	O
combine	O
reactive	B-KEY
agent	I-KEY
mechanisms	I-KEY
-LRB-	O
and	O
allocation	B-KEY
strategies	I-KEY
such	O
as	O
the	O
contract	B-KEY
net	I-KEY
-RRB-	O
with	O
appropriate	O
job	O
sequencing	O
heuristics	O
provide	O
the	O
best	O
performance	O
,	O
particularly	O
when	O
job	B-KEY
congestion	I-KEY
increases	O
on	O
a	O
shop-floor	O

Is	O
open	O
source	O
more	O
or	O
less	O
secure	O
?	O
Networks	O
dominate	O
today	O
's	O
computing	O
landscape	O
and	O
commercial	B-KEY
technical	I-KEY
protection	I-KEY
is	O
lagging	O
behind	O
attack	B-KEY
technology	I-KEY
.	O
As	O
a	O
result	O
,	O
protection	O
programme	O
success	O
depends	O
more	O
on	O
prudent	O
management	B-KEY
decisions	O
than	O
on	O
the	O
selection	O
of	O
technical	O
safeguards	O
.	O
The	O
paper	O
takes	O
a	O
management	B-KEY
view	O
of	O
protection	O
and	O
seeks	O
to	O
reconcile	O
the	O
need	O
for	O
security	O
with	O
the	O
limitations	O
of	O
technology	O

A	O
novel	O
robot	O
hand	O
with	O
embedded	B-KEY
shape	I-KEY
memory	I-KEY
alloy	I-KEY
actuators	I-KEY
Describes	O
the	O
development	O
of	O
an	O
active	B-KEY
robot	I-KEY
hand	I-KEY
,	O
which	O
allows	O
smooth	O
and	O
lifelike	B-KEY
motions	I-KEY
for	O
anthropomorphic	B-KEY
grasping	I-KEY
and	O
fine	B-KEY
manipulations	I-KEY
.	O
An	O
active	O
robot	O
finger	O
10	O
mm	O
in	O
outer	O
diameter	O
with	O
a	O
shape	O
memory	O
alloy	O
-LRB-	O
SMA	O
-RRB-	O
wire	O
actuator	O
embedded	O
in	O
the	O
finger	O
with	O
a	O
constant	O
distance	O
from	O
the	O
geometric	O
centre	O
of	O
the	O
finger	O
was	O
designed	O
and	O
fabricated	O
.	O
The	O
practical	O
specifications	O
of	O
the	O
SMA	O
wire	O
and	O
the	O
flexible	B-KEY
rod	I-KEY
were	O
determined	O
on	O
the	O
basis	O
of	O
a	O
series	O
of	O
formulae	O
.	O
The	O
active	B-KEY
finger	I-KEY
consists	O
of	O
two	O
bending	O
parts	O
,	O
the	O
SMA	O
actuators	O
and	O
a	O
connecting	O
part	O
.	O
The	O
mechanical	O
properties	O
of	O
the	O
bending	O
part	O
are	O
investigated	O
.	O
The	O
control	O
system	O
on	O
the	O
basis	O
of	O
resistance	B-KEY
feedback	I-KEY
is	O
also	O
presented	O
.	O
Finally	O
,	O
a	O
robot	O
hand	O
with	O
three	O
fingers	O
was	O
designed	O
and	O
the	O
grasping	O
experiment	O
was	O
carried	O
out	O
to	O
demonstrate	O
its	O
performance	O

Online	B-KEY
coverage	I-KEY
of	O
the	O
Olympic	B-KEY
Games	I-KEY
In	O
1956	O
a	O
new	O
medium	O
was	O
evolving	O
which	O
helped	O
shape	O
not	O
only	O
the	O
presentation	O
of	O
the	O
Games	O
to	O
a	O
worldwide	O
audience	O
,	O
but	O
created	O
entirely	O
new	O
avenues	O
for	O
marketing	B-KEY
and	O
sponsorship	B-KEY
which	O
changed	O
the	O
entire	O
economic	B-KEY
relevance	I-KEY
of	O
the	O
Games	O
.	O
The	O
medium	O
in	O
1956	O
was	O
television	O
,	O
and	O
the	O
medium	O
now	O
,	O
of	O
course	O
,	O
is	O
the	O
Internet	O
.	O
Not	O
since	O
1956	O
has	O
Olympic	O
coverage	O
been	O
so	O
impacted	O
by	O
the	O
onset	O
of	O
new	O
technology	O
as	O
the	O
current	O
Olympiad	B-KEY
has	O
been	O
.	O
But	O
now	O
the	O
IOC	B-KEY
finds	O
itself	O
in	O
another	O
set	O
of	O
circumstances	O
not	O
altogether	O
different	O
from	O
1956	O

IT	O
at	O
the	O
heart	O
of	O
joined-up	O
policing	O
Police	B-KEY
IT	I-KEY
is	O
to	O
shift	O
from	O
application-focused	O
to	O
component-based	O
technology	O
.	O
The	O
change	O
of	O
strategy	O
,	O
part	O
of	O
the	O
Valiant	B-KEY
Programme	I-KEY
,	O
will	O
make	O
information	O
held	O
by	O
individual	O
forces	O
available	O
on	O
a	O
national	O
basis	O

Generalized	B-KEY
mosaicing	I-KEY
:	O
wide	B-KEY
field	I-KEY
of	I-KEY
view	I-KEY
multispectral	I-KEY
imaging	I-KEY
We	O
present	O
an	O
approach	O
to	O
significantly	O
enhance	O
the	O
spectral	O
resolution	O
of	O
imaging	O
systems	O
by	O
generalizing	O
image	O
mosaicing	O
.	O
A	O
filter	O
transmitting	O
spatially	B-KEY
varying	I-KEY
spectral	I-KEY
bands	I-KEY
is	O
rigidly	O
attached	O
to	O
a	O
camera	O
.	O
As	O
the	O
system	O
moves	O
,	O
it	O
senses	O
each	O
scene	O
point	O
multiple	O
times	O
,	O
each	O
time	O
in	O
a	O
different	O
spectral	O
band	O
.	O
This	O
is	O
an	O
additional	O
dimension	O
of	O
the	O
generalized	B-KEY
mosaic	I-KEY
paradigm	O
,	O
which	O
has	O
demonstrated	O
yielding	O
high	O
radiometric	O
dynamic	O
range	O
images	O
in	O
a	O
wide	O
field	O
of	O
view	O
,	O
using	O
a	O
spatially	B-KEY
varying	I-KEY
density	I-KEY
filter	I-KEY
.	O
The	O
resulting	O
mosaic	O
represents	O
the	O
spectrum	O
at	O
each	O
scene	O
point	O
.	O
The	O
image	B-KEY
acquisition	I-KEY
is	O
as	O
easy	O
as	O
in	O
traditional	O
image	O
mosaics	O
.	O
We	O
derive	O
an	O
efficient	O
scene	B-KEY
sampling	I-KEY
rate	I-KEY
,	O
and	O
use	O
a	O
registration	B-KEY
method	I-KEY
that	O
accommodates	O
the	O
spatially	O
varying	O
properties	O
of	O
the	O
filter	O
.	O
Using	O
the	O
data	O
acquired	O
by	O
this	O
method	O
,	O
we	O
demonstrate	O
scene	B-KEY
rendering	I-KEY
under	O
different	O
simulated	B-KEY
illumination	I-KEY
spectra	I-KEY
.	O
We	O
are	O
also	O
able	O
to	O
infer	O
information	O
about	O
the	O
scene	B-KEY
illumination	I-KEY
.	O
The	O
approach	O
was	O
tested	O
using	O
a	O
standard	O
8-bit	O
black/white	O
video	O
camera	O
and	O
a	O
fixed	O
spatially	O
varying	O
spectral	O
-LRB-	O
interference	O
-RRB-	O
filter	O

A	O
knowledge-based	B-KEY
approach	I-KEY
for	O
business	B-KEY
process	I-KEY
reengineering	I-KEY
,	O
SHAMASH	B-KEY
We	O
present	O
an	O
overview	O
of	O
SHAMASH	B-KEY
,	O
a	O
process	B-KEY
modelling	I-KEY
tool	I-KEY
for	O
business	B-KEY
process	I-KEY
reengineering	I-KEY
.	O
The	O
main	O
features	O
that	O
differentiate	O
it	O
from	O
most	O
current	O
related	O
tools	O
are	O
its	O
ability	O
to	O
define	O
and	O
use	O
organisation	B-KEY
standards	I-KEY
,	O
functional	B-KEY
structure	I-KEY
,	O
and	O
develop	O
automatic	B-KEY
model	I-KEY
simulation	I-KEY
and	O
optimisation	B-KEY
.	O
SHAMASH	B-KEY
is	O
a	O
knowledge-based	B-KEY
system	I-KEY
,	O
and	O
we	O
include	O
a	O
discussion	O
on	O
how	O
knowledge	B-KEY
acquisition	I-KEY
takes	O
place	O
.	O
Furthermore	O
,	O
we	O
introduce	O
a	O
high	O
level	O
description	O
of	O
the	O
architecture	O
,	O
the	O
conceptual	B-KEY
model	I-KEY
,	O
and	O
other	O
important	O
modules	O
of	O
the	O
system	O

Induced-shear	O
piezoelectric	O
actuators	O
for	O
rotor	O
blade	O
trailing	O
edge	O
flaps	O
Much	O
of	O
the	O
current	O
rotorcraft	B-KEY
research	O
is	O
focused	O
on	O
improving	O
performance	O
by	O
reducing	O
unwanted	O
helicopter	B-KEY
noise	I-KEY
and	O
vibration	O
.	O
One	O
of	O
the	O
most	O
promising	O
active	O
rotorcraft	B-KEY
vibration	B-KEY
control	I-KEY
systems	O
is	O
an	O
active	B-KEY
trailing	I-KEY
edge	I-KEY
flap	I-KEY
.	O
In	O
this	O
paper	O
,	O
an	O
induced-shear	O
piezoelectric	B-KEY
tube	I-KEY
actuator	I-KEY
is	O
used	O
in	O
conjunction	O
with	O
a	O
simple	O
lever-cusp	B-KEY
hinge	I-KEY
amplification	I-KEY
device	I-KEY
to	O
generate	O
a	O
useful	O
combination	O
of	O
trailing	O
edge	O
flap	O
deflections	O
and	O
hinge	O
moments	O
.	O
A	O
finite-element	B-KEY
model	I-KEY
of	O
the	O
actuator	O
tube	O
and	O
trailing	O
edge	O
flap	O
-LRB-	O
including	O
aerodynamic	O
and	O
inertial	B-KEY
loading	I-KEY
-RRB-	O
was	O
used	O
to	O
guide	O
the	O
design	B-KEY
of	O
the	O
actuator-flap	O
system	O
.	O
A	O
full-scale	O
induced	O
shear	B-KEY
tube	I-KEY
actuator	I-KEY
flap	I-KEY
system	O
was	O
fabricated	O
and	O
bench	B-KEY
top	I-KEY
testing	I-KEY
was	O
conducted	O
to	O
validate	O
the	O
analysis	O
.	O
Hinge	O
moments	O
corresponding	O
to	O
various	O
rotor	O
speeds	O
were	O
applied	O
to	O
the	O
actuator	O
using	O
mechanical	O
springs	O
.	O
The	O
testing	O
demonstrated	O
that	O
for	O
an	O
applied	O
electric	O
field	O
of	O
3	O
kV	O
cm/sup	O
-1	O
/	O
the	O
tube	O
actuator	O
deflected	O
a	O
representative	O
full-scale	O
12	B-KEY
inch	I-KEY
flap	O
+	O
or-2	O
.8	O
degrees	O
at	O
0	O
rpm	O
and	O
+	O
or-1	O
.4	O
degrees	O
for	O
a	O
hinge	O
moment	O
simulating	O
a	O
400	O
rpm	O
condition	O
.	O
The	O
per	O
cent	O
error	O
between	O
the	O
predicted	O
and	O
experimental	O
full-scale	O
flap	O
deflections	O
ranged	O
from	O
4	O
%	O
-LRB-	O
low	O
rpm	O
-RRB-	O
to	O
12.5	O
%	O
-LRB-	O
large	O
rpm	O
-RRB-	O
.	O
Increasing	O
the	O
electric	O
field	O
to	O
4	O
kV	O
cm/sup	O
-1	O
/	O
results	O
in	O
+	O
or-2	O
.5	O
degrees	O
flap	O
deflection	O
at	O
a	O
rotation	O
speed	O
of	O
400	O
rpm	O
,	O
according	O
to	O
the	O
design	B-KEY
analysis	O
.	O
A	O
trade	O
study	O
was	O
conducted	O
to	O
compare	O
the	O
performance	O
of	O
the	O
piezoelectric	B-KEY
tube	I-KEY
actuator	I-KEY
to	O
the	O
state	O
of	O
the	O
art	O
in	O
trailing	O
edge	O
flap	O
actuators	O
and	O
indicated	O
that	O
the	O
induced-shear	B-KEY
tube	I-KEY
actuator	I-KEY
shows	O
promise	O
as	O
a	O
trailing	O
edge	O
flap	O
actuator	O

Selective	B-KEY
representing	I-KEY
and	O
world-making	B-KEY
We	O
discuss	O
the	O
thesis	O
of	O
selective	O
representing-the	O
idea	O
that	O
the	O
contents	O
of	O
the	O
mental	B-KEY
representations	I-KEY
had	O
by	O
organisms	B-KEY
are	O
highly	O
constrained	O
by	O
the	O
biological	O
niches	O
within	O
which	O
the	O
organisms	B-KEY
evolved	O
.	O
While	O
such	O
a	O
thesis	O
has	O
been	O
defended	O
by	O
several	O
authors	O
elsewhere	O
,	O
our	O
primary	O
concern	O
here	O
is	O
to	O
take	O
up	O
the	O
issue	O
of	O
the	O
compatibility	O
of	O
selective	B-KEY
representing	I-KEY
and	O
realism	B-KEY
.	O
We	O
hope	O
to	O
show	O
three	O
things	O
.	O
First	O
,	O
that	O
the	O
notion	O
of	O
selective	B-KEY
representing	I-KEY
is	O
fully	O
consistent	O
with	O
the	O
realist	O
idea	O
of	O
a	O
mind-independent	B-KEY
world	I-KEY
.	O
Second	O
,	O
that	O
not	O
only	O
are	O
these	O
two	O
consistent	O
,	O
but	O
that	O
the	O
latter	O
-LRB-	O
the	O
realist	O
conception	O
of	O
a	O
mind-independent	B-KEY
world	I-KEY
-RRB-	O
provides	O
the	O
most	O
powerful	O
perspective	O
from	O
which	O
to	O
motivate	O
and	O
understand	O
the	O
differing	O
perceptual	O
and	O
cognitive	B-KEY
profiles	I-KEY
themselves	O
.	O
Third	O
,	O
that	O
the	O
-LRB-	O
genuine	O
and	O
important	O
-RRB-	O
sense	O
in	O
which	O
organism	B-KEY
and	O
environment	O
may	O
together	O
constitute	O
an	O
integrated	O
system	O
of	O
scientific	O
interest	O
poses	O
no	O
additional	O
threat	O
to	O
the	O
realist	O
conception	O

Creating	O
Web-based	B-KEY
listings	I-KEY
of	O
electronic	B-KEY
journals	I-KEY
without	O
creating	O
extra	O
work	O
Creating	O
up-to-date	O
listings	O
of	O
electronic	B-KEY
journals	I-KEY
is	O
challenging	O
due	O
to	O
frequent	O
changes	O
in	O
titles	O
available	O
and	O
in	O
URLs	B-KEY
for	O
electronic	B-KEY
journal	I-KEY
titles	O
.	O
However	O
,	O
many	O
library	B-KEY
users	O
may	O
want	O
to	O
browse	O
Web	B-KEY
pages	I-KEY
which	O
contain	O
listings	O
of	O
electronic	B-KEY
journals	I-KEY
arranged	O
by	O
title	O
and/or	O
academic	O
disciplines	O
.	O
This	O
case	B-KEY
study	I-KEY
examines	O
the	O
development	O
of	O
a	O
system	O
which	O
automatically	O
exports	O
data	O
from	O
the	O
online	B-KEY
catalog	I-KEY
and	O
incorporates	O
it	O
into	O
dynamically-generated	O
Web	B-KEY
sites	I-KEY
.	O
These	O
sites	O
provide	O
multiple	O
access	O
points	O
for	O
journals	O
,	O
include	O
Web-based	O
interfaces	O
enabling	O
subject	O
specialists	O
to	O
manage	O
the	O
list	O
of	O
titles	O
which	O
appears	O
in	O
their	O
subject	O
area	O
.	O
Because	O
data	O
are	O
automatically	O
extracted	O
from	O
the	O
catalog	O
,	O
overlap	O
in	O
updating	O
titles	O
and	O
URLs	B-KEY
is	O
avoided	O
.	O
Following	O
the	O
creation	O
of	O
this	O
system	O
,	O
usage	O
of	O
electronic	B-KEY
journals	I-KEY
dramatically	O
increased	O
and	O
feedback	B-KEY
has	O
been	O
positive	O
.	O
Future	O
challenges	O
include	O
developing	O
more	O
frequent	O
updates	O
and	O
motivating	O
subject	O
specialists	O
to	O
more	O
regularly	O
monitor	O
new	O
titles	O

Angular	B-KEY
disparity	I-KEY
in	O
ETACT	O
scintimammography	O
Emission	O
tuned	O
aperture	O
computed	O
tomography	O
-LRB-	O
ETACT	O
-RRB-	O
has	O
been	O
previously	O
shown	O
to	O
have	O
the	O
potential	O
for	O
the	O
detection	O
of	O
small	B-KEY
tumors	I-KEY
-LRB-	O
<	O
1	O
cm	O
-RRB-	O
in	O
scintimammography	O
.	O
However	O
,	O
the	O
optimal	O
approach	O
to	O
the	O
application	O
of	O
ETACT	O
in	O
the	O
clinic	O
has	O
yet	O
to	O
be	O
determined	O
.	O
Therefore	O
,	O
we	O
sought	O
to	O
determine	O
the	O
effect	O
of	O
the	O
angular	B-KEY
disparity	I-KEY
between	O
the	O
ETACT	O
projections	O
on	O
image	B-KEY
quality	I-KEY
through	O
the	O
use	O
of	O
a	O
computer	B-KEY
simulation	I-KEY
.	O
A	O
small	O
,	O
spherical	B-KEY
tumor	I-KEY
of	O
variable	O
size	O
-LRB-	O
5	O
,	O
7.5	O
or	O
10	O
mm	O
-RRB-	O
was	O
placed	O
at	O
the	O
center	O
of	O
a	O
hemispherical	B-KEY
breast	I-KEY
-LRB-	O
15	O
cm	O
diameter	O
-RRB-	O
.	O
The	O
tumor	O
to	O
nontumor	O
ratio	O
was	O
either	O
5:1	O
or	O
10:1	O
.	O
The	O
detector	O
was	O
modeled	O
to	O
be	O
a	O
gamma	B-KEY
camera	I-KEY
fitted	O
with	O
a	O
4-mm-diam	O
pinhole	B-KEY
collimator	I-KEY
.	O
The	O
pinhole-to-detector	O
and	O
the	O
pinhole-to-tumor	B-KEY
distances	I-KEY
were	O
25	O
and	O
15	O
cm	O
,	O
respectively	O
.	O
A	O
ray	B-KEY
tracing	I-KEY
technique	I-KEY
was	O
used	O
to	O
generate	O
three	O
sets	O
of	O
projections	O
-LRB-	O
10	O
degrees	O
,	O
15	O
degrees	O
,	O
and	O
20	O
degrees	O
,	O
angular	B-KEY
disparity	I-KEY
-RRB-	O
.	O
These	O
data	O
were	O
blurred	O
to	O
a	O
resolution	O
consistent	O
with	O
the	O
4	O
mm	O
pinhole	O
.	O
The	O
TACT	O
reconstruction	O
method	O
was	O
used	O
to	O
reconstruct	O
these	O
three	O
image	B-KEY
sets	I-KEY
.	O
The	O
tumor	O
contrast	O
and	O
the	O
axial	B-KEY
spatial	I-KEY
resolution	I-KEY
was	O
measured	O
.	O
Smaller	O
angular	B-KEY
disparity	I-KEY
led	O
to	O
an	O
improvement	O
in	O
image	O
contrast	O
but	O
at	O
a	O
cost	O
of	O
degraded	O
axial	B-KEY
spatial	I-KEY
resolution	I-KEY
.	O
The	O
improvement	O
in	O
contrast	O
is	O
due	O
to	O
a	O
slight	O
improvement	O
in	O
the	O
in-plane	B-KEY
spatial	I-KEY
resolution	I-KEY
.	O
Since	O
improved	O
contrast	O
should	O
lead	O
to	O
better	O
tumor	O
detectability	O
,	O
smaller	O
angular	B-KEY
disparity	I-KEY
should	O
be	O
used	O
.	O
However	O
,	O
the	O
difference	O
in	O
contrast	O
between	O
10	O
degrees	O
and	O
15	O
degrees	O
was	O
very	O
slight	O
and	O
therefore	O
a	O
reasonable	O
clinical	B-KEY
choice	I-KEY
for	O
angular	B-KEY
disparity	I-KEY
is	O
15	O
degrees	O

Priming	O
the	O
pipeline	O
-LSB-	O
women	O
in	O
computer	O
science	O
careers	O
-RSB-	O
In	O
1997	O
The	B-KEY
Backyard	I-KEY
Project	I-KEY
,	O
a	O
pilot	O
program	O
of	O
the	O
Garnett	O
Foundation	O
,	O
was	O
instituted	O
to	O
encourage	O
high	B-KEY
school	I-KEY
girls	I-KEY
to	O
explore	O
careers	O
in	O
the	O
computer	O
industry	O
.	O
At	O
that	O
time	O
,	O
the	O
Garnett	O
Foundation	O
commissioned	O
the	O
Global	O
Strategy	O
Group	O
to	O
execute	O
a	O
survey	O
of	O
652	O
college-bound	B-KEY
high	I-KEY
school	I-KEY
students	I-KEY
-LRB-	O
grades	O
9	O
through	O
12	O
-RRB-	O
,	O
to	O
help	O
discover	O
directions	O
that	O
The	B-KEY
Backyard	I-KEY
Project	I-KEY
might	O
take	O
to	O
try	O
to	O
move	O
toward	O
the	O
mission	O
of	O
the	O
pilot	O
program	O
.	O
It	O
conducted	O
the	O
study	O
by	O
telephone	O
between	O
March	O
25	O
and	O
April	O
8	O
,	O
1997	O
in	O
the	O
Silicon	O
Valley	O
,	O
Boston	O
,	O
and	O
Austin	O
metropolitan	O
areas	O
.	O
It	O
conducted	O
all	O
interviews	O
using	O
a	O
random	O
digit	O
dialing	O
methodology	O
,	O
derived	O
from	O
a	O
file	O
of	O
American	O
households	O
with	O
high	O
incidences	O
of	O
adolescent	O
children	O
.	O
The	O
top	O
six	O
answers	O
from	O
girls	O
to	O
the	O
survey	O
question	O
``	O
why	O
are	O
girls	O
less	O
likely	O
to	O
pursue	O
computer	O
science	O
careers	O
?	O
''	O
in	O
order	O
of	O
perceived	O
importance	O
by	O
the	O
girls	O
were	O
:	O
not	O
enough	O
role	O
models	O
;	O
women	O
have	O
other	O
interests	O
;	O
did	O
n't	O
know	O
about	O
the	O
industry	O
;	O
limited	O
opportunity	O
;	O
negative	O
media	O
;	O
and	O
too	O
nerdy	O
.	O
These	O
responses	O
are	O
discussed	O

A	O
notion	O
of	O
non-interference	O
for	O
timed	B-KEY
automata	I-KEY
The	O
non-interference	O
property	O
of	O
concurrent	B-KEY
systems	I-KEY
is	O
a	O
security	B-KEY
property	I-KEY
concerning	O
the	O
flow	O
of	O
information	O
among	O
different	O
levels	O
of	O
security	O
of	O
the	O
system	O
.	O
In	O
this	O
paper	O
we	O
introduce	O
a	O
notion	O
of	O
timed	O
non-interference	O
for	O
real-time	B-KEY
systems	I-KEY
specified	O
by	O
timed	B-KEY
automata	I-KEY
.	O
The	O
notion	O
is	O
presented	O
using	O
an	O
automata	O
based	O
approach	O
and	O
then	O
it	O
is	O
characterized	O
also	O
by	O
operations	O
and	O
equivalence	O
between	O
timed	O
languages	O
.	O
The	O
definition	O
is	O
applied	O
to	O
an	O
example	O
of	O
a	O
time-critical	B-KEY
system	I-KEY
modeling	O
a	O
simplified	O
control	O
of	O
an	O
airplane	O

Winning	O
post	O
-LSB-	O
mail	B-KEY
systems	O
-RSB-	O
Businesses	O
that	O
take	O
their	O
mail	B-KEY
for	O
granted	O
can	O
end	O
up	O
wasting	O
money	O
as	O
well	O
as	O
opportunities	O
.	O
Mike	O
Stecyk	O
,	O
VP	O
of	O
marketing	O
and	O
lines	O
of	O
business	O
at	O
Pitney	B-KEY
Bowes	I-KEY
,	O
suggests	O
strategies	B-KEY
for	O
making	O
more	O
of	O
a	O
great	O
opportunity	O

A	O
re-examination	O
of	O
probability	B-KEY
matching	I-KEY
and	O
rational	B-KEY
choice	O
In	O
a	O
typical	O
probability	B-KEY
learning	I-KEY
task	I-KEY
participants	O
are	O
presented	O
with	O
a	O
repeated	O
choice	O
between	O
two	O
response	O
alternatives	O
,	O
one	O
of	O
which	O
has	O
a	O
higher	O
payoff	B-KEY
probability	I-KEY
than	O
the	O
other	O
.	O
Rational	B-KEY
choice	O
theory	O
requires	O
that	O
participants	O
should	O
eventually	O
allocate	O
all	O
their	O
responses	O
to	O
the	O
high-payoff	O
alternative	O
,	O
but	O
previous	O
research	O
has	O
found	O
that	O
people	O
fail	O
to	O
maximize	O
their	O
payoffs	O
.	O
Instead	O
,	O
it	O
is	O
commonly	O
observed	O
that	O
people	O
match	O
their	O
response	B-KEY
probabilities	I-KEY
to	O
the	O
payoff	B-KEY
probabilities	I-KEY
.	O
We	O
report	O
three	O
experiments	O
on	O
this	O
choice	B-KEY
anomaly	I-KEY
using	O
a	O
simple	O
probability	B-KEY
learning	I-KEY
task	I-KEY
in	O
which	O
participants	O
were	O
provided	O
with	O
-LRB-	O
i	O
-RRB-	O
large	B-KEY
financial	I-KEY
incentives	I-KEY
,	O
-LRB-	O
ii	O
-RRB-	O
meaningful	O
and	O
regular	O
feedback	B-KEY
,	O
and	O
-LRB-	O
iii	O
-RRB-	O
extensive	B-KEY
training	I-KEY
.	O
In	O
each	O
experiment	O
large	O
proportions	O
of	O
participants	O
adopted	O
the	O
optimal	B-KEY
response	I-KEY
strategy	I-KEY
and	O
all	O
three	O
of	O
the	O
factors	O
mentioned	O
above	O
contributed	O
to	O
this	O
.	O
The	O
results	O
are	O
supportive	O
of	O
rational	B-KEY
choice	O
theory	O

A	O
computational	B-KEY
model	I-KEY
of	O
learned	B-KEY
avoidance	I-KEY
behavior	I-KEY
in	O
a	O
one-way	B-KEY
avoidance	I-KEY
experiment	I-KEY
We	O
present	O
a	O
computational	B-KEY
model	I-KEY
of	O
learned	B-KEY
avoidance	I-KEY
behavior	I-KEY
in	O
a	O
one-way	B-KEY
avoidance	I-KEY
experiment	I-KEY
.	O
Our	O
model	O
employs	O
the	O
reinforcement	B-KEY
learning	I-KEY
paradigm	O
and	O
a	O
temporal-difference	B-KEY
algorithm	I-KEY
to	O
implement	O
both	O
classically	O
conditioned	O
and	O
instrumentally	B-KEY
conditioned	I-KEY
components	I-KEY
.	O
The	O
role	O
of	O
the	O
classically	B-KEY
conditioned	I-KEY
component	I-KEY
is	O
to	O
develop	O
an	O
expectation	O
of	O
future	O
benefit	O
that	O
is	O
a	O
function	O
of	O
the	O
learning	O
system	O
's	O
state	O
and	O
action	O
.	O
Competition	O
among	O
the	O
instrumentally	B-KEY
conditioned	I-KEY
components	I-KEY
determines	O
the	O
overt	O
behavior	O
generated	O
by	O
the	O
learning	O
system	O
.	O
Our	O
model	O
displays	O
,	O
in	O
simulation	O
,	O
the	O
reduced	B-KEY
latency	I-KEY
of	O
the	O
avoidance	O
behavior	O
during	O
learning	O
with	O
continuing	O
trials	O
and	O
the	O
resistance	O
to	O
extinction	O
of	O
the	O
avoidance	O
response	O
.	O
These	O
results	O
are	O
consistent	O
with	O
experimentally	O
observed	O
animal	B-KEY
behavior	I-KEY
.	O
Our	O
model	O
extends	O
the	O
traditional	B-KEY
two-process	I-KEY
learning	I-KEY
mechanism	I-KEY
of	O
Mowrer	O
-LRB-	O
1947	O
-RRB-	O
by	O
explicitly	O
defining	O
the	O
mechanisms	O
of	O
proprioceptive	B-KEY
feedback	I-KEY
,	O
an	O
internal	B-KEY
clock	I-KEY
,	O
and	O
generalization	O
over	O
the	O
action	O
space	O

Wireless-retail	O
financial	O
services	O
:	O
adoption	O
ca	O
n't	O
justify	O
the	O
cost	O
Slow	O
adoption	O
by	O
retail	O
investors	O
,	O
costly	O
services	O
and	O
bankrupt	O
vendors	O
has	O
prompted	O
banks	B-KEY
and	O
brokerage	B-KEY
firms	I-KEY
to	O
turn	O
off	O
their	O
wireless	B-KEY
applications	I-KEY

A	O
round	O
of	O
cash	O
,	O
a	O
pound	O
of	O
flesh	O
-LSB-	O
telecom	B-KEY
-RSB-	O
Despite	O
the	O
upheaval	O
across	O
telecom	B-KEY
,	O
venture	B-KEY
capital	I-KEY
firms	I-KEY
are	O
still	O
investing	O
in	O
start-ups	O
.	O
But	O
while	O
a	O
promising	O
idea	O
and	O
a	O
catchy	O
name	O
were	O
enough	O
to	O
guarantee	O
millions	O
in	O
funding	O
at	O
the	O
peak	O
of	O
the	O
dotcom	O
frenzy	O
,	O
now	O
start-ups	O
must	O
prove-their	O
long-term	O
viability	B-KEY
,	O
and	O
be	O
willing	O
to	O
concede	O
control	O
of	O
their	O
business	O
to	O
their	O
VC	O
suitors	O

A	O
new	O
voltage-vector	B-KEY
selection	I-KEY
algorithm	I-KEY
in	O
direct	B-KEY
torque	I-KEY
control	I-KEY
of	O
induction	B-KEY
motor	I-KEY
drives	I-KEY
AC	B-KEY
drives	I-KEY
based	O
on	O
direct	B-KEY
torque	I-KEY
control	I-KEY
of	O
induction	O
machines	O
allow	O
high	B-KEY
dynamic	I-KEY
performance	I-KEY
to	O
be	O
obtained	O
with	O
very	O
simple	O
control	O
schemes	O
.	O
The	O
drive	O
behavior	O
,	O
in	O
terms	O
of	O
current	O
,	O
flux	O
and	O
torque	B-KEY
ripple	I-KEY
,	O
is	O
dependent	O
on	O
the	O
utilised	O
voltage	B-KEY
vector	I-KEY
selection	I-KEY
strategy	I-KEY
and	O
the	O
operating	B-KEY
conditions	I-KEY
.	O
In	O
this	O
paper	O
a	O
new	O
voltage	O
vector	O
selection	O
algorithm	O
,	O
which	O
allows	O
a	O
sensible	O
reduction	O
of	O
the	O
RMS	B-KEY
value	I-KEY
of	O
the	O
stator	B-KEY
current	I-KEY
ripple	I-KEY
without	O
increasing	O
the	O
average	O
value	O
of	O
the	O
inverter	B-KEY
switching	I-KEY
frequency	I-KEY
and	O
without	O
the	O
need	O
of	O
a	O
PWM	O
pulse	O
generator	O
block	O
is	O
presented	O
Numerical	O
simulations	O
have	O
been	O
carried	O
out	O
to	O
validate	O
the	O
proposed	O
method	O

A	O
fast	O
implementation	O
of	O
correlation	B-KEY
of	O
long	B-KEY
data	I-KEY
sequences	I-KEY
for	O
coherent	B-KEY
receivers	I-KEY
Coherent	O
reception	O
depends	O
upon	O
matching	O
of	O
phase	O
between	O
the	O
transmitted	O
and	O
received	B-KEY
signal	I-KEY
.	O
Fast	O
convolution	O
techniques	O
based	O
on	O
fast	B-KEY
Fourier	I-KEY
transform	I-KEY
-LRB-	O
FFT	O
-RRB-	O
are	O
widely	O
used	O
for	O
extracting	O
time	B-KEY
delay	I-KEY
information	I-KEY
from	O
such	O
matching	O
.	O
The	O
latency	B-KEY
in	O
processing	O
a	O
large	O
data	O
window	O
of	O
the	O
received	B-KEY
signal	I-KEY
is	O
a	O
serious	O
overhead	O
for	O
mission	B-KEY
critical	I-KEY
real	I-KEY
time	I-KEY
applications	I-KEY
.	O
The	O
implementation	O
of	O
a	O
parallel	B-KEY
algorithm	I-KEY
for	O
correlation	B-KEY
of	O
long	B-KEY
data	I-KEY
sequences	I-KEY
in	O
multiprocessor	B-KEY
environment	I-KEY
is	O
demonstrated	O
here	O
.	O
The	O
algorithm	O
does	O
processing	O
while	O
acquiring	O
the	O
received	B-KEY
signal	I-KEY
and	O
reduces	O
the	O
computation	B-KEY
overhead	O
considerably	O
because	O
of	O
inherent	O
parallelism	O

A	O
hybrid	B-KEY
model	I-KEY
for	O
smoke	B-KEY
simulation	I-KEY
A	O
smoke	B-KEY
simulation	I-KEY
approach	O
based	O
on	O
the	O
integration	O
of	O
traditional	O
particle	O
systems	O
and	O
density	B-KEY
functions	I-KEY
is	O
presented	O
in	O
this	O
paper	O
.	O
By	O
attaching	O
a	O
density	B-KEY
function	I-KEY
to	O
each	O
particle	O
as	O
its	O
attribute	O
,	O
the	O
diffusion	O
of	O
smoke	O
can	O
be	O
described	O
by	O
the	O
variation	O
of	O
particles	O
'	O
density	B-KEY
functions	I-KEY
,	O
along	O
with	O
the	O
effect	O
on	O
airflow	O
by	O
controlling	O
particles	O
'	O
movement	O
and	O
fragmentation	O
.	O
In	O
addition	O
,	O
a	O
continuous	B-KEY
density	I-KEY
field	I-KEY
for	O
realistic	O
rendering	B-KEY
can	O
be	O
generated	O
quickly	O
through	O
the	O
look-up	B-KEY
tables	I-KEY
of	O
particle	O
's	O
density	B-KEY
functions	I-KEY
.	O
Compared	O
with	O
traditional	O
particle	O
systems	O
,	O
this	O
approach	O
can	O
describe	O
smoke	O
diffusion	O
,	O
and	O
provide	O
a	O
continuous	B-KEY
density	I-KEY
field	I-KEY
for	O
realistic	O
rendering	B-KEY
with	O
much	O
less	O
computation	O
.	O
A	O
quick	O
rendering	B-KEY
scheme	O
is	O
also	O
presented	O
in	O
this	O
paper	O
as	O
a	O
useful	O
preview	O
tool	O
for	O
tuning	O
appropriate	O
parameters	O
in	O
the	O
smoke	O
model	O

Three-dimensional	O
global	O
MHD	B-KEY
simulation	I-KEY
code	O
for	O
the	O
Earth	O
's	O
magnetosphere	O
using	O
HPF/JA	O
We	O
have	O
translated	O
a	O
three-dimensional	O
magnetohydrodynamic	O
-LRB-	O
MHD	O
-RRB-	O
simulation	O
code	O
of	O
the	O
Earth	O
's	O
magnetosphere	O
from	O
VPP	O
Fortran	O
to	O
HPF/JA	O
on	O
the	O
Fujitsu	B-KEY
VPP5000/56	I-KEY
vector-parallel	B-KEY
supercomputer	I-KEY
and	O
the	O
MHD	O
code	O
was	O
fully	O
vectorized	O
and	O
fully	O
parallelized	O
in	O
VPP	O
Fortran	O
.	O
The	O
entire	O
performance	O
and	O
capability	O
of	O
the	O
HPF	B-KEY
MHD	I-KEY
code	I-KEY
could	O
be	O
shown	O
to	O
be	O
almost	O
comparable	O
to	O
that	O
of	O
VPP	O
Fortran	O
.	O
A	O
three-dimensional	O
global	O
MHD	B-KEY
simulation	I-KEY
of	O
the	O
Earth	O
's	O
magnetosphere	O
was	O
performed	O
at	O
a	O
speed	O
of	O
over	O
400	O
Gflops	O
with	O
an	O
efficiency	O
of	O
76.5	O
%	O
using	O
56	O
processing	O
elements	O
of	O
the	O
Fujitsu	B-KEY
VPP5000/56	I-KEY
in	O
vector	O
and	O
parallel	B-KEY
computation	I-KEY
that	O
permitted	O
comparison	O
with	O
catalog	O
values	O
.	O
We	O
have	O
concluded	O
that	O
fluid	O
and	O
MHD	O
codes	O
that	O
are	O
fully	O
vectorized	O
and	O
fully	O
parallelized	O
in	O
VPP	O
Fortran	O
can	O
be	O
translated	O
with	O
relative	O
ease	O
to	O
HPF/JA	O
,	O
and	O
a	O
code	O
in	O
HPF/JA	O
may	O
be	O
expected	O
to	O
perform	O
comparably	O
to	O
the	O
same	O
code	O
written	O
in	O
VPP	O
Fortran	O

ERP	B-KEY
systems	I-KEY
implementation	I-KEY
:	O
Best	B-KEY
practices	I-KEY
in	O
Canadian	B-KEY
government	I-KEY
organizations	I-KEY
ERP	O
-LRB-	O
Enterprise	B-KEY
resource	I-KEY
planning	I-KEY
-RRB-	O
systems	O
implementation	O
is	O
a	O
complex	O
exercise	O
in	O
organizational	O
innovation	O
and	O
change	O
management	O
.	O
Government	O
organizations	O
are	O
increasing	O
their	O
adoption	O
of	O
these	O
systems	O
for	O
various	O
benefits	O
such	O
as	O
integrated	B-KEY
real-time	I-KEY
information	I-KEY
,	O
better	O
administration	B-KEY
,	O
and	O
result-based	B-KEY
management	I-KEY
.	O
Government	O
organizations	O
,	O
due	O
to	O
their	O
social	B-KEY
obligations	I-KEY
,	O
higher	O
legislative	O
and	O
public	B-KEY
accountability	I-KEY
,	O
and	O
unique	O
culture	O
face	O
many	O
specific	O
challenges	O
in	O
the	O
transition	O
to	O
enterprise	O
systems	O
.	O
This	O
motivated	O
the	O
authors	O
to	O
explore	O
the	O
key	O
considerations	O
and	O
typical	O
activities	O
in	O
government	O
organizations	O
adopting	O
ERP	O
systems	O
.	O
The	O
article	O
adopts	O
the	O
innovation	B-KEY
process	I-KEY
theory	I-KEY
framework	I-KEY
as	O
well	O
as	O
the	O
-LRB-	O
Markus	O
&	O
Tanis	O
,	O
2000	O
-RRB-	O
model	O
as	O
a	O
basis	O
to	O
delineate	O
the	O
ERP	O
adoption	O
process	O
.	O
Although	O
,	O
each	O
adopting	O
organization	O
has	O
a	O
distinct	O
set	O
of	O
objectives	O
for	O
its	O
systems	O
,	O
the	O
study	O
found	O
many	O
similarities	O
in	O
motivations	O
,	O
concerns	O
,	O
and	O
strategies	O
across	O
organizations	O

War	O
games	O
:	O
The	O
truth	O
-LSB-	O
network	B-KEY
security	B-KEY
-RSB-	O
With	O
al	O
Qaeda	O
on	O
the	O
tip	O
of	O
tongues	O
around	O
the	O
world	O
,	O
find	O
out	O
how	O
terror	O
groups	O
could	O
target	O
your	O
network	B-KEY
.	O
What	O
are	O
the	O
dangers	O
and	O
how	O
do	O
you	O
fight	O
them	O
?	O

Efficient	O
cellular	B-KEY
automata	I-KEY
based	I-KEY
versatile	I-KEY
multiplier	I-KEY
for	O
GF	O
-LRB-	O
2/sup	O
m	O
/	O
-RRB-	O
In	O
this	O
paper	O
,	O
a	O
low-complexity	B-KEY
programmable	I-KEY
cellular	I-KEY
automata	I-KEY
-LRB-	O
PCA	O
-RRB-	O
based	O
versatile	O
modular	O
multiplier	O
in	O
GF	O
-LRB-	O
2/sup	O
m	O
/	O
-RRB-	O
is	O
presented	O
.	O
The	O
proposed	O
versatile	O
multiplier	O
increases	O
flexibility	O
by	O
using	O
the	O
same	O
multiplier	O
in	O
different	O
security	B-KEY
environments	I-KEY
,	O
and	O
it	O
reduces	O
the	O
user	O
's	O
cost	O
.	O
Moreover	O
,	O
the	O
multiplier	O
can	O
be	O
easily	O
extended	O
to	O
high	O
order	O
of	O
m	O
for	O
more	O
security	O
,	O
and	O
low-cost	O
serial	O
implementation	O
is	O
feasible	O
in	O
restricted	B-KEY
computing	I-KEY
environments	I-KEY
,	O
such	O
as	O
smart	B-KEY
cards	I-KEY
and	O
wireless	B-KEY
devices	I-KEY

Algebraic	B-KEY
conditions	I-KEY
for	O
high-order	O
convergent	O
deferred	B-KEY
correction	I-KEY
schemes	I-KEY
based	O
on	O
Runge-Kutta-Nystrom	O
methods	O
for	O
second	O
order	O
boundary	O
value	O
problems	O
In	O
-LSB-	O
T.	O
Van	O
Hecke	O
,	O
M.	O
Van	O
Daele	O
,	O
J.	O
Comp	O
.	O
Appl	O
.	O
Math.	O
,	O
vol	O
.	O
132	O
,	O
p.	O
107-125	O
,	O
-LRB-	O
2001	O
-RRB-	O
-RSB-	O
the	O
investigation	O
of	O
high-order	O
convergence	O
of	O
deferred	B-KEY
correction	I-KEY
schemes	I-KEY
for	O
the	O
numerical	O
solution	O
of	O
second	B-KEY
order	I-KEY
nonlinear	I-KEY
two-point	I-KEY
boundary	I-KEY
value	I-KEY
problems	I-KEY
not	O
containing	O
the	O
first	O
derivative	O
,	O
is	O
made	O
.	O
The	O
derivation	O
of	O
the	O
algebraic	B-KEY
conditions	I-KEY
to	O
raise	O
the	O
increase	O
of	O
order	O
by	O
the	O
deferred	B-KEY
correction	I-KEY
scheme	I-KEY
was	O
based	O
on	O
Taylor	B-KEY
series	I-KEY
expansions	I-KEY
.	O
In	O
this	O
paper	O
we	O
describe	O
a	O
more	O
elegant	O
way	O
by	O
means	O
of	O
P-series	O
to	O
obtain	O
this	O
necessary	O
conditions	O
and	O
generalize	O
this	O
idea	O
to	O
equations	O
of	O
the	O
form	O
y	O
''	O
=	O
f	O
-LRB-	O
t	O
,	O
y	O
,	O
y	O
'	O
-RRB-	O

Fractional	B-KEY
motion	I-KEY
control	I-KEY
:	O
application	O
to	O
an	O
XY	B-KEY
cutting	I-KEY
table	I-KEY
In	O
path	B-KEY
tracking	I-KEY
design	I-KEY
,	O
the	O
dynamic	O
of	O
actuators	B-KEY
must	O
be	O
taken	O
into	O
account	O
in	O
order	O
to	O
reduce	O
overshoots	O
appearing	O
for	O
small	O
displacements	O
.	O
A	O
new	O
approach	O
to	O
path	O
tracking	O
using	O
fractional	B-KEY
differentiation	I-KEY
is	O
proposed	O
with	O
its	O
application	O
on	O
a	O
XY	B-KEY
cutting	I-KEY
table	I-KEY
.	O
It	O
permits	O
the	O
generation	O
of	O
optimal	B-KEY
movement	O
reference-input	O
leading	O
to	O
a	O
minimum	B-KEY
path	I-KEY
completion	I-KEY
time	I-KEY
,	O
taking	O
into	O
account	O
both	O
maximum	O
velocity	O
,	O
acceleration	O
and	O
torque	O
and	O
the	O
bandwidth	O
of	O
the	O
closed-loop	B-KEY
system	I-KEY
.	O
Fractional	B-KEY
differentiation	I-KEY
is	O
used	O
here	O
through	O
a	O
Davidson-Cole	B-KEY
filter	I-KEY
.	O
A	O
methodology	O
aiming	O
at	O
improving	O
the	O
accuracy	O
especially	O
on	O
checkpoints	O
is	O
presented	O
.	O
The	O
reference-input	O
obtained	O
is	O
compared	O
with	O
spline	B-KEY
function	I-KEY
.	O
Both	O
are	O
applied	O
to	O
an	O
XY	B-KEY
cutting	I-KEY
table	I-KEY
model	O
and	O
actuator	B-KEY
outputs	O
compared	O

Web	B-KEY
content	I-KEY
extraction	I-KEY
.	O
A	O
WhizBang	O
!	O
approach	O
The	O
extraction	O
technology	O
that	O
Whizbang	O
uses	O
consists	O
of	O
a	O
unique	O
approach	O
to	O
scouring	O
the	O
Web	O
for	O
current	O
,	O
very	O
specific	O
forms	O
of	O
information	O
.	O
FlipDog	B-KEY
,	O
for	O
example	O
,	O
checks	O
company	B-KEY
Web	I-KEY
sites	I-KEY
for	O
hyperlinks	O
to	O
pages	O
that	O
list	O
job	O
opportunities	O
.	O
It	O
then	O
crawls	O
to	O
the	O
deeper	O
page	O
and	O
,	O
using	O
the	O
WhizBang	O
!	O
Extraction	O
Framework	O
,	O
extracts	O
the	O
key	O
elements	O
of	O
the	O
postings	O
,	O
such	O
as	O
job	O
title	O
,	O
name	O
of	O
employer	O
,	O
job	O
category	O
,	O
and	O
job	O
function	O
.	O
Click	O
on	O
a	O
job	O
and	O
you	O
are	O
transferred	O
to	O
the	O
company	B-KEY
Web	I-KEY
site	I-KEY
to	O
view	O
the	O
job	B-KEY
description	I-KEY
as	O
it	O
appears	O
there	O

Virtual	O
projects	O
at	O
Halden	O
-LSB-	O
Reactor	O
Project	O
-RSB-	O
The	O
Halden	O
man-machine	O
systems	O
-LRB-	O
MMS	O
-RRB-	O
programme	O
for	O
2002	O
is	O
intended	O
to	O
address	O
issues	O
related	O
to	O
human	B-KEY
factors	I-KEY
,	O
control	B-KEY
room	I-KEY
design	I-KEY
,	O
computer-based	B-KEY
support	I-KEY
system	I-KEY
areas	O
and	O
system	O
safety	B-KEY
and	O
reliability	B-KEY
.	O
The	O
Halden	O
MMS	O
programme	O
is	O
intended	O
to	O
address	O
extensive	O
experimental	O
work	O
in	O
the	O
human	B-KEY
factors	I-KEY
,	O
control	B-KEY
room	I-KEY
design	I-KEY
and	O
computer-based	B-KEY
support	I-KEY
system	I-KEY
areas	O
.	O
The	O
work	O
is	O
based	O
on	O
experiments	O
and	O
demonstrations	O
carried	O
out	O
in	O
the	O
experimental	O
facility	O
HAMMLAB	O
.	O
Pilot-versions	O
of	O
several	O
operator	O
aids	O
are	O
adopted	O
and	O
integrated	O
to	O
the	O
HAMMLAB	O
simulators	O
and	O
demonstrated	O
in	O
a	O
full	O
dynamic	O
setting	O
.	O
The	O
Halden	O
virtual	B-KEY
reality	I-KEY
laboratory	O
has	O
recently	O
become	O
an	O
integral	O
and	O
important	O
part	O
of	O
the	O
programme	O

L/sub	O
p	O
/	O
stability	O
and	O
linearization	O
A	O
theorem	O
by	O
Hadamard	O
gives	O
a	O
two-part	O
condition	O
under	O
which	O
a	O
map	O
from	O
one	O
Banach	B-KEY
space	I-KEY
to	O
another	O
is	O
a	O
homeomorphism	O
.	O
The	O
theorem	O
,	O
while	O
often	O
very	O
useful	O
,	O
is	O
incomplete	O
in	O
the	O
sense	O
that	O
it	O
does	O
not	O
explicitly	O
specify	O
the	O
family	O
of	O
maps	O
for	O
which	O
the	O
condition	O
is	O
met	O
.	O
Recently	O
,	O
under	O
a	O
typically	O
weak	O
additional	O
assumption	O
on	O
the	O
map	O
,	O
it	O
was	O
shown	O
that	O
Hadamard	O
's	O
condition	O
is	O
met	O
if	O
and	O
only	O
if	O
the	O
map	O
is	O
a	O
homeomorphism	O
with	O
a	O
Lipschitz	B-KEY
continuous	I-KEY
inverse	I-KEY
.	O
Here	O
,	O
an	O
application	O
is	O
given	O
concerning	O
the	O
relation	O
between	O
the	O
L/sub	O
p	O
/	O
stability	O
-LRB-	O
with	O
1	O
<	O
or	O
=	O
p	O
<	O
infinity	O
-RRB-	O
of	O
a	O
nonlinear	B-KEY
system	I-KEY
and	O
the	O
stability	O
of	O
related	O
linear	B-KEY
systems	I-KEY
.	O
We	O
also	O
give	O
a	O
result	O
that	O
directs	O
attention	O
to	O
a	O
fundamental	O
limitation	O
concerning	O
what	O
can	O
be	O
proved	O
about	O
linearization	O
and	O
stability	O
for	O
a	O
related	O
familiar	O
family	O
of	O
feedback	B-KEY
systems	I-KEY

Getting	O
the	O
most	O
out	O
of	O
intrusion	B-KEY
detection	I-KEY
systems	I-KEY
Intrusion	B-KEY
detection	I-KEY
systems	I-KEY
-LRB-	O
IDS	O
-RRB-	O
can	O
play	O
a	O
very	O
valuable	O
role	O
in	O
the	O
defence	O
of	O
a	O
network	O
.	O
However	O
,	O
it	O
is	O
important	O
to	O
understand	O
not	O
just	O
what	O
it	O
will	O
do	O
-LRB-	O
and	O
how	O
it	O
does	O
it	O
-RRB-	O
-	O
but	O
what	O
it	O
wo	O
n't	O
do	O
-LRB-	O
and	O
why	O
-RRB-	O
.	O
This	O
article	O
does	O
not	O
go	O
into	O
the	O
technical	O
working	O
of	O
IDS	O
in	O
too	O
much	O
detail	O
,	O
rather	O
it	O
limits	O
itself	O
to	O
a	O
discussion	O
of	O
some	O
of	O
the	O
capabilities	O
and	O
failings	O
of	O
the	O
technology	O

The	O
real	O
story	O
behind	O
Calpoint	B-KEY
-LSB-	O
telecom	O
-RSB-	O
A	O
former	O
Qwest	B-KEY
executive	O
sheds	O
light	O
on	O
the	O
carrier	O
's	O
controversial	O
deal	O
with	O
Calpoint	B-KEY
.	O
Discusses	O
why	O
Calpoint	B-KEY
gets	O
a	O
monthly	O
check	O
from	O
Quest	O
,	O
regardless	O
of	O
whether	O
it	O
provides	O
services	O

Cache	B-KEY
invalidation	I-KEY
and	O
replacement	O
strategies	O
for	O
location-dependent	O
data	O
in	O
mobile	O
environments	O
Mobile	O
location-dependent	B-KEY
information	I-KEY
services	O
-LRB-	O
LDISs	O
-RRB-	O
have	O
become	O
increasingly	O
popular	O
in	O
recent	O
years	O
.	O
However	O
,	O
data	B-KEY
caching	I-KEY
strategies	O
for	O
LDISs	O
have	O
thus	O
far	O
received	O
little	O
attention	O
.	O
In	O
this	O
paper	O
,	O
we	O
study	O
the	O
issues	O
of	O
cache	B-KEY
invalidation	I-KEY
and	O
cache	B-KEY
replacement	I-KEY
for	O
location-dependent	O
data	O
under	O
a	O
geometric	O
location	O
model	O
.	O
We	O
introduce	O
a	O
new	O
performance	O
criterion	O
,	O
called	O
caching	O
efficiency	O
,	O
and	O
propose	O
a	O
generic	O
method	O
for	O
location-dependent	O
cache	B-KEY
invalidation	I-KEY
strategies	O
.	O
In	O
addition	O
,	O
two	O
cache	B-KEY
replacement	I-KEY
policies	O
,	O
PA	O
and	O
PAID	O
,	O
are	O
proposed	O
.	O
Unlike	O
the	O
conventional	O
replacement	O
policies	O
,	O
PA	O
and	O
PAID	O
take	O
into	O
consideration	O
the	O
valid	O
scope	O
area	O
of	O
a	O
data	O
value	O
.	O
We	O
conduct	O
a	O
series	O
of	O
simulation	O
experiments	O
to	O
study	O
the	O
performance	O
of	O
the	O
proposed	O
caching	O
schemes	O
.	O
The	O
experimental	O
results	O
show	O
that	O
the	O
proposed	O
location-dependent	O
invalidation	O
scheme	O
is	O
very	O
effective	O
and	O
the	O
PA	O
and	O
PAID	O
policies	O
significantly	O
outperform	O
the	O
conventional	O
replacement	O
policies	O

Adaptable	B-KEY
dialog	I-KEY
boxes	I-KEY
for	O
cross-platform	O
programming	O
The	O
author	O
presents	O
a	O
framework	O
for	O
building	O
dialog	B-KEY
boxes	I-KEY
that	O
adapt	O
to	O
the	O
look	O
and	O
feel	O
of	O
their	O
platform	O
.	O
This	O
method	O
also	O
helps	O
with	O
a	O
few	O
related	O
problems	O
:	O
specifying	O
cross-platform	B-KEY
resources	I-KEY
and	O
handling	O
dialog	B-KEY
size	I-KEY
changes	I-KEY
due	O
to	O
localization	B-KEY
.	O
He	O
uses	O
a	O
combination	O
of	O
XML	B-KEY
,	O
automatic	B-KEY
layout	I-KEY
,	O
and	O
run-time	B-KEY
dialog	I-KEY
creation	I-KEY
to	O
give	O
you	O
most	O
of	O
the	O
benefits	O
of	O
platform-specific	B-KEY
resources	I-KEY
,	O
without	O
the	O
associated	O
pain	O
.	O
Source	O
code	O
with	O
an	O
implementation	O
of	O
the	O
layout	O
engine	O
for	O
Mac	B-KEY
OS	I-KEY
9.1	I-KEY
-LRB-	O
``	O
Carbon	O
''	O
-RRB-	O
,	O
Mac	B-KEY
OS	I-KEY
X	I-KEY
,	O
and	O
Microsoft	B-KEY
Windows	I-KEY
can	O
be	O
downloaded	O
from	O
the	O
CUJ	O
website	O
at	O
<	O
www.cuj.com/code	O
>	O
.	O
You	O
can	O
use	O
this	O
code	O
as	O
is	O
,	O
or	O
as	O
a	O
starting	O
point	O
for	O
your	O
own	O
more	O
complete	O
implementation	O

Modelling	O
of	O
complete	B-KEY
robot	I-KEY
dynamics	I-KEY
based	O
on	O
a	O
multi-dimensional	O
,	O
RBF-like	O
neural	B-KEY
architecture	I-KEY
A	O
neural	O
network	O
based	O
identification	O
approach	O
of	O
manipulator	B-KEY
dynamics	I-KEY
is	O
presented	O
.	O
For	O
a	O
structured	O
modelling	O
,	O
RBF-like	O
static	B-KEY
neural	I-KEY
networks	I-KEY
are	O
used	O
in	O
order	O
to	O
represent	O
and	O
adapt	O
all	O
model	O
parameters	O
with	O
their	O
non-linear	O
dependences	O
on	O
the	O
joint	O
positions	O
.	O
The	O
neural	B-KEY
architecture	I-KEY
is	O
hierarchically	O
organised	O
to	O
reach	O
optimal	O
adjustment	O
to	O
structural	O
a	O
priori-knowledge	O
about	O
the	O
identification	O
problem	O
.	O
The	O
model	O
structure	O
is	O
substantially	O
simplified	O
by	O
general	B-KEY
system	I-KEY
analysis	I-KEY
independent	O
of	O
robot	O
type	O
.	O
But	O
also	O
a	O
lot	O
of	O
specific	O
features	O
of	O
the	O
utilised	O
experimental	O
robot	O
are	O
taken	O
into	O
account	O
.	O
A	O
fixed	O
,	O
grid	O
based	O
neuron	O
placement	O
together	O
with	O
application	O
of	O
B-spline	B-KEY
polynomial	I-KEY
basis	I-KEY
functions	I-KEY
is	O
utilised	O
favourably	O
for	O
a	O
very	O
effective	O
recursive	B-KEY
implementation	I-KEY
of	O
the	O
neural	B-KEY
architecture	I-KEY
.	O
Thus	O
,	O
an	O
online	B-KEY
identification	I-KEY
of	O
a	O
dynamic	B-KEY
model	I-KEY
is	O
submitted	O
for	O
a	O
complete	B-KEY
6	I-KEY
joint	I-KEY
industrial	I-KEY
robot	I-KEY

Storage	B-KEY
functionals	I-KEY
and	O
Lyapunov	O
functions	O
for	O
passive	B-KEY
dynamical	I-KEY
systems	I-KEY
For	O
nonlinear	B-KEY
time-invariant	I-KEY
input-output	I-KEY
dynamical	I-KEY
systems	I-KEY
the	O
passivity	B-KEY
conditions	I-KEY
are	O
obtained	O
under	O
some	O
restrictions	O
.	O
The	O
conditions	O
imply	O
storage	B-KEY
functions	I-KEY
satisfying	O
a	O
dissipation	B-KEY
inequality	I-KEY
.	O
A	O
class	O
of	O
storage	B-KEY
functions	I-KEY
allowing	O
unique	O
reconstruction	O
of	O
a	O
passive	B-KEY
dynamical	I-KEY
system	I-KEY
is	O
defined	O
.	O
These	O
results	O
are	O
illustrated	O
by	O
an	O
example	O
of	O
a	O
linear	B-KEY
system	I-KEY
with	O
fading	B-KEY
memory	I-KEY
.	O
An	O
important	O
,	O
for	O
practical	O
application	O
,	O
class	O
of	O
the	O
linear	O
relaxation	O
systems	O
without	O
direct	O
input-output	O
interaction	O
is	O
considered	O
.	O
A	O
necessary	B-KEY
condition	I-KEY
for	O
dynamical	O
systems	O
to	O
be	O
of	O
the	O
relaxation	O
type	O
is	O
obtained	O
for	O
this	O
class	O
.	O
The	O
condition	O
is	O
connected	O
with	O
the	O
existence	O
of	O
a	O
unique	B-KEY
quadratic	I-KEY
Lyapunov	I-KEY
function	I-KEY
satisfying	O
the	O
complete	B-KEY
monotonicity	I-KEY
condition	I-KEY
.	O
This	O
unique	O
Lyapunov	O
function	O
corresponds	O
to	O
a	O
``	O
standard	O
''	O
thermodynamic	B-KEY
potential	I-KEY
in	O
a	O
compact	O
family	O
of	O
potentials	O
in	O
the	O
nonequilibrium	B-KEY
thermodynamics	I-KEY
.	O
The	O
results	O
obtained	O
can	O
be	O
useful	O
in	O
automatic	B-KEY
control	I-KEY
,	O
mechanics	B-KEY
of	O
viscoelastic	B-KEY
materials	I-KEY
,	O
and	O
various	O
applications	O
in	O
physics	O
and	O
the	O
system	O
theory	O

Smart	O
collision	B-KEY
information	I-KEY
processing	I-KEY
sensors	O
for	O
fast	B-KEY
moving	I-KEY
objects	I-KEY
In	O
this	O
technical	O
note	O
we	O
survey	O
the	O
area	O
of	O
smart	O
collision	B-KEY
information	I-KEY
processing	I-KEY
sensors	O
.	O
We	O
review	O
the	O
existing	O
technologies	O
to	O
detect	O
collision	O
or	O
overlap	O
between	O
fast	O
moving	O
physical	O
objects	O
or	O
objects	O
in	O
virtual	B-KEY
environments	I-KEY
,	O
physical	B-KEY
environments	I-KEY
or	O
a	O
combination	O
of	O
physical	O
and	O
virtual	O
objects	O
.	O
We	O
report	O
developments	O
in	O
the	O
collision	B-KEY
detection	I-KEY
of	O
fast	B-KEY
moving	I-KEY
objects	I-KEY
at	O
discrete	B-KEY
time	I-KEY
steps	I-KEY
such	O
as	O
two	O
consecutive	B-KEY
time	I-KEY
frames	I-KEY
,	O
as	O
well	O
as	O
continuous	B-KEY
time	I-KEY
intervals	I-KEY
such	O
as	O
in	O
an	O
interframe	B-KEY
collision	I-KEY
detection	I-KEY
system	O
.	O
Our	O
discussion	O
of	O
computational	O
techniques	O
in	O
this	O
paper	O
is	O
limited	O
to	O
convex	B-KEY
objects	I-KEY
.	O
Techniques	O
exist	O
however	O
to	O
efficiently	O
decompose	O
non-convex	O
objects	O
into	O
convex	O
objects	O
.	O
We	O
also	O
discuss	O
the	O
tracking	B-KEY
technologies	O
for	O
objects	O
from	O
the	O
standpoint	O
of	O
collision	B-KEY
detection	I-KEY
or	O
avoidance	O

Tracking	B-KEY
control	I-KEY
of	O
the	O
flexible	B-KEY
slider-crank	I-KEY
mechanism	I-KEY
system	I-KEY
under	O
impact	B-KEY
The	O
variable	B-KEY
structure	I-KEY
control	I-KEY
-LRB-	O
VSC	O
-RRB-	O
and	O
the	O
stabilizer	B-KEY
design	I-KEY
by	O
using	O
the	O
pole	B-KEY
placement	I-KEY
technique	I-KEY
are	O
applied	O
to	O
the	O
tracking	B-KEY
control	I-KEY
of	O
the	O
flexible	O
slider-crank	O
mechanism	O
under	O
impact	B-KEY
.	O
The	O
VSC	O
strategy	O
is	O
employed	O
to	O
track	O
the	O
crank	B-KEY
angular	I-KEY
position	I-KEY
and	O
speed	O
,	O
while	O
the	O
stabilizer	B-KEY
design	I-KEY
is	O
involved	O
to	O
suppress	O
the	O
flexible	B-KEY
vibrations	I-KEY
simultaneously	O
.	O
From	O
the	O
theoretical	O
impact	B-KEY
consideration	O
,	O
three	O
approaches	O
including	O
the	O
generalized	B-KEY
momentum	I-KEY
balance	I-KEY
-LRB-	O
GMB	O
-RRB-	O
,	O
the	O
continuous	B-KEY
force	I-KEY
model	I-KEY
-LRB-	O
CFM	O
-RRB-	O
,	O
and	O
the	O
CFM	O
associated	O
with	O
the	O
effective	B-KEY
mass	I-KEY
compensation	I-KEY
EMC	O
are	O
adopted	O
,	O
and	O
are	O
derived	O
on	O
the	O
basis	O
of	O
the	O
energy	O
and	O
impulse-momentum	O
conservations	O
.	O
Simulation	O
results	O
are	O
provided	O
to	O
demonstrate	O
the	O
performance	O
of	O
the	O
motor-controller	O
flexible	O
slider-crank	O
mechanism	O
not	O
only	O
accomplishing	O
good	O
tracking	B-KEY
trajectory	I-KEY
of	O
the	O
crank	O
angle	O
,	O
but	O
also	O
eliminating	O
vibrations	O
of	O
the	O
flexible	B-KEY
connecting	I-KEY
rod	I-KEY

A	O
method	O
for	O
geometrical	O
verification	O
of	O
dynamic	B-KEY
intensity	I-KEY
modulated	I-KEY
radiotherapy	I-KEY
using	O
a	O
scanning	O
electronic	O
portal	O
imaging	O
device	O
In	O
order	O
to	O
guarantee	O
the	O
safe	B-KEY
delivery	I-KEY
of	O
dynamic	B-KEY
intensity	I-KEY
modulated	I-KEY
radiotherapy	I-KEY
-LRB-	O
IMRT	O
-RRB-	O
,	O
verification	O
of	O
the	O
leaf	B-KEY
trajectories	I-KEY
during	O
the	O
treatment	O
is	O
necessary	O
.	O
Our	O
aim	O
in	O
this	O
study	O
is	O
to	O
develop	O
a	O
method	O
for	O
on-line	B-KEY
verification	I-KEY
of	O
leaf	B-KEY
trajectories	I-KEY
using	O
an	O
electronic	O
portal	O
imaging	O
device	O
with	O
scanning	B-KEY
read-out	I-KEY
,	O
independent	O
of	O
the	O
multileaf	O
collimator	O
.	O
Examples	O
of	O
such	O
scanning	O
imagers	O
are	O
electronic	O
portal	O
imaging	O
devices	O
-LRB-	O
EPIDs	O
-RRB-	O
based	O
on	O
liquid-filled	B-KEY
ionization	I-KEY
chambers	I-KEY
and	O
those	O
based	O
on	O
amorphous	O
silicon	O
.	O
Portal	O
images	O
were	O
acquired	O
continuously	O
with	O
a	O
liquid-filled	B-KEY
ionization	I-KEY
chamber	I-KEY
EPID	O
during	O
the	O
delivery	O
,	O
together	O
with	O
the	O
signal	O
of	O
treatment	O
progress	O
that	O
is	O
generated	O
by	O
the	O
accelerator	O
.	O
For	O
each	O
portal	O
image	O
,	O
the	O
prescribed	O
leaf	O
and	O
diaphragm	B-KEY
positions	I-KEY
were	O
computed	O
from	O
the	O
dynamic	O
prescription	O
and	O
the	O
progress	O
information	O
.	O
Motion	B-KEY
distortion	I-KEY
effects	I-KEY
of	O
the	O
leaves	O
are	O
corrected	O
based	O
on	O
the	O
treatment	O
progress	O
that	O
is	O
recorded	O
for	O
each	O
image	O
row	O
.	O
The	O
aperture	O
formed	O
by	O
the	O
prescribed	O
leaves	O
and	O
diaphragms	O
is	O
used	O
as	O
the	O
reference	B-KEY
field	I-KEY
edge	I-KEY
,	O
while	O
the	O
actual	O
field	O
edge	O
is	O
found	O
using	O
a	O
maximum-gradient	O
edge	O
detector	O
.	O
The	O
errors	O
in	O
leaf	O
and	O
diaphragm	B-KEY
position	I-KEY
are	O
found	O
from	O
the	O
deviations	O
between	O
the	O
reference	B-KEY
field	I-KEY
edge	I-KEY
and	O
the	O
detected	O
field	O
edge	O
.	O
Earlier	O
measurements	O
of	O
the	O
dynamic	O
EPID	O
response	O
show	O
that	O
the	O
accuracy	O
of	O
the	O
detected	O
field	O
edge	O
is	O
better	O
than	O
1	O
mm	O
.	O
To	O
ensure	O
that	O
the	O
verification	O
is	O
independent	O
of	O
inaccuracies	O
in	O
the	O
acquired	O
progress	O
signal	O
,	O
the	O
signal	O
was	O
checked	O
with	O
diode	O
measurements	O
beforehand	O
.	O
The	O
method	O
was	O
tested	O
on	O
three	O
different	O
dynamic	O
prescriptions	O
.	O
Using	O
the	O
described	O
method	O
,	O
we	O
correctly	O
reproduced	O
the	O
distorted	B-KEY
field	I-KEY
edges	I-KEY
.	O
Verifying	O
a	O
single	O
portal	O
image	O
took	O
0.1	O
s	O
on	O
an	O
866	O
MHz	O
personal	O
computer	O
.	O
Two	O
flaws	O
in	O
the	O
control	B-KEY
system	I-KEY
of	O
our	O
experimental	O
dynamic	B-KEY
multileaf	I-KEY
collimator	I-KEY
were	O
correctly	O
revealed	O
with	O
our	O
method	O
.	O
First	O
,	O
the	O
errors	O
in	O
leaf	B-KEY
position	I-KEY
increase	O
with	O
leaf	O
speed	O
,	O
indicating	O
a	O
delay	O
of	O
approximately	O
0.8	O
s	O
in	O
the	O
control	B-KEY
system	I-KEY
.	O
Second	O
,	O
the	O
accuracy	O
of	O
the	O
leaves	O
and	O
diaphragms	O
depends	O
on	O
the	O
direction	O
of	O
motion	O
.	O
In	O
conclusion	O
,	O
the	O
described	O
verification	O
method	O
is	O
suitable	O
for	O
detailed	O
verification	O
of	O
leaf	B-KEY
trajectories	I-KEY
during	O
dynamic	O
IMRT	O

Verification	O
of	O
timed	O
automata	O
based	O
on	O
similarity	O
The	O
paper	O
presents	O
a	O
modification	O
of	O
the	O
standard	O
partitioning	B-KEY
technique	I-KEY
to	O
generate	O
abstract	B-KEY
state	I-KEY
spaces	I-KEY
preserving	O
similarity	O
for	O
Timed	O
Automata	O
.	O
Since	O
this	O
relation	O
is	O
weaker	O
than	O
bisimilarity	B-KEY
,	O
most	O
of	O
the	O
obtained	O
models	O
-LRB-	O
state	O
spaces	O
-RRB-	O
are	O
smaller	O
than	O
bisimilar	B-KEY
ones	O
,	O
but	O
still	O
preserve	O
the	O
universal	B-KEY
fragments	I-KEY
of	O
branching	B-KEY
time	I-KEY
temporal	I-KEY
logics	I-KEY
.	O
The	O
theoretical	O
results	O
are	O
exemplified	O
for	O
strong	O
,	O
delay	O
,	O
and	O
observational	B-KEY
simulation	I-KEY
relations	I-KEY

Four-terminal	B-KEY
quantum	I-KEY
resistor	I-KEY
network	I-KEY
for	O
electron-wave	B-KEY
computing	I-KEY
Interconnected	B-KEY
ultrathin	I-KEY
conducting	I-KEY
wires	I-KEY
or	O
,	O
equivalently	O
,	O
interconnected	O
quasi-one-dimensional	O
electron	O
waveguides	O
,	O
which	O
form	O
a	O
quantum	O
resistor	O
network	O
,	O
are	O
presented	O
here	O
in	O
four-terminal	O
configurations	O
.	O
The	O
transmission	B-KEY
behaviors	I-KEY
through	O
such	O
four-terminal	O
networks	O
are	O
evaluated	O
and	O
classified	O
.	O
In	O
addition	O
,	O
we	O
show	O
that	O
such	O
networks	O
can	O
be	O
used	O
as	O
the	O
basic	O
building	O
blocks	O
for	O
a	O
possible	O
massive	O
wave	O
computing	O
machine	O
in	O
the	O
future	O
.	O
In	O
a	O
network	O
,	O
each	O
interconnection	O
,	O
a	O
node	O
point	O
,	O
is	O
an	O
elastic	O
scatterer	O
that	O
routes	O
the	O
electron	O
wave	O
.	O
Routing	O
and	O
rerouting	B-KEY
of	O
electron	O
waves	O
in	O
a	O
network	O
is	O
described	O
in	O
the	O
framework	O
of	O
quantum	O
transport	O
from	O
Landauer-Buttiker	B-KEY
theory	I-KEY
in	O
the	O
presence	O
of	O
multiple	B-KEY
elastic	I-KEY
scatterers	I-KEY
.	O
Transmissions	O
through	O
various	O
types	O
of	O
four-terminal	O
generalized	O
clean	O
Aharonov-Bohm	B-KEY
rings	I-KEY
are	O
investigated	O
at	O
zero	O
temperature	O
.	O
Useful	O
logic	B-KEY
functions	I-KEY
are	O
gathered	O
based	O
on	O
the	O
transmission	B-KEY
probability	I-KEY
to	O
each	O
terminal	O
with	O
the	O
use	O
of	O
the	O
Buttiker	B-KEY
symmetry	I-KEY
rule	I-KEY
.	O
In	O
the	O
generalized	O
rings	O
,	O
even	O
and	O
odd	O
numbers	O
of	O
terminals	O
can	O
possess	O
some	O
distinctly	O
different	O
transmission	O
characteristics	O
as	O
we	O
have	O
shown	O
here	O
and	O
earlier	O
.	O
Just	O
as	O
an	O
even	O
or	O
odd	O
number	O
of	O
atoms	O
in	O
a	O
ring	O
is	O
an	O
important	O
quantity	O
for	O
classifying	O
the	O
transmission	B-KEY
behavior	I-KEY
,	O
we	O
show	O
here	O
that	O
whether	O
the	O
number	O
of	O
terminals	O
is	O
an	O
even	O
or	O
an	O
odd	O
number	O
is	O
just	O
as	O
important	O
in	O
understanding	O
the	O
physics	O
of	O
transmission	O
through	O
such	O
a	O
ring	O
.	O
Furthermore	O
,	O
we	O
show	O
that	O
there	O
are	O
three	O
basic	O
classes	O
of	O
four-terminal	O
rings	O
and	O
the	O
scaling	O
relation	O
for	O
each	O
class	O
is	O
provided	O
.	O
In	O
particular	O
,	O
the	O
existence	O
of	O
equitransmission	O
among	O
all	O
four	O
terminals	O
is	O
shown	O
here	O
.	O
This	O
particular	O
physical	O
phenomena	O
can	O
not	O
exist	O
in	O
any	O
three-terminal	O
ring	O
.	O
Comparisons	O
and	O
discussions	O
of	O
transmission	O
characteristics	O
between	O
three-terminal	O
and	O
four-terminal	O
rings	O
are	O
also	O
presented	O
.	O
The	O
node-equation	O
approach	O
by	O
considering	O
the	O
Kirchhoff	B-KEY
current	I-KEY
conservation	I-KEY
law	I-KEY
at	O
each	O
node	O
point	O
is	O
used	O
for	O
this	O
analysis	O
.	O
Many	O
useful	O
logic	B-KEY
functions	I-KEY
for	O
electron-wave	B-KEY
computing	I-KEY
are	O
shown	O
here	O
.	O
In	O
particular	O
,	O
we	O
show	O
that	O
a	O
full	O
adder	O
can	O
be	O
constructed	O
very	O
simply	O
using	O
the	O
equitransmission	B-KEY
property	I-KEY
of	O
the	O
four-terminal	O
ring	O
.	O
This	O
is	O
in	O
sharp	O
contrast	O
with	O
circuits	O
based	O
on	O
transistor	O
logic	O

Stochastic	B-KEY
optimization	I-KEY
of	O
acoustic	B-KEY
response	I-KEY
-	O
a	O
numerical	O
and	O
experimental	O
comparison	O
The	O
objective	O
of	O
the	O
work	O
presented	O
is	O
to	O
compare	O
results	O
from	O
numerical	B-KEY
optimization	I-KEY
with	O
experimental	O
data	O
and	O
to	O
highlight	O
and	O
discuss	O
the	O
differences	O
between	O
two	O
fundamentally	O
different	O
optimization	O
methods	O
.	O
The	O
problem	O
domain	O
is	O
minimization	O
of	O
acoustic	O
emission	O
and	O
the	O
structure	B-KEY
used	O
in	O
the	O
work	O
is	O
a	O
closed	B-KEY
cylinder	I-KEY
with	O
forced	B-KEY
vibration	I-KEY
of	O
one	O
end	O
.	O
The	O
optimization	O
method	O
used	O
in	O
this	O
paper	O
is	O
simulated	B-KEY
annealing	I-KEY
-LRB-	O
SA	O
-RRB-	O
,	O
a	O
stochastic	O
method	O
.	O
The	O
results	O
are	O
compared	O
with	O
those	O
from	O
a	O
gradient-based	B-KEY
method	I-KEY
used	O
on	O
the	O
same	O
structure	B-KEY
in	O
an	O
earlier	O
paper	O
-LRB-	O
Tinnsten	O
,	O
2000	O
-RRB-	O

Robotically	O
enhanced	O
placement	O
of	O
left	B-KEY
ventricular	I-KEY
epicardial	I-KEY
electrodes	I-KEY
during	O
implantation	O
of	O
a	O
biventricular	O
implantable	O
cardioverter	O
defibrillator	O
system	O
Biventricular	O
pacing	O
has	O
gained	O
increasing	O
acceptance	O
in	O
advanced	B-KEY
heart	I-KEY
failure	I-KEY
patients	I-KEY
.	O
One	O
major	O
limitation	O
of	O
this	O
therapy	O
is	O
positioning	O
the	O
left	O
ventricular	O
stimulation	O
lead	O
via	O
the	O
coronary	B-KEY
sinus	I-KEY
.	O
This	O
report	O
demonstrates	O
the	O
feasibility	O
of	O
totally	B-KEY
endoscopic	I-KEY
direct	I-KEY
placement	I-KEY
of	O
an	O
epicardial	O
stimulation	O
lead	O
on	O
the	O
left	O
ventricle	O
using	O
the	O
daVinci	B-KEY
surgical	I-KEY
system	I-KEY

A	O
note	O
on	O
an	O
axiomatization	O
of	O
the	O
core	O
of	O
market	O
games	O
As	O
shown	O
by	O
Peleg	O
-LRB-	O
1993	O
-RRB-	O
,	O
the	O
core	O
of	O
market	O
games	O
is	O
characterized	O
by	O
nonemptiness	O
,	O
individual	B-KEY
rationality	I-KEY
,	O
superadditivity	O
,	O
the	O
weak	B-KEY
reduced	I-KEY
game	I-KEY
property	I-KEY
,	O
the	O
converse	B-KEY
reduced	I-KEY
game	I-KEY
property	I-KEY
,	O
and	O
weak	B-KEY
symmetry	I-KEY
.	O
It	O
was	O
not	O
known	O
whether	O
weak	B-KEY
symmetry	I-KEY
was	O
logically	O
independent	O
.	O
With	O
the	O
help	O
of	O
a	O
certain	O
transitive	B-KEY
4-person	I-KEY
TU	I-KEY
game	I-KEY
,	O
it	O
is	O
shown	O
that	O
weak	B-KEY
symmetry	I-KEY
is	O
redundant	B-KEY
in	O
this	O
result	O
.	O
Hence	O
,	O
the	O
core	O
on	O
market	O
games	O
is	O
axiomatized	O
by	O
the	O
remaining	O
five	O
properties	O
,	O
if	O
the	O
universe	O
of	O
players	O
contains	O
at	O
least	O
four	O
members	O

Quantum	B-KEY
limit	I-KEY
on	O
computational	B-KEY
time	I-KEY
and	O
speed	O
We	O
investigate	O
if	O
physical	O
laws	O
can	O
impose	O
limits	O
on	O
computational	B-KEY
time	I-KEY
and	O
speed	O
of	O
a	O
quantum	B-KEY
computer	I-KEY
built	O
from	O
elementary	O
particles	O
.	O
We	O
show	O
that	O
the	O
product	O
of	O
the	O
speed	O
and	O
the	O
running	O
time	O
of	O
a	O
quantum	B-KEY
computer	I-KEY
is	O
limited	O
by	O
the	O
type	O
of	O
fundamental	B-KEY
interactions	I-KEY
present	O
inside	O
the	O
system	O
.	O
This	O
will	O
help	O
us	O
to	O
decide	O
as	O
to	O
what	O
type	O
of	O
interaction	O
should	O
be	O
allowed	O
in	O
building	O
quantum	B-KEY
computers	I-KEY
in	O
achieving	O
the	O
desired	O
speed	O

The	O
development	O
of	O
virtual	O
reality	O
therapy	O
-LRB-	O
VRT	O
-RRB-	O
system	O
for	O
the	O
treatment	O
of	O
acrophobia	O
and	O
therapeutic	B-KEY
case	I-KEY
Virtual	O
reality	O
therapy	O
-LRB-	O
VRT	O
-RRB-	O
,	O
based	O
on	O
this	O
sophisticated	O
technology	O
,	O
has	O
been	O
used	O
in	O
the	O
treatment	O
of	O
subjects	O
diagnosed	O
with	O
acrophobia	O
,	O
a	O
disorder	O
that	O
is	O
characterized	O
by	O
marked	O
anxiety	O
upon	O
exposure	O
to	O
heights	O
and	O
avoidance	O
of	O
heights	O
.	O
Conventional	O
VR	O
systems	O
for	O
the	O
treatment	O
of	O
acrophobia	O
have	O
limitations	O
,	O
over-costly	O
devices	O
or	O
somewhat	O
unrealistic	O
graphic	O
scenes	O
.	O
The	O
goal	O
of	O
this	O
study	O
was	O
to	O
develop	O
an	O
inexpensive	O
and	O
more	O
realistic	B-KEY
virtual	I-KEY
environment	I-KEY
-LRB-	O
VE	O
-RRB-	O
in	O
which	O
to	O
perform	O
exposure	B-KEY
therapy	I-KEY
for	O
acrophobia	O
.	O
It	O
is	O
based	O
on	O
a	O
personal	B-KEY
computer	I-KEY
,	O
and	O
a	O
virtual	B-KEY
scene	I-KEY
of	O
a	O
bunge-jump	O
tower	O
in	O
the	O
middle	O
of	O
a	O
large	O
city	O
.	O
The	O
virtual	O
scenario	O
includes	O
an	O
open	O
lift	O
surrounded	O
by	O
props	O
beside	O
a	O
tower	O
,	O
which	O
allows	O
the	O
patient	O
to	O
feel	O
a	O
sense	O
of	O
heights	O
.	O
The	O
effectiveness	O
of	O
the	O
VE	O
was	O
evaluated	O
through	O
the	O
clinical	B-KEY
treatment	I-KEY
of	O
a	O
subject	O
who	O
was	O
suffering	O
from	O
the	O
fear	O
of	O
heights	O
.	O
As	O
a	O
result	O
,	O
it	O
was	O
proved	O
that	O
this	O
VR	O
environment	O
was	O
effective	O
and	O
realistic	O
at	O
overcoming	O
acrophobia	O
according	O
not	O
only	O
to	O
the	O
comparison	O
results	O
of	O
a	O
variety	O
of	O
questionnaires	O
before	O
and	O
after	O
treatment	O
but	O
also	O
to	O
the	O
subject	O
's	O
comments	O
that	O
the	O
VE	O
seemed	O
to	O
evoke	O
more	O
fearful	O
feelings	O
than	O
the	O
real	O
situation	O

Oracle	O
's	O
Suite	O
grows	O
up	O
Once	O
a	O
low-cost	O
Web	O
offering	O
,	O
Oracle	O
's	O
Small	O
Business	O
Suite	O
now	O
carries	O
a	O
price	O
tag	O
to	O
justify	O
VAR	O
interest	O

From	O
the	O
DOS	O
dog	O
days	O
to	O
e-filing	B-KEY
-LSB-	O
law	O
firms	O
-RSB-	O
The	O
poster	O
child	O
for	O
a	O
successful	O
e-filing	B-KEY
venture	O
is	O
the	O
Case	B-KEY
Management	I-KEY
and	I-KEY
Electronic	I-KEY
Case	I-KEY
File	I-KEY
system	I-KEY
now	O
rolling	O
through	O
the	O
district	O
and	O
bankruptcy	O
courts	O
.	O
A	O
project	O
of	O
the	O
Administrative	O
Office	O
of	O
the	O
United	B-KEY
States	I-KEY
Courts	I-KEY
,	O
CM/ECF	O
is	O
a	O
loud	O
proponent	O
of	O
the	O
benefits	O
of	O
the	O
PDF	B-KEY
approach	O
and	O
it	O
has	O
a	O
full	O
head	O
of	O
steam	O
.	O
Present	O
plans	O
are	O
for	O
all	O
federal	O
courts	O
to	O
implement	O
CM/ECF	O
by	O
2005	O
.	O
That	O
means	O
a	O
radical	O
shift	O
in	O
methodology	O
and	O
tools	O
for	O
a	O
lot	O
of	O
lawyers	O
.	O
It	O
also	O
means	O
that	O
you	O
should	O
get	O
cozy	O
with	O
Acrobat	O
real	O
soon	O

Operations	O
that	O
do	O
not	O
disturb	O
partially	B-KEY
known	I-KEY
quantum	I-KEY
states	I-KEY
Consider	O
a	O
situation	O
in	O
which	O
a	O
quantum	B-KEY
system	I-KEY
is	O
secretly	O
prepared	O
in	O
a	O
state	O
chosen	O
from	O
the	O
known	O
set	O
of	O
states	O
.	O
We	O
present	O
a	O
principle	O
that	O
gives	O
a	O
definite	O
distinction	O
between	O
the	O
operations	O
that	O
preserve	O
the	O
states	O
of	O
the	O
system	O
and	O
those	O
that	O
disturb	O
the	O
states	O
.	O
The	O
principle	O
is	O
derived	O
by	O
alternately	O
applying	O
a	O
fundamental	O
property	O
of	O
classical	B-KEY
signals	I-KEY
and	O
a	O
fundamental	O
property	O
of	O
quantum	O
ones	O
.	O
The	O
principle	O
can	O
be	O
cast	O
into	O
a	O
simple	O
form	O
by	O
using	O
a	O
decomposition	O
of	O
the	O
relevant	O
Hilbert	B-KEY
space	I-KEY
,	O
which	O
is	O
uniquely	O
determined	O
by	O
the	O
set	O
of	O
possible	O
states	O
.	O
The	O
decomposition	O
implies	O
the	O
classification	O
of	O
the	O
degrees	B-KEY
of	I-KEY
freedom	I-KEY
of	O
the	O
system	O
into	O
three	O
parts	O
depending	O
on	O
how	O
they	O
store	O
the	O
information	O
on	O
the	O
initially	O
chosen	O
state	O
:	O
one	O
storing	O
it	O
classically	O
,	O
one	O
storing	O
it	O
nonclassically	O
,	O
and	O
the	O
other	O
one	O
storing	O
no	O
information	O
.	O
Then	O
the	O
principle	O
states	O
that	O
the	O
nonclassical	B-KEY
part	I-KEY
is	O
inaccessible	O
and	O
the	O
classical	O
part	O
is	O
read-only	O
if	O
we	O
are	O
to	O
preserve	O
the	O
state	O
of	O
the	O
system	O
.	O
From	O
this	O
principle	O
,	O
many	O
types	O
of	O
no-cloning	O
,	O
no-broadcasting	O
,	O
and	O
no-imprinting	O
conditions	O
can	O
easily	O
be	O
derived	O
in	O
general	O
forms	O
including	O
mixed	O
states	O
.	O
It	O
also	O
gives	O
a	O
unified	O
view	O
on	O
how	O
various	O
schemes	O
of	O
quantum	B-KEY
cryptography	I-KEY
work	O
.	O
The	O
principle	O
helps	O
one	O
to	O
derive	O
optimum	O
amount	O
of	O
resources	O
-LRB-	O
bits	B-KEY
,	O
qubits	B-KEY
,	O
and	O
ebits	B-KEY
-RRB-	O
required	O
in	O
data	O
compression	O
or	O
in	O
quantum	B-KEY
teleportation	I-KEY
of	O
mixed-state	B-KEY
ensembles	I-KEY

The	O
role	O
and	O
future	O
of	O
subject	B-KEY
classification	I-KEY
:	O
the	O
exploitation	O
of	O
resources	O
It	O
is	O
imperative	O
that	O
the	O
library	B-KEY
information	I-KEY
systems	I-KEY
-LRB-	O
LIS	B-KEY
-RRB-	O
profession	O
and	O
LIS	B-KEY
educators	O
appreciate	O
fully	O
the	O
contribution	O
that	O
classification	O
makes	O
to	O
the	O
discipline	O
and	O
that	O
it	O
is	O
no	O
longer	O
seen	O
as	O
the	O
domain	O
of	O
the	O
academic	O
,	O
isolated	O
theorist	O
,	O
but	O
becomes	O
an	O
integral	O
part	O
of	O
our	O
understanding	O
of	O
the	O
contribution	O
that	O
the	O
LIS	B-KEY
community	O
can	O
make	O
to	O
society	O
as	O
a	O
whole	O
-	O
as	O
well	O
as	O
to	O
particular	O
areas	O
such	O
as	O
legal	B-KEY
information	I-KEY

Access	O
matters	O
Discusses	O
accessibility	B-KEY
needs	I-KEY
of	O
people	O
with	O
disabilities	O
,	O
both	O
from	O
the	O
perspective	O
of	O
getting	O
the	O
information	O
from	O
I&R	O
programs	O
-LRB-	O
including	O
accessible	B-KEY
Web	I-KEY
sites	I-KEY
,	O
TTY	B-KEY
access	I-KEY
,	O
Braille	B-KEY
,	O
and	O
other	O
mechanisms	O
-RRB-	O
and	O
from	O
the	O
perspective	O
of	O
being	O
aware	O
of	O
accessibility	B-KEY
needs	I-KEY
when	O
referring	O
clients	O
to	O
resources	O
.	O
Includes	O
information	O
on	O
ADA	B-KEY
legislation	I-KEY
requiring	O
accessibility	O
to	O
public	B-KEY
places	I-KEY
and	O
recommends	O
several	O
organizations	O
and	O
Web	O
sites	O
for	O
additional	O
information	O

Implementation	O
and	O
performance	B-KEY
evaluation	I-KEY
of	O
a	O
FIFO	B-KEY
queue	I-KEY
class	B-KEY
library	I-KEY
for	O
time	O
warp	O
The	O
authors	O
describe	O
the	O
implementation	O
,	O
use	O
,	O
and	O
performance	B-KEY
evaluation	I-KEY
of	O
a	O
FIFO	B-KEY
queue	I-KEY
class	B-KEY
library	I-KEY
by	O
means	O
of	O
a	O
high-performance	O
,	O
easy-to-use	B-KEY
interface	I-KEY
employed	O
for	O
queuing	B-KEY
simulations	I-KEY
in	O
parallel	B-KEY
discrete	I-KEY
simulations	I-KEY
based	O
on	O
the	O
time	O
warp	O
method	O
.	O
Various	O
general-purpose	B-KEY
simulation	I-KEY
libraries	I-KEY
and	O
languages	O
have	O
been	O
proposed	O
,	O
and	O
among	O
these	O
some	O
have	O
the	O
advantage	O
of	O
not	O
requiring	O
users	O
to	O
define	O
anything	O
other	O
than	O
the	O
state	B-KEY
vector	I-KEY
,	O
and	O
not	O
needing	O
awareness	O
of	O
rollback	O
under	O
a	O
platform	O
which	O
performs	O
state	O
control	O
based	O
on	O
copies	O
.	O
However	O
,	O
because	O
the	O
state	B-KEY
vectors	I-KEY
must	O
be	O
defined	O
as	O
simple	O
data	O
structures	O
without	O
pointers	O
,	O
dynamic	B-KEY
data	I-KEY
structures	I-KEY
such	O
as	O
a	O
FIFO	B-KEY
queue	I-KEY
can	O
not	O
be	O
handled	O
directly	O
.	O
Under	O
the	O
proposed	O
class	B-KEY
library	I-KEY
,	O
both	O
the	O
platform	O
and	O
the	O
user	O
can	O
handle	O
such	O
structures	O
in	O
the	O
same	O
fashion	O
that	O
embedded	B-KEY
data	I-KEY
structures	I-KEY
are	O
handled	O
.	O
In	O
addition	O
,	O
instead	O
of	O
all	O
stored	O
data	O
,	O
just	O
the	O
operational	B-KEY
history	I-KEY
can	O
be	O
stored	O
and	O
recovered	O
efficiently	O
at	O
an	O
effectively	O
minimal	O
cost	O
by	O
taking	O
advantage	O
of	O
the	O
first-in-first-out	B-KEY
characteristics	I-KEY
of	O
the	O
above	O
data	O
structures	O
.	O
When	O
the	O
kernel	O
deletes	O
past	O
state	O
histories	O
during	O
a	O
simulation	O
,	O
garbage	B-KEY
collection	I-KEY
is	O
also	O
performed	O
transparently	O
using	O
the	O
corresponding	O
method	O

Adaptive	B-KEY
wavelet	I-KEY
methods	I-KEY
.	O
II	O
.	O
Beyond	O
the	O
elliptic	B-KEY
case	I-KEY
This	O
paper	O
is	O
concerned	O
with	O
the	O
design	O
and	O
analysis	O
of	O
adaptive	B-KEY
wavelet	I-KEY
methods	I-KEY
for	O
systems	O
of	O
operator	B-KEY
equations	I-KEY
.	O
Its	O
main	O
accomplishment	O
is	O
to	O
extend	O
the	O
range	O
of	O
applicability	O
of	O
the	O
adaptive	O
wavelet-based	O
method	O
developed	O
previously	O
for	O
symmetric	O
positive	O
definite	O
problems	O
to	O
indefinite	O
or	O
unsymmetric	O
systems	O
of	O
operator	B-KEY
equations	I-KEY
.	O
This	O
is	O
accomplished	O
by	O
first	O
introducing	O
techniques	O
-LRB-	O
such	O
as	O
the	O
least	B-KEY
squares	I-KEY
formulation	I-KEY
developed	O
previously	O
-RRB-	O
that	O
transform	O
the	O
original	O
-LRB-	O
continuous	O
-RRB-	O
problem	O
into	O
an	O
equivalent	O
infinite	O
system	O
of	O
equations	O
which	O
is	O
now	O
well-posed	O
in	O
the	O
Euclidean	B-KEY
metric	I-KEY
.	O
It	O
is	O
then	O
shown	O
how	O
to	O
utilize	O
adaptive	O
techniques	O
to	O
solve	O
the	O
resulting	O
infinite	O
system	O
of	O
equations	O
.	O
It	O
is	O
shown	O
that	O
for	O
a	O
wide	O
range	O
of	O
problems	O
,	O
this	O
new	O
adaptive	O
method	O
performs	O
with	O
asymptotically	B-KEY
optimal	I-KEY
complexity	I-KEY
,	O
i.e.	O
,	O
it	O
recovers	O
an	O
approximate	O
solution	O
with	O
desired	O
accuracy	O
at	O
a	O
computational	O
expense	O
that	O
stays	O
proportional	O
to	O
the	O
number	O
of	O
terms	O
in	O
a	O
corresponding	O
wavelet-best	O
N-term	B-KEY
approximation	I-KEY
.	O
An	O
important	O
advantage	O
of	O
this	O
adaptive	O
approach	O
is	O
that	O
it	O
automatically	O
stabilizes	O
the	O
numerical	O
procedure	O
so	O
that	O
,	O
for	O
instance	O
,	O
compatibility	O
constraints	O
on	O
the	O
choice	O
of	O
trial	O
spaces	O
,	O
like	O
the	O
LBB	O
condition	O
,	O
no	O
longer	O
arise	O

A	O
gendered	O
view	O
of	O
computer	O
professionals	O
:	O
preliminary	O
results	O
of	O
a	O
survey	O
The	O
under-representation	O
of	O
women	O
in	O
the	O
computing	B-KEY
profession	I-KEY
in	O
many	O
parts	O
the	O
western	O
world	O
has	O
received	O
our	O
attention	O
through	O
numerous	O
publications	O
,	O
the	O
noticeable	O
low	O
representation	O
of	O
women	O
at	O
computer	O
science	O
conferences	O
and	O
in	O
the	O
lecture	O
halls	O
.	O
Over	O
the	O
past	O
two	O
decades	O
,	O
the	O
situation	O
had	O
become	O
worse	O
.	O
This	O
paper	O
seeks	O
to	O
add	O
to	O
the	O
dialogue	O
by	O
presenting	O
preliminary	O
findings	O
from	O
a	O
research	O
project	O
conducted	O
in	O
four	O
countries	O
.	O
The	O
aim	O
of	O
this	O
research	O
was	O
to	O
gain	O
an	O
insight	O
into	O
the	O
perceptions	O
future	O
computer	O
professionals	O
hold	O
on	O
the	O
category	O
of	O
employment	B-KEY
loosely	O
defined	O
under	O
the	O
term	O
of	O
``	O
a	O
computer	O
professional	O
.	O
''	O
One	O
goal	O
was	O
to	O
get	O
insight	O
into	O
whether	O
or	O
not	O
there	O
is	O
a	O
difference	O
between	O
female	O
and	O
mate	B-KEY
students	I-KEY
regarding	O
their	O
view	O
of	O
computer	O
professionals	O
.	O
Other	O
goals	O
were	O
to	O
determine	O
if	O
there	O
was	O
any	O
difference	O
between	O
female	O
and	O
male	O
students	O
in	O
different	O
parts	O
of	O
the	O
world	O
,	O
as	O
well	O
as	O
who	O
or	O
what	O
most	O
influences	O
the	O
students	O
to	O
undertake	O
their	O
courses	O
in	O
computing	O

Development	O
of	O
a	O
3.5	B-KEY
inch	I-KEY
magneto-optical	B-KEY
disk	I-KEY
with	O
a	O
capacity	O
of	O
2.3	B-KEY
GB	I-KEY
The	O
recording	B-KEY
capacity	I-KEY
of	O
GIGAMO	B-KEY
media	I-KEY
was	O
enlarged	O
from	O
1.3	O
GB	O
to	O
2.3	B-KEY
GB	I-KEY
for	O
3.5	B-KEY
inch	I-KEY
magneto-optical	O
-LRB-	O
MO	O
-RRB-	O
disks	O
while	O
maintaining	O
downward	O
compatibility	O
.	O
For	O
the	O
new	O
GIGAMO	O
technology	O
,	O
a	O
land	O
and	O
groove	O
recording	O
method	O
was	O
applied	O
in	O
addition	O
to	O
magnetically	B-KEY
induced	I-KEY
super	I-KEY
resolution	I-KEY
-LRB-	O
MSR	B-KEY
-RRB-	O
media	O
.	O
Furthermore	O
,	O
a	O
novel	O
address	B-KEY
format	I-KEY
suitable	O
for	O
the	O
land	O
and	O
groove	O
recording	O
method	O
was	O
adopted	O
.	O
The	O
specifications	O
of	O
the	O
new	O
GIGAMO	B-KEY
media	I-KEY
were	O
examined	O
to	O
satisfy	O
requirements	O
for	O
practical	O
use	O
with	O
respect	O
to	O
margins	O
.	O
Durability	O
of	O
more	O
than	O
10/sup	O
6	O
/	O
rewritings	O
and	O
an	O
enough	O
lifetime	B-KEY
were	O
confirmed	O

Manufacturing	B-KEY
data	I-KEY
analysis	I-KEY
of	O
machine	B-KEY
tool	I-KEY
errors	I-KEY
within	O
a	O
contemporary	B-KEY
small	I-KEY
manufacturing	I-KEY
enterprise	I-KEY
The	O
main	O
focus	O
of	O
the	O
paper	O
is	O
directed	O
at	O
the	O
determination	O
of	O
manufacturing	O
errors	O
within	O
the	O
contemporary	O
smaller	O
manufacturing	O
enterprise	O
sector	O
.	O
The	O
manufacturing	O
error	O
diagnosis	O
is	O
achieved	O
through	O
the	O
manufacturing	B-KEY
data	I-KEY
analysis	I-KEY
of	O
the	O
results	O
obtained	O
from	O
the	O
inspection	B-KEY
of	O
the	O
component	O
on	O
a	O
co-ordinate	B-KEY
measuring	I-KEY
machine	I-KEY
.	O
This	O
manufacturing	B-KEY
data	I-KEY
analysis	I-KEY
activity	O
adopts	O
a	O
feature-based	B-KEY
approach	I-KEY
and	O
is	O
conducted	O
through	O
the	O
application	O
of	O
a	O
forward	B-KEY
chaining	I-KEY
expert	I-KEY
system	I-KEY
,	O
called	O
the	O
product	B-KEY
data	I-KEY
analysis	I-KEY
distributed	I-KEY
diagnostic	I-KEY
expert	I-KEY
system	I-KEY
,	O
which	O
forms	O
part	O
of	O
a	O
larger	O
prototype	O
feedback	O
system	O
entitled	O
the	O
production	O
data	O
analysis	O
framework	O
.	O
The	O
paper	O
introduces	O
the	O
manufacturing	O
error	O
categorisations	O
that	O
are	O
associated	O
with	O
milling	B-KEY
type	I-KEY
operations	I-KEY
,	O
knowledge	B-KEY
acquisition	I-KEY
and	O
representation	O
,	O
conceptual	B-KEY
structure	I-KEY
and	O
operating	B-KEY
procedure	I-KEY
of	O
the	O
prototype	O
manufacturing	B-KEY
data	I-KEY
analysis	I-KEY
facility	O
.	O
The	O
paper	O
concludes	O
with	O
a	O
brief	O
evaluation	O
of	O
the	O
logic	O
employed	O
through	O
the	O
simulation	O
of	O
manufacturing	O
error	O
scenarios	O
.	O
This	O
prototype	O
manufacturing	B-KEY
data	I-KEY
analysis	I-KEY
expert	O
system	O
provides	O
a	O
valuable	O
aid	O
for	O
the	O
rapid	O
diagnosis	O
and	O
elimination	O
of	O
manufacturing	O
errors	O
on	O
a	O
3-axis	B-KEY
vertical	I-KEY
machining	I-KEY
centre	I-KEY
in	O
an	O
environment	O
where	O
operator	O
expertise	O
is	O
limited	O

Vibration	O
control	O
of	O
the	O
rotating	B-KEY
flexible-shaft/multi-flexible-disk	I-KEY
system	I-KEY
with	O
the	O
eddy-current	B-KEY
damper	I-KEY
In	O
this	O
paper	O
,	O
the	O
rotating	B-KEY
flexible-Timoshenko-shaft/flexible-disk	I-KEY
coupling	I-KEY
system	I-KEY
is	O
formulated	O
by	O
applying	O
the	O
assumed-mode	B-KEY
method	I-KEY
into	O
the	O
kinetic	O
and	O
strain	O
energies	O
,	O
and	O
the	O
virtual	B-KEY
work	I-KEY
done	O
by	O
the	O
eddy-current	B-KEY
damper	I-KEY
.	O
From	O
Lagrange	O
's	O
equations	O
,	O
the	O
resulting	O
discretized	B-KEY
equations	I-KEY
of	I-KEY
motion	I-KEY
can	O
be	O
simplified	O
as	O
a	O
bilinear	B-KEY
system	I-KEY
-LRB-	O
BLS	O
-RRB-	O
.	O
Introducing	O
the	O
control	O
laws	O
,	O
including	O
the	O
quadratic	O
,	O
nonlinear	O
and	O
optimal	B-KEY
feedback	I-KEY
control	I-KEY
laws	I-KEY
,	O
into	O
the	O
BLS	O
,	O
it	O
is	O
found	O
that	O
the	O
eddy-current	B-KEY
damper	I-KEY
can	O
be	O
used	O
to	O
suppress	O
flexible	O
and	O
shear	B-KEY
vibrations	I-KEY
simultaneously	O
,	O
and	O
the	O
system	O
is	O
globally	O
asymptotically	O
stable	O
.	O
Numerical	O
results	O
are	O
provided	O
to	O
validate	O
the	O
theoretical	O
analysis	O

TCP	B-KEY
explicit	I-KEY
congestion	I-KEY
notification	I-KEY
over	O
ATM-UBR	B-KEY
:	O
a	O
simulation	B-KEY
study	O
The	O
enhancement	O
of	O
transmission	O
control	O
protocol	O
's	O
-LRB-	O
TCP	O
's	O
-RRB-	O
congestion	B-KEY
control	I-KEY
mechanisms	I-KEY
using	O
explicit	O
congestion	O
notification	O
-LRB-	O
ECN	O
-RRB-	O
over	O
asynchronous	O
transfer	O
mode	O
-LRB-	O
ATM	O
-RRB-	O
networks	O
is	O
overviewed	O
.	O
TCP	O
's	O
congestion	O
control	O
is	O
enhanced	O
so	O
that	O
congestion	O
is	O
indicated	O
by	O
not	O
only	O
packet	B-KEY
losses	I-KEY
as	O
is	O
currently	O
the	O
case	O
but	O
an	O
agent	B-KEY
implemented	O
at	O
the	O
ATM	B-KEY
network	I-KEY
's	O
edge	O
as	O
well	O
.	O
The	O
novel	O
idea	O
uses	O
EFCI	O
-LRB-	O
explicit	O
forward	O
congestion	O
indication	O
-RRB-	O
bits	O
-LRB-	O
available	O
in	O
every	O
ATM	O
cell	O
header	O
-RRB-	O
to	O
generalize	O
the	O
ECN	O
response	O
to	O
the	O
UBR	O
-LRB-	O
unspecified	O
bit	O
rate	O
-RRB-	O
service	O
,	O
notify	O
congestion	O
,	O
and	O
adjust	O
the	O
credit-based	B-KEY
window	I-KEY
size	I-KEY
of	O
the	O
TCR	O
.	O
The	O
authors	O
'	O
simulation	B-KEY
experiments	O
show	O
that	O
TCP	O
ECN	O
achieves	O
significantly	O
lower	O
cell	B-KEY
loss	I-KEY
,	O
packet	B-KEY
retransmissions	I-KEY
,	O
and	O
buffer	B-KEY
utilization	I-KEY
,	O
and	O
exhibits	O
better	O
throughput	B-KEY
than	O
-LRB-	O
non-ECN	O
-RRB-	O
TCP	O
Reno	O

ICANN	B-KEY
and	O
Internet	B-KEY
governance	I-KEY
:	O
leveraging	O
technical	B-KEY
coordination	I-KEY
to	O
realize	O
global	B-KEY
public	I-KEY
policy	I-KEY
The	O
Internet	B-KEY
Corporation	I-KEY
for	I-KEY
Assigned	I-KEY
Names	I-KEY
and	I-KEY
Numbers	I-KEY
-LRB-	O
ICANN	B-KEY
-RRB-	O
was	O
created	O
in	O
1998	O
to	O
perform	O
technical	B-KEY
coordination	I-KEY
of	O
the	O
Internet	O
.	O
ICANN	B-KEY
also	O
lays	O
the	O
foundations	O
for	O
governance	O
,	O
creating	O
capabilities	O
for	O
promulgating	O
and	O
enforcing	O
global	B-KEY
regulations	I-KEY
on	O
Internet	B-KEY
use	I-KEY
.	O
ICANN	B-KEY
leverages	O
the	O
capabilities	O
in	O
the	O
Internet	B-KEY
domain	I-KEY
name	I-KEY
system	I-KEY
-LRB-	O
DNS	O
-RRB-	O
to	O
implement	O
four	O
mechanisms	O
of	O
governance	O
:	O
authority	O
,	O
law	O
,	O
sanctions	O
,	O
and	O
jurisdictions	O
.	O
These	O
governance-related	B-KEY
features	I-KEY
are	O
embodied	O
in	O
seemingly	O
technical	O
features	O
of	O
ICANN	B-KEY
's	O
institutional	B-KEY
design	I-KEY
.	O
Recognition	O
of	O
ICANN	B-KEY
's	O
governance	O
mechanisms	O
allows	O
us	O
to	O
better	O
understand	O
the	O
Internet	O
's	O
emerging	O
regulatory	B-KEY
regime	I-KEY

A	O
new	O
approach	O
to	O
the	O
d-MC	B-KEY
problem	I-KEY
Many	O
real-world	O
systems	O
are	O
multi-state	B-KEY
systems	I-KEY
composed	O
of	O
multi-state	B-KEY
components	I-KEY
in	O
which	O
the	O
reliability	O
can	O
be	O
computed	O
in	O
terms	O
of	O
the	O
lower	O
bound	O
points	O
of	O
level	O
d	O
,	O
called	O
d-Mincuts	B-KEY
-LRB-	O
d-MCs	O
-RRB-	O
.	O
Such	O
systems	O
-LRB-	O
electric	O
power	O
,	O
transportation	O
,	O
etc.	O
-RRB-	O
may	O
be	O
regarded	O
as	O
flow	B-KEY
networks	I-KEY
whose	O
arcs	O
have	O
independent	O
,	O
discrete	O
,	O
limited	O
and	O
multi-valued	O
random	O
capacities	O
.	O
In	O
this	O
paper	O
,	O
all	O
MCs	O
are	O
assumed	O
to	O
be	O
known	O
in	O
advance	O
,	O
and	O
the	O
authors	O
focused	O
on	O
how	O
to	O
verify	O
each	O
d-MC	O
candidate	O
before	O
using	O
d-MCs	O
to	O
calculate	O
the	O
network	O
reliability	O
.	O
The	O
proposed	O
algorithm	O
is	O
more	O
efficient	O
than	O
existing	O
algorithms	O
.	O
The	O
algorithm	O
runs	O
in	O
O	O
-LRB-	O
p	O
sigma	O
mn	O
-RRB-	O
time	O
,	O
a	O
significant	O
improvement	O
over	O
the	O
previous	O
O	O
-LRB-	O
p	O
sigma	O
m/sup	O
2	O
/	O
-RRB-	O
time	B-KEY
bounds	I-KEY
based	O
on	O
max-flow/min-cut	B-KEY
,	O
where	O
p	O
and	O
or	O
are	O
the	O
number	O
of	O
MCs	O
and	O
d-MC	O
candidates	O
,	O
respectively	O
.	O
It	O
is	O
simple	O
,	O
intuitive	O
and	O
uses	O
no	O
complex	O
data	O
structures	O
.	O
An	O
example	O
is	O
given	O
to	O
show	O
how	O
all	O
d-MC	O
candidates	O
are	O
found	O
and	O
verified	O
by	O
the	O
proposed	O
algorithm	O
.	O
Then	O
the	O
reliability	O
of	O
this	O
example	O
is	O
computed	O

Unlocking	O
the	O
clubhouse	O
:	O
the	O
Carnegie	O
Mellon	O
experience	O
In	O
the	O
fall	O
of	O
1995	O
,	O
just	O
seven	O
of	O
95	O
students	B-KEY
entering	O
the	O
undergraduate	B-KEY
program	I-KEY
in	O
computer	O
science	O
at	O
Carnegie	B-KEY
Mellon	I-KEY
University	I-KEY
were	O
women	B-KEY
.	O
In	O
2000	O
,	O
54	O
of	O
130	O
,	O
or	O
42	O
%	O
,	O
were	O
women	B-KEY
.	O
What	O
happened	O
?	O
This	O
article	O
presents	O
a	O
brief	O
history	B-KEY
of	O
the	O
transformation	O
at	O
Carnegie	O
Mellon	O
's	O
School	O
of	O
Computer	O
Science	O
,	O
and	O
the	O
research	B-KEY
project	I-KEY
that	O
lay	O
behind	O
it	O

A	O
knowledge	B-KEY
management	I-KEY
framework	I-KEY
for	O
the	O
support	O
of	O
decision	O
making	O
in	O
humanitarian	O
assistance/disaster	O
relief	O
The	O
major	O
challenge	O
in	O
current	O
humanitarian	O
assistance/disaster	O
relief	O
-LRB-	O
HA/DR	O
-RRB-	O
efforts	O
is	O
that	O
diverse	O
information	O
and	O
knowledge	O
are	O
widely	O
distributed	O
and	O
owned	O
by	O
different	O
organizations	O
.	O
These	O
resources	O
are	O
not	O
efficiently	O
organized	B-KEY
and	O
utilized	O
during	O
HA/DR	O
operations	O
.	O
We	O
present	O
a	O
knowledge	B-KEY
management	I-KEY
framework	I-KEY
that	O
integrates	O
multiple	O
information	B-KEY
technologies	I-KEY
to	O
collect	O
,	O
analyze	O
,	O
and	O
manage	O
information	O
and	O
knowledge	O
for	O
supporting	O
decision	O
making	O
in	O
HA/DR	O
.	O
The	O
framework	O
will	O
help	O
identify	O
the	O
information	B-KEY
needs	I-KEY
,	O
be	O
aware	O
of	O
a	O
disaster	O
situation	O
,	O
and	O
provide	O
decision-makers	O
with	O
useful	O
relief	O
recommendations	O
based	O
on	O
past	O
experience	O
.	O
A	O
comprehensive	O
,	O
consistent	O
and	O
authoritative	O
knowledge	O
base	O
within	O
the	O
framework	O
will	O
facilitate	O
knowledge	B-KEY
sharing	I-KEY
and	O
reuse	O
.	O
This	O
framework	O
can	O
also	O
be	O
applied	O
to	O
other	O
similar	O
real-time	B-KEY
decision-making	I-KEY
environments	I-KEY
,	O
such	O
as	O
crisis	B-KEY
management	I-KEY
and	O
emergency	B-KEY
medical	I-KEY
assistance	I-KEY

Bad	B-KEY
pixel	I-KEY
identification	I-KEY
by	O
means	O
of	O
principal	B-KEY
components	I-KEY
analysis	I-KEY
Bad	O
pixels	O
are	O
defined	O
as	O
those	O
pixels	O
showing	O
a	O
temporal	B-KEY
evolution	I-KEY
of	O
the	O
signal	O
different	O
from	O
the	O
rest	O
of	O
the	O
pixels	O
of	O
a	O
given	O
array	O
.	O
Principal	B-KEY
component	I-KEY
analysis	I-KEY
helps	O
us	O
to	O
understand	O
the	O
definition	O
of	O
a	O
statistical	B-KEY
distance	I-KEY
associated	O
with	O
each	O
pixels	O
,	O
and	O
using	O
this	O
distance	O
it	O
is	O
possible	O
to	O
identify	O
those	O
pixels	O
labeled	O
as	O
bad	O
pixels	O
.	O
The	O
spatiality	O
of	O
a	O
pixel	O
is	O
also	O
calculated	O
.	O
An	O
assumption	O
about	O
the	O
normality	O
of	O
the	O
distribution	O
of	O
the	O
distances	O
of	O
the	O
pixels	O
is	O
revised	O
.	O
Although	O
the	O
influence	O
on	O
the	O
robustness	B-KEY
of	O
the	O
identification	B-KEY
algorithm	I-KEY
is	O
negligible	O
,	O
the	O
definition	O
of	O
a	O
parameter	O
related	O
with	O
this	O
nonnormality	O
helps	O
to	O
identify	O
those	O
principal	O
components	O
and	O
eigenimages	B-KEY
responsible	O
for	O
the	O
departure	O
from	O
a	O
multinormal	B-KEY
distribution	I-KEY
.	O
The	O
method	O
for	O
identifying	O
the	O
bad	O
pixels	O
is	O
successfully	O
applied	O
to	O
a	O
set	O
of	O
frames	O
obtained	O
from	O
a	O
CCD	O
visible	O
and	O
a	O
focal	B-KEY
plane	I-KEY
array	I-KEY
-LRB-	O
FPA	O
-RRB-	O
IR	B-KEY
camera	I-KEY

It	O
's	O
time	O
to	O
buy	O
There	O
is	O
an	O
upside	O
to	O
a	O
down	O
economy	O
:	O
over-zealous	O
suppliers	B-KEY
are	O
willing	O
to	O
make	O
deals	O
that	O
were	O
unthinkable	O
a	O
few	O
years	O
ago	O
.	O
That	O
's	O
because	O
vendors	B-KEY
are	O
experiencing	O
the	O
same	O
money	B-KEY
squeeze	I-KEY
as	O
manufacturers	O
,	O
which	O
makes	O
the	O
year	O
2002	O
the	O
perfect	O
time	O
to	O
invest	O
in	O
new	O
technology	O
.	O
The	O
author	O
states	O
that	O
when	O
negotiating	B-KEY
the	O
deal	O
,	O
provisions	O
for	O
unexpected	B-KEY
costs	I-KEY
,	O
an	O
exit	B-KEY
strategy	I-KEY
,	O
and	O
even	O
shared	B-KEY
risk	I-KEY
with	O
the	O
vendor	B-KEY
should	O
be	O
on	O
the	O
table	O

Lifting	B-KEY
factorization	I-KEY
of	O
discrete	O
W	O
transform	O
A	O
general	O
method	O
is	O
proposed	O
to	O
factor	O
the	O
type-IV	O
discrete	O
W	O
transform	O
-LRB-	O
DWT-IV	O
-RRB-	O
into	O
lifting	O
steps	O
and	O
additions	O
.	O
Then	O
,	O
based	O
on	O
the	O
relationships	O
among	O
various	O
types	O
of	O
DWTs	B-KEY
,	O
four	O
types	O
of	O
DWTs	B-KEY
are	O
factored	O
into	O
lifting	O
steps	O
and	O
additions	O
.	O
After	O
approximating	O
the	O
lifting	B-KEY
matrices	I-KEY
,	O
we	O
get	O
four	O
types	O
of	O
new	O
integer	O
DWTs	B-KEY
-LRB-	O
IntDWT-I	O
,	O
IntDWT-II	O
,	O
IntDWT-III	O
,	O
and	O
IntDWT-IV	O
-RRB-	O
which	O
are	O
floating-point	O
multiplication	O
free	O
.	O
Integer-to-integer	O
transforms	O
-LRB-	O
II-DWT	O
-RRB-	O
,	O
which	O
approximate	O
to	O
DWT	O
,	O
are	O
also	O
proposed	O
.	O
Fast	O
algorithms	O
are	O
given	O
for	O
the	O
new	O
transforms	O
and	O
their	O
computational	B-KEY
complexities	I-KEY
are	O
analyzed	O

Disposable	O
mobiles	O
After	O
many	O
delays	O
,	O
the	O
reusable	B-KEY
,	O
recyclable	B-KEY
,	O
disposable	B-KEY
mobile	I-KEY
phone	I-KEY
is	O
finally	O
going	O
on	O
sale	O
in	O
the	O
US	O
.	O
But	O
with	O
a	O
business	O
model	O
largely	O
dependent	O
on	O
niche	O
markets	O
,	O
Elizabeth	O
Biddlecombe	O
asks	O
if	O
these	O
simplified	B-KEY
handsets	I-KEY
will	O
be	O
good	O
enough	O
to	O
survive	O
a	O
brutal	O
market	O

Knowledge	B-KEY
organisation	I-KEY
of	O
product	B-KEY
design	I-KEY
blackboard	I-KEY
systems	I-KEY
via	O
graph	B-KEY
decomposition	I-KEY
Knowledge	B-KEY
organisation	I-KEY
plays	O
an	O
important	O
role	O
in	O
building	O
a	O
knowledge-based	B-KEY
product	I-KEY
design	I-KEY
blackboard	O
system	O
.	O
Well-organised	O
knowledge	O
sources	O
will	O
facilitate	O
the	O
effectiveness	O
and	O
efficiency	O
of	O
communication	O
and	O
data	B-KEY
exchange	I-KEY
in	O
a	O
blackboard	O
system	O
.	O
In	O
a	O
previous	O
investigation	O
,	O
an	O
approach	O
for	O
constructing	O
blackboard	O
systems	O
for	O
product	O
design	O
using	O
a	O
non-directed	O
graph	B-KEY
decomposition	I-KEY
algorithm	O
was	O
proposed	O
.	O
In	O
this	O
paper	O
,	O
the	O
relationship	O
between	O
graph	B-KEY
decomposition	I-KEY
and	O
the	O
resultant	O
blackboard	O
system	O
is	O
further	O
studied	O
.	O
A	O
case	B-KEY
study	I-KEY
of	O
a	O
number	O
of	O
hypothetical	O
blackboard	O
systems	O
that	O
comprise	O
different	O
knowledge	B-KEY
organisations	I-KEY
is	O
provided	O

Flow	B-KEY
measurement	I-KEY
-	O
future	O
directions	O
Interest	O
in	O
the	O
flow	O
of	O
liquids	O
and	O
its	O
measurement	O
can	O
be	O
traced	O
back	O
to	O
early	O
studies	O
by	O
the	O
Egyptians	B-KEY
,	O
the	O
Chinese	B-KEY
and	O
the	O
Romans	B-KEY
.	O
Since	O
these	O
early	O
times	O
the	O
science	O
of	O
flow	B-KEY
measurement	I-KEY
has	O
undergone	O
a	O
massive	O
change	O
but	O
during	O
the	O
last	O
25	O
years	O
or	O
so	O
-LRB-	O
1977-2002	O
-RRB-	O
it	O
has	O
matured	O
enormously	O
.	O
One	O
of	O
the	O
principal	O
reasons	O
for	O
this	O
is	O
that	O
higher	O
accuracies	O
and	O
reliabilities	O
have	O
been	O
demanded	O
by	O
industry	O
in	O
the	O
measurement	O
of	O
fiscal	B-KEY
transfers	I-KEY
and	O
today	O
there	O
is	O
vigorous	O
interest	O
in	O
the	O
subject	O
from	O
both	O
the	O
flowmeter	B-KEY
manufacturer	I-KEY
and	O
user	O
viewpoints	O
.	O
This	O
interest	O
is	O
coupled	O
with	O
the	O
development	O
of	O
advanced	B-KEY
computer	I-KEY
techniques	I-KEY
in	O
fluid	B-KEY
mechanics	I-KEY
together	O
with	O
the	O
application	O
of	O
increasingly	O
sophisticated	O
electronics	O

Computing	O
transient	B-KEY
gating	I-KEY
charge	I-KEY
movement	I-KEY
of	O
voltage-dependent	O
ion	O
channels	O
The	O
opening	O
of	O
voltage-gated	O
sodium	O
,	O
potassium	O
,	O
and	O
calcium	O
ion	B-KEY
channels	I-KEY
has	O
a	O
steep	O
relationship	O
with	O
voltage	O
.	O
In	O
response	O
to	O
changes	O
in	O
the	O
transmembrane	B-KEY
voltage	I-KEY
,	O
structural	O
movements	O
of	O
an	O
ion	B-KEY
channel	I-KEY
that	O
precede	O
channel	O
opening	O
generate	O
a	O
capacitative	O
gating	B-KEY
current	I-KEY
.	O
The	O
net	O
gating	O
charge	O
displacement	O
due	O
to	O
membrane	O
depolarization	O
is	O
an	O
index	O
of	O
the	O
voltage	O
sensitivity	O
of	O
the	O
ion	B-KEY
channel	I-KEY
activation	O
process	O
.	O
Understanding	O
the	O
molecular	O
basis	O
of	O
voltage-dependent	O
gating	O
of	O
ion	B-KEY
channels	I-KEY
requires	O
the	O
measurement	O
and	O
computation	O
of	O
the	O
gating	O
charge	O
,	O
Q	O
.	O
We	O
derive	O
a	O
simple	O
and	O
accurate	O
semianalytic	O
approach	O
to	O
computing	O
the	O
voltage	O
dependence	O
of	O
transient	B-KEY
gating	I-KEY
charge	I-KEY
movement	I-KEY
-LRB-	O
Q-V	O
relationship	O
-RRB-	O
of	O
discrete	O
Markov	O
state	O
models	O
of	O
ion	O
channels	O
using	O
matrix	O
methods	O
.	O
This	O
approach	O
allows	O
rapid	O
computation	O
of	O
Q-V	O
curves	O
for	O
finite	O
and	O
infinite	O
length	O
step	O
depolarizations	O
and	O
is	O
consistent	O
with	O
experimentally	O
measured	O
transient	O
gating	O
charge	O
.	O
This	O
computational	O
approach	O
was	O
applied	O
to	O
Shaker	O
potassium	O
channel	O
gating	O
,	O
including	O
the	O
impact	O
of	O
inactivating	B-KEY
particles	O
on	O
potassium	O
channel	O
gating	B-KEY
currents	I-KEY

Greenberger-Horne-Zeilinger	B-KEY
paradoxes	I-KEY
for	O
many	B-KEY
qubits	I-KEY
We	O
construct	O
Greenberger-Horne-Zeilinger	O
-LRB-	O
GHZ	O
-RRB-	O
contradictions	O
for	O
three	O
or	O
more	O
parties	O
sharing	O
an	O
entangled	B-KEY
state	I-KEY
,	O
the	O
dimension	O
of	O
each	O
subsystem	O
being	O
an	O
even	O
integer	O
d	O
.	O
The	O
simplest	O
example	O
that	O
goes	O
beyond	O
the	O
standard	O
GHZ	B-KEY
paradox	I-KEY
-LRB-	O
three	O
qubits	O
-RRB-	O
involves	O
five	O
ququats	O
-LRB-	O
d	O
=	O
4	O
-RRB-	O
.	O
We	O
then	O
examine	O
the	O
criteria	O
that	O
a	O
GHZ	B-KEY
paradox	I-KEY
must	O
satisfy	O
in	O
order	O
to	O
be	O
genuinely	O
M	O
partite	O
and	O
d	O
dimensional	O

Node-capacitated	B-KEY
ring	I-KEY
routing	I-KEY
We	O
consider	O
the	O
node-capacitated	B-KEY
routing	I-KEY
problem	I-KEY
in	O
an	O
undirected	B-KEY
ring	I-KEY
network	I-KEY
along	O
with	O
its	O
fractional	B-KEY
relaxation	I-KEY
,	O
the	O
node-capacitated	B-KEY
multicommodity	I-KEY
flow	I-KEY
problem	I-KEY
.	O
For	O
the	O
feasibility	B-KEY
problem	I-KEY
,	O
Farkas	O
'	O
lemma	O
provides	O
a	O
characterization	O
for	O
general	O
undirected	B-KEY
graphs	I-KEY
,	O
asserting	O
roughly	O
that	O
there	O
exists	O
such	O
a	O
flow	O
if	O
and	O
only	O
if	O
the	O
so-called	O
distance	B-KEY
inequality	I-KEY
holds	O
for	O
every	O
choice	O
of	O
distance	B-KEY
functions	I-KEY
arising	O
from	O
nonnegative	B-KEY
node	I-KEY
weights	I-KEY
.	O
For	O
rings	O
,	O
this	O
-LRB-	O
straightforward	O
-RRB-	O
result	O
will	O
be	O
improved	O
in	O
two	O
ways	O
.	O
We	O
prove	O
that	O
,	O
independent	O
of	O
the	O
integrality	O
of	O
node	O
capacities	O
,	O
it	O
suffices	O
to	O
require	O
the	O
distance	B-KEY
inequality	I-KEY
only	O
for	O
distances	O
arising	O
from	O
-LRB-	O
0-1-2	O
-RRB-	O
-	O
valued	O
node	O
weights	O
,	O
a	O
requirement	O
that	O
will	O
be	O
called	O
the	O
double-cut	B-KEY
condition	I-KEY
.	O
Moreover	O
,	O
for	O
integer-valued	B-KEY
node	I-KEY
capacities	I-KEY
,	O
the	O
double-cut	B-KEY
condition	I-KEY
implies	O
the	O
existence	O
of	O
a	O
half-integral	B-KEY
multicommodity	I-KEY
flow	I-KEY
.	O
In	O
this	O
case	O
there	O
is	O
even	O
an	O
integer-valued	B-KEY
multicommodity	I-KEY
flow	I-KEY
that	O
violates	O
each	O
node	O
capacity	O
by	O
at	O
most	O
one	O
.	O
Our	O
approach	O
gives	O
rise	O
to	O
a	O
combinatorial	O
,	O
strongly	O
polynomial	O
algorithm	O
to	O
compute	O
either	O
a	O
violating	B-KEY
double-cut	I-KEY
or	O
a	O
node-capacitated	O
multicommodity	O
flow	O
.	O
A	O
relation	O
of	O
the	O
problem	O
to	O
its	O
edge-capacitated	O
counterpart	O
will	O
also	O
be	O
explained	O

Kontiki	B-KEY
.	O
Shortcuts	O
for	O
content	O
's	O
trip	O
to	O
the	O
edge	O
When	O
electronic	B-KEY
files	I-KEY
get	O
zapped	O
from	O
one	O
location	O
to	O
another	O
,	O
you	O
probably	O
are	O
n't	O
thinking	O
about	O
the	O
physical	O
distance	O
they	O
must	O
travel-or	O
how	O
that	O
distance	O
might	O
affect	O
the	O
time	O
it	O
takes	O
to	O
get	O
there	O
.	O
But	O
if	O
you	O
work	O
for	O
CDN	O
company	O
Kontiki	B-KEY
,	O
this	O
is	O
just	O
about	O
all	O
you	O
think	O
about	O
.	O
Championing	O
a	O
P2P-like	O
``	O
bandwidth	O
harvesting	O
''	O
technology	O
,	O
Kontiki	B-KEY
has	O
figured	O
out	O
how	O
to	O
not	O
only	O
quickly	O
distribute	O
content	O
to	O
the	O
``	O
edge	O
''	O
but	O
to	O
utilize	O
a	O
combination	O
of	O
centralized	B-KEY
servers	I-KEY
and	O
a	O
network	O
of	O
enduser	B-KEY
machines	I-KEY
to	O
collect	O
,	O
or	O
``	O
harvest	O
,	O
''	O
underutilized	B-KEY
bandwidth	I-KEY
and	O
make	O
redundant	B-KEY
file	I-KEY
requests	I-KEY
more	O
efficient	O

Numerical	B-KEY
simulation	I-KEY
of	O
information	B-KEY
recovery	I-KEY
in	O
quantum	B-KEY
computers	I-KEY
Decoherence	O
is	O
the	O
main	O
problem	O
to	O
be	O
solved	O
before	O
quantum	B-KEY
computers	I-KEY
can	O
be	O
built	O
.	O
To	O
control	O
decoherence	O
,	O
it	O
is	O
possible	O
to	O
use	O
error	B-KEY
correction	I-KEY
methods	I-KEY
,	O
but	O
these	O
methods	O
are	O
themselves	O
noisy	O
quantum	B-KEY
computation	I-KEY
processes	O
.	O
In	O
this	O
work	O
,	O
we	O
study	O
the	O
ability	O
of	O
Steane	O
's	O
and	O
Shor	O
's	O
fault-tolerant	B-KEY
recovering	I-KEY
methods	I-KEY
,	O
as	O
well	O
as	O
a	O
modification	O
of	O
Steane	O
's	O
ancilla	B-KEY
network	I-KEY
,	O
to	O
correct	O
errors	O
in	O
qubits	B-KEY
.	O
We	O
test	O
a	O
way	O
to	O
measure	O
correctly	O
ancilla	O
's	O
fidelity	O
for	O
these	O
methods	O
,	O
and	O
state	O
the	O
possibility	O
of	O
carrying	O
out	O
an	O
effective	O
error	O
correction	O
through	O
a	O
noisy	B-KEY
quantum	I-KEY
channel	I-KEY
,	O
even	O
using	O
noisy	B-KEY
error	I-KEY
correction	I-KEY
methods	I-KEY

Implementation	O
of	O
universal	B-KEY
quantum	I-KEY
gates	I-KEY
based	O
on	O
nonadiabatic	O
geometric	O
phases	O
We	O
propose	O
an	O
experimentally	O
feasible	O
scheme	O
to	O
achieve	O
quantum	B-KEY
computation	I-KEY
based	O
on	O
nonadiabatic	B-KEY
geometric	I-KEY
phase	I-KEY
shifts	I-KEY
,	O
in	O
which	O
a	O
cyclic	B-KEY
geometric	I-KEY
phase	I-KEY
is	O
used	O
to	O
realize	O
a	O
set	O
of	O
universal	B-KEY
quantum	I-KEY
gates	I-KEY
.	O
Physical	O
implementation	O
of	O
this	O
set	O
of	O
gates	O
is	O
designed	O
for	O
Josephson	B-KEY
junctions	I-KEY
and	O
for	O
NMR	B-KEY
systems	I-KEY
.	O
Interestingly	O
,	O
we	O
find	O
that	O
the	O
nonadiabatic	B-KEY
phase	I-KEY
shift	I-KEY
may	O
be	O
independent	O
of	O
the	O
operation	B-KEY
time	I-KEY
under	O
appropriate	O
controllable	O
conditions	O
.	O
A	O
remarkable	O
feature	O
of	O
the	O
present	O
nonadiabatic	B-KEY
geometric	I-KEY
gates	I-KEY
is	O
that	O
there	O
is	O
no	O
intrinsic	O
limitation	O
on	O
the	O
operation	B-KEY
time	I-KEY

The	O
development	O
and	O
evaluation	O
of	O
SHOKE2000	B-KEY
:	O
the	O
PCI-based	O
FPGA	O
card	O
This	O
paper	O
describes	O
a	O
PCI-based	O
FPGA	O
card	O
,	O
SHOKE2000	O
,	O
which	O
was	O
developed	O
in	O
order	O
to	O
study	O
reconfigurable	O
computing	O
.	O
Since	O
the	O
latest	O
field	B-KEY
programmable	I-KEY
gate	I-KEY
arrays	I-KEY
-LRB-	O
FPGA	B-KEY
-RRB-	O
consist	O
of	O
input/output	O
-LRB-	O
I/O	O
-RRB-	O
configurable	O
blocks	O
as	O
well	O
as	O
internal	O
configurable	O
logic	O
blocks	O
,	O
they	O
not	O
only	O
realize	O
various	O
user	B-KEY
logic	I-KEY
circuits	I-KEY
but	O
also	O
connect	O
with	O
popular	O
I/O	B-KEY
standards	I-KEY
easily	O
.	O
These	O
features	O
enable	O
FPGA	B-KEY
to	O
connect	O
several	O
devices	O
with	O
different	O
interfaces	B-KEY
,	O
and	O
thus	O
new	O
reconfigurable	O
systems	O
would	O
be	O
realizable	O
by	O
connecting	O
the	O
FPGA	B-KEY
with	O
devices	O
such	O
as	O
digital	B-KEY
signal	I-KEY
processors	I-KEY
-LRB-	O
DSP	B-KEY
-RRB-	O
and	O
analog	B-KEY
devices	I-KEY
.	O
This	O
paper	O
describes	O
the	O
basic	O
functions	O
of	O
SHOKE2000	B-KEY
,	O
which	O
was	O
developed	O
for	O
realizing	O
hybrid	B-KEY
reconfigurable	I-KEY
systems	I-KEY
consisting	O
of	O
FPGA	B-KEY
,	O
DSP	B-KEY
,	O
and	O
analog	B-KEY
devices	I-KEY
.	O
We	O
also	O
present	O
application	O
examples	O
of	O
SHOKE2000	B-KEY
,	O
including	O
a	O
simple	O
image	B-KEY
recognition	I-KEY
application	I-KEY
,	O
a	O
distributed	B-KEY
shared	I-KEY
memory	I-KEY
computer	I-KEY
cluster	I-KEY
,	O
and	O
teaching	B-KEY
materials	I-KEY
for	O
computer	B-KEY
education	I-KEY

A	O
building	B-KEY
block	I-KEY
approach	I-KEY
to	O
automated	O
engineering	O
Shenandoah	B-KEY
Valley	I-KEY
Electric	I-KEY
Cooperative	I-KEY
-LRB-	O
SVEC	O
,	O
Mt.	O
Crawford	O
,	O
Virginia	O
,	O
US	O
-RRB-	O
recognized	O
the	O
need	O
to	O
automate	O
engineering	O
functions	O
and	O
create	O
an	O
interactive	O
model	B-KEY
of	I-KEY
its	O
distribution	O
system	B-KEY
in	I-KEY
the	O
early	O
1990s	O
.	O
It	O
had	O
used	O
Milsoft	O
's	O
DA	O
software	O
for	O
more	O
than	O
10	O
years	O
to	O
make	O
engineering	O
studies	O
,	O
and	O
had	O
a	O
Landis	O
and	O
Gyr	O
SCADA	O
system	O
and	O
a	O
hybrid	O
load	O
management	O
system	O
for	O
controlling	O
water	O
heater	O
switches	O
.	O
With	O
the	O
development	O
of	O
GIS	B-KEY
and	O
facilities	O
management	O
-LRB-	O
FM	O
-RRB-	O
applications	O
,	O
SVEC	O
decided	O
this	O
should	O
be	O
the	O
basis	O
for	O
an	O
information	O
system	O
that	O
would	O
model	O
its	O
physical	O
plant	O
and	O
interface	O
with	O
its	O
accounting	O
and	O
billing	B-KEY
systems	I-KEY
.	O
It	O
could	O
add	O
applications	O
such	O
as	O
outage	B-KEY
management	I-KEY
,	O
staking	B-KEY
,	O
line	B-KEY
design	I-KEY
and	O
metering	B-KEY
to	O
use	O
this	O
information	O
and	O
interface	O
with	O
these	O
databases	B-KEY
.	O
However	O
,	O
based	O
on	O
SVEC	O
's	O
size	O
it	O
was	O
not	O
feasible	O
to	O
implement	O
a	O
sophisticated	O
and	O
expensive	O
GIS/FM	O
system	O
.	O
Over	O
the	O
past	O
nine	O
years	O
,	O
SVEC	O
has	O
had	O
success	O
with	O
a	O
building	B-KEY
block	I-KEY
approach	I-KEY
,	O
and	O
its	O
customers	O
and	O
employees	O
are	O
realizing	O
the	O
benefits	O
of	O
the	O
automated	O
applications	O
.	O
This	O
building	B-KEY
block	I-KEY
approach	I-KEY
is	O
discussed	O
in	O
this	O
article	O
including	O
the	O
GIS	B-KEY
,	O
outage	B-KEY
management	I-KEY
system	O
,	O
MapViewer	B-KEY
and	O
a	O
staking	B-KEY
package	O
.	O
The	O
lessons	O
learned	O
and	O
future	O
expansion	O
are	O
discussed	O

Second	O
term	O
-LSB-	O
International	B-KEY
Telecommunication	I-KEY
Union	I-KEY
-RSB-	O
Later	O
this	O
month	O
Yoshio	O
Utsumi	O
is	O
expected	O
to	O
be	O
re-elected	O
for	O
a	O
second	O
four	O
year	O
term	O
as	O
secretary	O
general	O
of	O
the	O
International	B-KEY
Telecommunication	I-KEY
Union	I-KEY
.	O
Here	O
he	O
talks	O
to	O
Matthew	O
May	O
about	O
getting	O
involved	O
in	O
internet	B-KEY
addressing	I-KEY
,	O
the	O
prospects	O
for	O
3g	B-KEY
,	O
the	O
need	O
for	O
further	O
reform	O
of	O
his	O
organisation	O
...	O
and	O
the	O
translating	B-KEY
telephone	I-KEY

Dementing	B-KEY
disorders	I-KEY
:	O
volumetric	O
measurement	O
of	O
cerebrospinal	O
fluid	O
to	O
distinguish	O
normal	O
from	O
pathologic	O
finding	O
-	O
feasibility	O
study	O
We	O
have	O
demonstrated	O
that	O
automated	B-KEY
methods	I-KEY
to	O
describe	O
the	O
severity	O
and	O
distribution	O
of	O
cerebral	O
atrophy	O
are	O
capable	O
of	O
providing	O
diagnostic	B-KEY
information	I-KEY
in	O
the	O
classification	O
of	O
neurodegenerative	O
diseases	O

Fast	O
accurate	O
MEG	B-KEY
source	I-KEY
localization	I-KEY
using	O
a	O
multilayer	B-KEY
perceptron	I-KEY
trained	O
with	O
real	B-KEY
brain	I-KEY
noise	I-KEY
Iterative	B-KEY
gradient	I-KEY
methods	I-KEY
such	O
as	O
Levenberg-Marquardt	O
-LRB-	O
LM	O
-RRB-	O
are	O
in	O
widespread	O
use	O
for	O
source	O
localization	O
from	O
electroencephalographic	O
-LRB-	O
EEG	O
-RRB-	O
and	O
magnetoencephalographic	O
-LRB-	O
MEG	O
-RRB-	O
signals	O
.	O
Unfortunately	O
,	O
LM	O
depends	O
sensitively	O
on	O
the	O
initial	O
guess	O
,	O
necessitating	O
repeated	O
runs	O
.	O
This	O
,	O
combined	O
with	O
LM	O
's	O
high	O
per-step	O
cost	O
,	O
makes	O
its	O
computational	O
burden	O
quite	O
high	O
.	O
To	O
reduce	O
this	O
burden	O
,	O
we	O
trained	O
a	O
multilayer	B-KEY
perceptron	I-KEY
-LRB-	O
MLP	O
-RRB-	O
as	O
a	O
realtime	O
localizer	O
.	O
We	O
used	O
an	O
analytical	B-KEY
model	I-KEY
of	O
quasistatic	B-KEY
electromagnetic	I-KEY
propagation	I-KEY
through	O
a	O
spherical	B-KEY
head	I-KEY
to	O
map	O
randomly	O
chosen	O
dipoles	O
to	O
sensor	O
activities	O
according	O
to	O
the	O
sensor	O
geometry	O
of	O
a	O
4D	O
Neuroimaging	O
Neuromag-122	O
MEG	O
system	O
,	O
and	O
trained	O
a	O
MLP	O
to	O
invert	O
this	O
mapping	O
in	O
the	O
absence	O
of	O
noise	O
or	O
in	O
the	O
presence	O
of	O
various	O
sorts	O
of	O
noise	O
such	O
as	O
white	B-KEY
Gaussian	I-KEY
noise	I-KEY
,	O
correlated	B-KEY
noise	I-KEY
,	O
or	O
real	B-KEY
brain	I-KEY
noise	I-KEY
.	O
A	O
MLP	O
structure	O
was	O
chosen	O
to	O
trade	O
off	O
computation	O
and	O
accuracy	O
.	O
This	O
MLP	O
was	O
trained	O
four	O
times	O
,	O
with	O
each	O
type	O
of	O
noise	O
.	O
We	O
measured	O
the	O
effects	O
of	O
initial	O
guesses	O
on	O
LM	O
performance	O
,	O
which	O
motivated	O
a	O
hybrid	O
MLP-start-LM	O
method	O
,	O
in	O
which	O
the	O
trained	O
MLP	O
initializes	O
LM	O
.	O
We	O
also	O
compared	O
the	O
localization	O
performance	O
of	O
LM	O
,	O
MLPs	O
,	O
and	O
hybrid	O
MLP-start-LMs	O
for	O
realistic	O
brain	O
signals	O
.	O
Trained	O
MLPs	O
are	O
much	O
faster	O
than	O
other	O
methods	O
,	O
while	O
the	O
hybrid	O
MLP-start-LMs	O
are	O
faster	O
and	O
more	O
accurate	O
than	O
fixed-4-start-LM	O
.	O
In	O
particular	O
,	O
the	O
hybrid	O
MLP-start-LM	O
initialized	O
by	O
a	O
MLP	O
trained	O
with	O
the	O
real	B-KEY
brain	I-KEY
noise	I-KEY
dataset	O
is	O
60	O
times	O
faster	O
and	O
is	O
comparable	O
in	O
accuracy	O
to	O
random-20-start-LM	O
,	O
and	O
this	O
hybrid	O
system	O
-LRB-	O
localization	O
error	O
:	O
0.28	O
cm	O
,	O
computation	O
time	O
:	O
36	O
ms	O
-RRB-	O
shows	O
almost	O
as	O
good	O
performance	O
as	O
optimal-1-start-LM	O
-LRB-	O
localization	O
error	O
:	O
0.23	O
cm	O
,	O
computation	O
time	O
:	O
22	O
ms	O
-RRB-	O
,	O
which	O
initializes	O
LM	O
with	O
the	O
correct	O
dipole	O
location	O
.	O
MLPs	O
trained	O
with	O
noise	O
perform	O
better	O
than	O
the	O
MLP	O
trained	O
without	O
noise	O
,	O
and	O
the	O
MLP	O
trained	O
with	O
real	B-KEY
brain	I-KEY
noise	I-KEY
is	O
almost	O
as	O
good	O
an	O
initial	O
guesser	O
for	O
LM	O
as	O
the	O
correct	O
dipole	O
location	O

Molecular	B-KEY
descriptor	I-KEY
selection	I-KEY
combining	O
genetic	B-KEY
algorithms	I-KEY
and	O
fuzzy	B-KEY
logic	I-KEY
:	O
application	O
to	O
database	B-KEY
mining	I-KEY
procedures	O
A	O
new	O
algorithm	O
,	O
devoted	O
to	O
molecular	B-KEY
descriptor	I-KEY
selection	I-KEY
in	O
the	O
context	O
of	O
data	B-KEY
mining	I-KEY
problems	O
,	O
has	O
been	O
developed	O
.	O
This	O
algorithm	O
is	O
based	O
on	O
the	O
concepts	O
of	O
genetic	B-KEY
algorithms	I-KEY
-LRB-	O
GA	O
-RRB-	O
for	O
descriptor	B-KEY
hyperspace	I-KEY
exploration	I-KEY
and	O
combined	O
with	O
a	O
stepwise	B-KEY
approach	I-KEY
to	O
get	O
local	B-KEY
convergence	I-KEY
.	O
Its	O
selection	O
power	O
was	O
evaluated	O
by	O
a	O
fitness	B-KEY
function	I-KEY
derived	O
from	O
a	O
fuzzy	B-KEY
clustering	I-KEY
method	I-KEY
.	O
Different	O
training	O
and	O
test	B-KEY
sets	I-KEY
were	O
randomly	O
generated	O
at	O
each	O
GA	O
generation	O
.	O
The	O
fitness	B-KEY
score	I-KEY
was	O
derived	O
by	O
combining	O
the	O
scores	O
of	O
the	O
training	O
and	O
test	B-KEY
sets	I-KEY
.	O
The	O
ability	O
of	O
the	O
proposed	O
algorithm	O
to	O
select	O
relevant	O
subsets	O
of	O
descriptors	O
was	O
tested	O
on	O
two	O
data	O
sets	O
.	O
The	O
first	O
one	O
,	O
an	O
academic	O
example	O
,	O
corresponded	O
to	O
the	O
artificial	O
problem	O
of	O
Bullseye	B-KEY
,	O
the	O
second	O
was	O
a	O
real	O
data	O
set	O
including	O
114	O
olfactory	B-KEY
compounds	I-KEY
divided	O
into	O
three	O
odor	B-KEY
categories	I-KEY
.	O
In	O
both	O
cases	O
,	O
the	O
proposed	O
method	O
allowed	O
to	O
improve	O
the	O
separation	O
between	O
the	O
different	O
data	O
set	O
classes	O

Finding	O
performance	O
bugs	O
with	O
the	O
TNO	O
HPF	O
benchmark	B-KEY
suite	I-KEY
High-Performance	B-KEY
Fortran	I-KEY
-LRB-	O
HPF	O
-RRB-	O
has	O
been	O
designed	O
to	O
provide	O
portable	B-KEY
performance	I-KEY
on	O
distributed	B-KEY
memory	I-KEY
machines	I-KEY
.	O
An	O
important	O
aspect	O
of	O
portable	B-KEY
performance	I-KEY
is	O
the	O
behavior	O
of	O
the	O
available	O
HPF	B-KEY
compilers	I-KEY
.	O
Ideally	O
,	O
a	O
programmer	O
may	O
expect	O
comparable	O
performance	O
between	O
different	O
HPF	B-KEY
compilers	I-KEY
,	O
given	O
the	O
same	O
program	O
and	O
the	O
same	O
machine	O
.	O
To	O
test	O
the	O
performance	B-KEY
portability	I-KEY
between	O
compilers	O
,	O
we	O
have	O
designed	O
a	O
special	O
benchmark	B-KEY
suite	I-KEY
,	O
called	O
the	O
TNO	O
HPF	O
benchmark	B-KEY
suite	I-KEY
.	O
It	O
consists	O
of	O
a	O
set	O
of	O
HPF	O
programs	O
that	O
test	O
various	O
aspects	O
of	O
efficient	O
parallel	O
code	O
generation	O
.	O
The	O
benchmark	B-KEY
suite	I-KEY
consists	O
of	O
a	O
number	O
of	O
template	O
programs	O
that	O
are	O
used	O
to	O
generate	O
test	O
programs	O
with	O
different	O
array	O
sizes	O
,	O
alignments	O
,	O
distributions	O
,	O
and	O
iteration	O
spaces	O
.	O
It	O
ranges	O
from	O
very	O
simple	O
assignments	O
to	O
more	O
complex	O
assignments	O
such	O
as	O
triangular	O
iteration	O
spaces	O
,	O
convex	O
iteration	O
spaces	O
,	O
coupled	O
subscripts	O
,	O
and	O
indirection	O
arrays	O
.	O
We	O
have	O
run	O
the	O
TNO	O
HPF	O
benchmark	B-KEY
suite	I-KEY
on	O
three	O
compilers	O
:	O
the	O
PREPARE	O
prototype	O
compiler	O
,	O
the	O
PGI-HPF	O
compiler	O
,	O
and	O
the	O
GMD	O
Adaptor	O
HPF	O
compiler	O
.	O
Results	O
show	O
performance	O
differences	O
that	O
can	O
be	O
quite	O
large	O
-LRB-	O
up	O
to	O
two	O
orders	O
of	O
magnitude	O
for	O
the	O
same	O
test	O
program	O
-RRB-	O
.	O
Closer	O
inspection	O
reveals	O
that	O
the	O
origin	O
of	O
most	O
of	O
the	O
differences	O
in	O
performance	O
is	O
due	O
to	O
differences	O
in	O
local	O
enumeration	O
and	O
storage	O
of	O
distributed	O
array	O
elements	O

The	O
quadratic	B-KEY
0-1	I-KEY
knapsack	I-KEY
problem	I-KEY
with	O
series-parallel	B-KEY
support	I-KEY
We	O
consider	O
various	O
special	O
cases	O
of	O
the	O
quadratic	B-KEY
0-1	I-KEY
knapsack	I-KEY
problem	I-KEY
-LRB-	O
QKP	O
-RRB-	O
for	O
which	O
the	O
underlying	B-KEY
graph	I-KEY
structure	I-KEY
is	O
fairly	O
simple	O
.	O
For	O
the	O
variant	O
with	O
edge	O
series-parallel	O
graphs	O
,	O
we	O
give	O
a	O
dynamic	B-KEY
programming	I-KEY
algorithm	I-KEY
with	O
pseudo-polynomial	B-KEY
time	I-KEY
complexity	I-KEY
,	O
and	O
a	O
fully	B-KEY
polynomial	I-KEY
time	I-KEY
approximation	I-KEY
scheme	I-KEY
.	O
In	O
strong	O
contrast	O
to	O
this	O
,	O
the	O
variant	O
with	O
vertex	O
series-parallel	O
graphs	O
is	O
shown	O
to	O
be	O
strongly	O
NP-complete	O

Integrated	B-KEY
support	I-KEY
based	O
on	O
task	B-KEY
models	I-KEY
for	O
the	O
design	O
,	O
evaluation	O
,	O
and	O
documentation	O
of	O
interactive	B-KEY
safety-critical	I-KEY
systems	I-KEY
:	O
a	O
case	O
study	O
in	O
the	O
air-traffic	B-KEY
control	I-KEY
domain	I-KEY
This	O
paper	O
presents	O
an	O
approach	O
to	O
using	O
task	B-KEY
models	I-KEY
in	O
both	O
the	O
design	O
and	O
the	O
evaluation	O
phases	O
of	O
interactive	O
safety-critical	O
applications	O
.	O
We	O
explain	O
how	O
it	O
is	O
possible	O
to	O
use	O
information	O
contained	O
in	O
task	B-KEY
models	I-KEY
to	O
support	O
the	O
design	O
and	O
development	O
of	O
effective	O
user	B-KEY
interfaces	I-KEY
.	O
Moreover	O
,	O
we	O
show	O
how	O
task	B-KEY
models	I-KEY
can	O
also	O
support	O
a	O
systematic	O
inspection-based	B-KEY
usability	I-KEY
assessment	I-KEY
by	O
examining	O
possible	O
deviations	O
that	O
can	O
occur	O
while	O
users	O
interact	O
with	O
the	O
system	O
,	O
an	O
important	O
issue	O
especially	O
when	O
coping	O
with	O
the	O
peculiar	O
requirements	O
of	O
safety-critical	O
applications	O
.	O
Such	O
evaluation	O
provides	O
useful	O
technical	B-KEY
documentation	I-KEY
to	O
help	O
users	O
achieve	O
an	O
in-depth	O
understanding	O
of	O
the	O
system	O
and	O
its	O
design	O
rationale	O
.	O
Lastly	O
,	O
a	O
description	O
of	O
the	O
application	O
of	O
our	O
approach	O
to	O
a	O
real	O
case	O
study	O
in	O
the	O
air-traffic	B-KEY
control	I-KEY
domain	I-KEY
will	O
illustrate	O
the	O
main	O
features	O
of	O
the	O
proposed	O
method	O
.	O
In	O
particular	O
,	O
we	O
discuss	O
examples	O
taken	O
from	O
an	O
application	O
for	O
air-traffic	O
controllers	O
in	O
an	O
aerodrome	O
supported	O
by	O
graphical	B-KEY
user	I-KEY
interfaces	I-KEY
for	O
data-link	O
communications	O
with	O
pilots	O

Copyright	O
management	O
in	O
the	O
digital	B-KEY
age	I-KEY
Listening	O
to	O
and	O
buying	O
music	O
online	O
is	O
becoming	O
increasingly	O
popular	O
with	O
consumers	O
.	O
So	O
much	O
so	O
that	O
Merrill	O
Lynch	O
forecasts	O
the	O
value	O
of	O
the	O
online	O
music	O
market	O
will	O
explode	O
from	O
$	O
8	O
million	O
in	O
2001	O
to	O
$	O
1,409	O
million	O
in	O
2005	O
.	O
But	O
online	O
delivery	O
is	O
not	O
without	O
problems	O
;	O
the	O
issue	O
of	O
copyright	O
management	O
in	O
particular	O
has	O
become	O
a	O
serious	O
thorn	O
in	O
the	O
side	O
for	O
digital	B-KEY
content	I-KEY
creators	I-KEY
.	O
Martin	O
Brass	O
,	O
ex	O
-	O
music	O
producer	O
and	O
senior	O
industry	O
consultant	O
at	O
Syntegra	O
,	O
explains	O

Mustering	O
motivation	B-KEY
to	O
enact	O
decisions	O
:	O
how	O
decision	B-KEY
process	I-KEY
characteristics	I-KEY
influence	O
goal	B-KEY
realization	I-KEY
Decision	B-KEY
scientists	I-KEY
tend	O
to	O
focus	O
mainly	O
on	O
decision	O
antecedents	O
,	O
studying	O
how	O
people	O
make	O
decisions	O
.	O
Action	B-KEY
psychologists	I-KEY
,	O
in	O
contrast	O
,	O
study	O
post-decision	B-KEY
issues	I-KEY
,	O
investigating	O
how	O
decisions	O
,	O
once	O
formed	O
,	O
are	O
maintained	O
,	O
protected	O
,	O
and	O
enacted	O
.	O
Through	O
the	O
research	O
presented	O
here	O
,	O
we	O
seek	O
to	O
bridge	O
these	O
two	O
disciplines	O
,	O
proposing	O
that	O
the	O
process	O
by	O
which	O
decisions	O
are	O
reached	O
motivates	B-KEY
subsequent	O
pursuit	O
and	O
benefits	O
eventual	O
realization	O
.	O
We	O
identify	O
three	O
characteristics	O
of	O
the	O
decision	O
process	O
-LRB-	O
DP	O
-RRB-	O
as	O
having	O
motivation-mustering	B-KEY
potential	I-KEY
:	O
DP	O
effort	O
investment	O
,	O
DP	O
importance	O
,	O
and	O
DP	O
confidence	O
.	O
Through	O
two	O
field	O
studies	O
tracking	O
participants	O
'	O
decision	O
processes	O
,	O
pursuit	O
and	O
realization	O
,	O
we	O
find	O
that	O
after	O
controlling	O
for	O
the	O
influence	O
of	O
the	O
motivational	B-KEY
mechanisms	O
of	O
goal	B-KEY
intention	I-KEY
and	O
implementation	O
intention	O
,	O
the	O
three	O
decision	B-KEY
process	I-KEY
characteristics	I-KEY
significantly	O
influence	O
the	O
successful	O
enactment	O
of	O
the	O
chosen	O
decision	O
directly	O
.	O
The	O
theoretical	O
and	O
practical	O
implications	O
of	O
these	O
findings	O
are	O
considered	O
and	O
future	O
research	B-KEY
opportunities	I-KEY
are	O
identified	O

North	B-KEY
American	I-KEY
carrier	I-KEY
survey	I-KEY
:	O
simply	O
the	O
best	O
Network	O
Magazine	O
carried	O
out	O
a	O
North	B-KEY
American	I-KEY
carrier	I-KEY
survey	I-KEY
.	O
Thousands	O
of	O
network	O
engineers	O
gave	O
information	O
on	O
providers	O
'	O
strengths	O
and	O
weaknesses	O
across	O
seven	O
services	O
:	O
private	B-KEY
lines	I-KEY
,	O
frame	B-KEY
relay	I-KEY
,	O
ATM	B-KEY
,	O
VPNs	B-KEY
,	O
dedicated	B-KEY
Internet	I-KEY
access	I-KEY
,	O
Ethernet	B-KEY
services	I-KEY
,	O
and	O
Web	B-KEY
hosting	I-KEY
.	O
Respondents	O
also	O
ranked	O
providers	O
on	O
their	O
ability	O
to	O
perform	O
in	O
up	O
to	O
eight	O
categories	O
including	O
customer	B-KEY
service	I-KEY
,	O
reliability	B-KEY
,	O
and	O
price	B-KEY
.	O
Users	O
rated	O
more	O
than	O
a	O
dozen	O
providers	O
for	O
each	O
survey	O
.	O
Carriers	O
needed	O
to	O
receive	O
at	O
least	O
30	O
votes	O
for	O
inclusion	O
in	O
the	O
survey	O
.	O
Readers	O
were	O
asked	O
to	O
rate	O
carriers	O
on	O
up	O
to	O
nine	O
categories	O
using	O
a	O
scale	O
of	O
1	O
-LRB-	O
unacceptable	O
-RRB-	O
to	O
5	O
-LRB-	O
excellent	O
-RRB-	O
.	O
Not	O
all	O
categories	O
are	O
equally	O
important	O
.	O
To	O
try	O
and	O
get	O
at	O
these	O
differences	O
,	O
Network	O
Magazine	O
asked	O
readers	O
to	O
assign	O
a	O
weight	O
to	O
each	O
category	O
.	O
The	O
big	O
winners	O
were	O
VPNs	B-KEY

Phase	B-KEY
conditions	I-KEY
for	O
Schur	B-KEY
polynomials	I-KEY
The	O
rate	B-KEY
of	I-KEY
change	I-KEY
of	I-KEY
phase	I-KEY
of	O
a	O
real	O
or	O
complex	O
Schur	B-KEY
polynomial	I-KEY
,	O
evaluated	O
along	O
the	O
unit	O
circle	O
traversed	O
counterclockwise	O
,	O
is	O
strictly	O
positive	O
.	O
For	O
polynomials	O
with	O
real	B-KEY
coefficients	I-KEY
,	O
this	O
bound	O
can	O
be	O
tightened	O
.	O
These	O
and	O
some	O
other	O
fundamental	O
bounds	O
on	O
the	O
rate	B-KEY
of	I-KEY
change	I-KEY
of	I-KEY
phase	I-KEY
are	O
derived	O
here	O
,	O
using	O
the	O
Tchebyshev	B-KEY
representation	I-KEY
of	O
the	O
image	O
of	O
a	O
real	O
polynomial	O
evaluated	O
on	O
the	O
unit	O
circle	O

Steinmetz	B-KEY
system	I-KEY
design	I-KEY
under	O
unbalanced	O
conditions	O
This	O
paper	O
studies	O
and	O
develops	O
general	O
analytical	O
expressions	O
to	O
obtain	O
three-phase	B-KEY
current	I-KEY
symmetrization	I-KEY
under	O
unbalanced	B-KEY
voltage	I-KEY
conditions	I-KEY
.	O
It	O
proposes	O
two	O
procedures	O
for	O
this	O
symmetrization	O
:	O
the	O
application	O
of	O
the	O
traditional	O
expressions	O
assuming	O
symmetry	O
conditions	O
and	O
the	O
use	O
of	O
optimization	B-KEY
methods	I-KEY
based	O
on	O
the	O
general	B-KEY
analytical	I-KEY
equations	I-KEY
.	O
Specifically	O
,	O
the	O
paper	O
applies	O
and	O
evaluates	O
these	O
methods	O
to	O
analyze	O
the	O
Steinmetz	B-KEY
system	I-KEY
design	I-KEY
.	O
Several	O
graphics	O
evaluating	O
the	O
error	O
introduced	O
by	O
assumption	O
of	O
balanced	O
voltage	O
in	O
the	O
design	O
are	O
plotted	O
and	O
an	O
example	O
is	O
studied	O
to	O
compare	O
both	O
procedures	O
.	O
In	O
the	O
example	O
the	O
necessity	O
to	O
apply	O
the	O
optimization	O
techniques	O
in	O
highly	O
unbalanced	O
conditions	O
is	O
demonstrated	O

Convergence	O
of	O
Runge-Kutta	O
methods	O
for	O
nonlinear	O
parabolic	O
equations	O
We	O
study	O
time	B-KEY
discretizations	I-KEY
of	O
fully	O
nonlinear	B-KEY
parabolic	I-KEY
differential	I-KEY
equations	I-KEY
.	O
Our	O
analysis	O
uses	O
the	O
fact	O
that	O
the	O
linearization	B-KEY
along	O
the	O
exact	O
solution	O
is	O
a	O
uniformly	B-KEY
sectorial	I-KEY
operator	I-KEY
.	O
We	O
derive	O
smooth	O
and	O
nonsmooth-data	O
error	O
estimates	O
for	O
the	O
backward	O
Euler	O
method	O
,	O
and	O
we	O
prove	O
convergence	O
for	O
strongly	O
A	O
-LRB-	O
v	O
-RRB-	O
-	O
stable	O
Runge-Kutta	O
methods	O
.	O
For	O
the	O
latter	O
,	O
the	O
order	O
of	O
convergence	O
for	O
smooth	O
solutions	O
is	O
essentially	O
determined	O
by	O
the	O
stage	O
order	O
of	O
the	O
method	O
.	O
Numerical	O
examples	O
illustrating	O
the	O
convergence	O
estimates	O
are	O
presented	O

A	O
study	O
of	O
computer	B-KEY
attitudes	I-KEY
of	O
non-computing	O
students	O
of	O
technical	B-KEY
colleges	I-KEY
in	O
Brunei	O
Darussalam	O
The	O
study	O
surveyed	B-KEY
268	O
non-computing	O
students	O
among	O
three	O
technical	B-KEY
colleges	I-KEY
in	O
Brunei	O
Darussalam	O
.	O
The	O
study	O
validated	O
an	O
existing	O
instrument	O
to	O
measure	O
computer	B-KEY
attitudes	I-KEY
of	O
non-computing	O
students	O
,	O
and	O
identified	O
factors	O
that	O
contributed	O
to	O
the	O
formation	O
of	O
their	O
attitudes	O
.	O
The	O
findings	O
show	O
that	O
computer	B-KEY
experience	I-KEY
and	O
educational	B-KEY
qualification	I-KEY
are	O
associated	O
with	O
students	O
'	O
computer	B-KEY
attitudes	I-KEY
.	O
In	O
contrast	O
,	O
variables	O
such	O
as	O
gender	B-KEY
,	O
age	B-KEY
,	O
ownership	O
of	O
a	O
personal	O
computer	O
-LRB-	O
PC	O
-RRB-	O
,	O
geographical	O
location	O
of	O
institution	O
,	O
and	O
prior	O
computer	B-KEY
training	I-KEY
appeared	O
to	O
have	O
no	O
impact	O
on	O
computer	B-KEY
attitudes	I-KEY

Formalising	O
optimal	B-KEY
feature	I-KEY
weight	I-KEY
setting	I-KEY
in	O
case	B-KEY
based	I-KEY
diagnosis	I-KEY
as	O
linear	B-KEY
programming	I-KEY
problems	O
Many	O
approaches	O
to	O
case	B-KEY
based	I-KEY
reasoning	I-KEY
-LRB-	O
CBR	O
-RRB-	O
exploit	O
feature	O
weight	O
setting	O
algorithms	O
to	O
reduce	O
the	O
sensitivity	O
to	O
distance	B-KEY
functions	I-KEY
.	O
We	O
demonstrate	O
that	O
optimal	B-KEY
feature	I-KEY
weight	I-KEY
setting	I-KEY
in	O
a	O
special	O
kind	O
of	O
CBR	O
problems	O
can	O
be	O
formalised	O
as	O
linear	B-KEY
programming	I-KEY
problems	O
.	O
Therefore	O
,	O
the	O
optimal	O
weight	O
settings	O
can	O
be	O
calculated	O
in	O
polynomial	B-KEY
time	I-KEY
instead	O
of	O
searching	B-KEY
in	O
exponential	B-KEY
weight	I-KEY
space	I-KEY
using	O
heuristics	B-KEY
to	O
get	O
sub-optimal	O
settings	O
.	O
We	O
also	O
demonstrate	O
that	O
our	O
approach	O
can	O
be	O
used	O
to	O
solve	O
classification	B-KEY
problems	O

The	O
case	O
for	O
activity	B-KEY
based	I-KEY
management	I-KEY
In	O
today	O
's	O
stormy	O
economic	O
climate	O
businesses	O
need	O
Activity	B-KEY
Based	I-KEY
Management	I-KEY
-LRB-	O
ABM	O
-RRB-	O
more	O
than	O
ever	O
before	O
.	O
In	O
an	O
economic	O
downturn	O
it	O
is	O
a	O
vital	O
tool	O
for	O
pinpointing	O
a	O
business	O
'	O
most	O
profitable	O
customers	O
,	O
products	O
,	O
regions	O
or	O
channels	O
,	O
as	O
well	O
as	O
uncovering	O
the	O
costs	O
of	O
individual	O
business	O
processes	O
that	O
may	O
need	O
to	O
be	O
improved	O
in	O
order	O
to	O
drive	O
higher	O
profit	O
levels	O
.	O
Changes	O
may	O
be	O
afoot	O
in	O
the	O
ABM	O
market	O
,	O
but	O
Armstrong	B-KEY
Laing	I-KEY
Group	I-KEY
CEO	O
Mike	O
Sherratt	O
argues	O
that	O
businesses	O
need	O
specialists	O
with	O
an	O
ABM	O
focus	O
to	O
keep	O
up	O
with	O
their	O
requirements	O
in	O
such	O
a	O
climate	O
.	O
He	O
looks	O
at	O
what	O
benefits	O
a	O
`	O
best-of-breed	O
'	O
ABM	O
system	O
can	O
offer	O
businesses	O
and	O
contends	O
that	O
businesses	O
must	O
choose	O
carefully	O
when	O
going	O
down	O
the	O
ABM	O
route	O
-	O
and	O
also	O
ask	O
themselves	O
the	O
question	O
whether	O
`	O
generalist	O
'	O
organisations	O
will	O
be	O
able	O
to	O
deliver	O
the	O
best	O
possible	O
ABM	O
solution	O

Vibration	B-KEY
control	I-KEY
of	O
structure	O
by	O
using	O
tuned	O
mass	O
damper	O
-LRB-	O
development	O
of	O
system	O
which	O
suppress	O
displacement	O
of	O
auxiliary	O
mass	O
-RRB-	O
In	O
vibration	B-KEY
control	I-KEY
of	O
a	O
structure	O
by	O
using	O
an	O
active	O
tuned	O
mass	O
damper	O
-LRB-	O
ATMD	O
-RRB-	O
,	O
stroke	O
of	O
the	O
auxiliary	O
mass	O
is	O
so	O
limited	O
that	O
it	O
is	O
difficult	O
to	O
control	O
the	O
vibration	O
in	O
the	O
case	O
of	O
large	O
disturbance	O
input	O
.	O
In	O
this	O
paper	O
,	O
two	O
methods	O
are	O
proposed	O
for	O
the	O
problem	O
.	O
One	O
of	O
the	O
methods	O
is	O
a	O
switching	O
control	B-KEY
system	O
by	O
two	O
types	O
of	O
controllers	B-KEY
.	O
One	O
of	O
the	O
controllers	B-KEY
is	O
a	O
normal	O
controller	B-KEY
under	O
small	O
relative	O
displacement	O
of	O
the	O
auxiliary	O
mass	O
,	O
and	O
the	O
other	O
is	O
not	O
effective	O
only	O
for	O
first	O
mode	O
of	O
vibration	O
under	O
large	O
relative	O
displacement	O
of	O
the	O
auxiliary	O
mass	O
.	O
New	O
variable	O
gain	O
control	B-KEY
system	O
is	O
constructed	O
by	O
switching	O
these	O
two	O
controllers	O
.	O
The	O
other	O
method	O
is	O
the	O
brake	B-KEY
system	I-KEY
.	O
In	O
active	O
vibration	B-KEY
control	I-KEY
,	O
it	O
is	O
necessary	O
to	O
use	O
actuator	O
for	O
active	O
control	O
.	O
By	O
using	O
the	O
actuator	B-KEY
,	O
the	O
proposed	O
system	O
puts	O
on	O
the	O
brake	O
to	O
suppress	O
displacement	O
increase	O
of	O
the	O
auxiliary	O
mass	O
under	O
large	O
disturbance	O
input	O
.	O
Finally	O
,	O
the	O
systems	O
are	O
designed	O
and	O
the	O
effectiveness	O
of	O
the	O
systems	O
is	O
confirmed	O
by	O
the	O
simulation	O

A	O
fuzzy-soft	B-KEY
learning	I-KEY
vector	I-KEY
quantization	I-KEY
for	O
control	B-KEY
chart	I-KEY
pattern	I-KEY
recognition	I-KEY
This	O
paper	O
presents	O
a	O
supervised	B-KEY
competitive	I-KEY
learning	I-KEY
network	I-KEY
approach	I-KEY
,	O
called	O
a	O
fuzzy-soft	B-KEY
learning	I-KEY
vector	I-KEY
quantization	I-KEY
,	O
for	O
control	B-KEY
chart	I-KEY
pattern	I-KEY
recognition	I-KEY
.	O
Unnatural	B-KEY
patterns	I-KEY
in	O
control	O
charts	O
mean	O
that	O
there	O
are	O
some	O
unnatural	O
causes	O
for	O
variations	O
in	O
statistical	B-KEY
process	I-KEY
control	I-KEY
-LRB-	O
SPC	B-KEY
-RRB-	O
.	O
Hence	O
,	O
control	B-KEY
chart	I-KEY
pattern	I-KEY
recognition	I-KEY
becomes	O
more	O
important	O
in	O
SPC.	O
In	O
order	O
to	O
detect	O
effectively	O
the	O
patterns	O
for	O
the	O
six	O
main	O
types	O
of	O
control	O
charts	O
,	O
Pham	O
and	O
Oztemel	O
-LRB-	O
1994	O
-RRB-	O
described	O
a	O
class	O
of	O
pattern	O
recognizers	O
for	O
control	O
charts	O
based	O
on	O
the	O
learning	O
vector	O
quantization	O
-LRB-	O
LVQ	O
-RRB-	O
such	O
as	O
LVQ	O
,	O
LVQ2	O
and	O
LVQ-X	O
etc.	O
.	O
In	O
this	O
paper	O
,	O
we	O
propose	O
a	O
new	O
supervised	B-KEY
LVQ	I-KEY
for	O
control	O
charts	O
based	O
on	O
a	O
fuzzy-soft	O
competitive	O
learning	O
network	O
.	O
The	O
proposed	O
fuzzy-soft	O
LVQ	O
-LRB-	O
FS-LVQ	O
-RRB-	O
uses	O
a	O
fuzzy	B-KEY
relaxation	I-KEY
technique	I-KEY
and	O
simultaneously	O
updates	O
all	O
neurons	O
.	O
It	O
can	O
increase	O
correct	B-KEY
recognition	I-KEY
accuracy	I-KEY
and	O
also	O
decrease	O
the	O
learning	B-KEY
time	I-KEY
.	O
Comparisons	O
between	O
LVQ	O
,	O
LVQ-X	O
and	O
FS-LVQ	O
are	O
made	O

Direct	B-KEY
self	I-KEY
control	I-KEY
with	O
minimum	O
torque	O
ripple	O
and	O
high	O
dynamics	O
for	O
a	O
double	O
three-level	O
GTO	O
inverter	O
drive	O
A	O
highly	B-KEY
dynamic	I-KEY
control	I-KEY
scheme	I-KEY
with	O
very	O
low	O
torque	O
ripple-direct	O
self	O
control	O
-LRB-	O
DSC	O
-RRB-	O
with	O
torque	O
hysteresis	O
control-for	O
very	O
high-power	O
medium-voltage	O
induction	O
motor	O
drives	O
fed	O
by	O
a	O
double	O
three-level	O
inverter	O
-LRB-	O
D3LI	O
-RRB-	O
is	O
presented	O
.	O
In	O
this	O
arrangement	O
,	O
two	O
three-level	O
inverters	O
that	O
are	O
connected	O
in	O
parallel	O
at	O
their	O
DC	O
sides	O
are	O
feeding	O
the	O
open	B-KEY
motor	I-KEY
windings	I-KEY
.	O
The	O
DSC	O
,	O
well	O
known	O
from	O
two	O
-	O
and	O
three-level	O
inverters	O
,	O
is	O
adapted	O
to	O
the	O
D3LI	O
and	O
optimized	O
for	O
a	O
minimum	O
torque	O
ripple	O
.	O
An	O
18-corner	O
trajectory	O
is	O
chosen	O
for	O
the	O
stator	B-KEY
flux	I-KEY
of	O
the	O
induction	O
machine	O
since	O
it	O
is	O
approaching	O
the	O
ideal	O
circle	O
much	O
better	O
than	O
the	O
hexagon	O
known	O
from	O
DSC	O
for	O
two-level	O
inverters	O
,	O
without	O
any	O
detriment	O
to	O
the	O
torque	O
ripple	O
.	O
The	O
machine	O
and	O
inverter	O
control	O
are	O
explained	O
and	O
the	O
proposed	O
torque	B-KEY
quality	I-KEY
and	O
dynamics	O
are	O
verified	O
by	O
measurements	O
on	O
a	O
180-kW	O
laboratory	O
drive	O

A	O
genetic	O
approach	O
to	O
the	O
optimization	O
of	O
automatic	O
generation	O
control	O
parameters	O
for	O
power	O
systems	O
This	O
paper	O
presents	O
a	O
method	O
based	O
on	O
genetic	B-KEY
algorithm	I-KEY
for	O
the	O
automatic	O
generation	O
control	O
of	O
power	O
systems	O
.	O
The	O
technique	O
is	O
applied	O
to	O
control	O
a	O
system	O
,	O
which	O
includes	O
two	O
areas	O
tied	O
together	O
through	O
a	O
power	B-KEY
line	I-KEY
.	O
As	O
a	O
consequence	O
of	O
continuous	B-KEY
load	I-KEY
variation	I-KEY
,	O
the	O
frequency	O
of	O
the	O
power	O
system	O
changes	O
with	O
time	O
.	O
In	O
conventional	O
studies	O
,	O
frequency	B-KEY
transients	I-KEY
are	O
minimized	O
by	O
using	O
integral	O
controllers	O
and	O
thus	O
zero	O
steady-state	O
error	O
is	O
obtained	O
.	O
In	O
this	O
paper	O
,	O
integral	B-KEY
controller	I-KEY
gains	I-KEY
and	O
frequency	B-KEY
bias	I-KEY
factors	I-KEY
are	O
determined	O
by	O
using	O
the	O
genetic	B-KEY
algorithm	I-KEY
.	O
The	O
results	O
of	O
simulation	O
reveal	O
the	O
application	O
of	O
the	O
genetic	B-KEY
algorithm	I-KEY
having	O
easy	O
implementation	O
to	O
find	O
the	O
global	O
optimum	O
values	O
of	O
the	O
control	O
parameters	O

Design	O
,	O
analysis	O
and	O
testing	O
of	O
some	O
parallel	B-KEY
two-step	I-KEY
W-methods	I-KEY
for	O
stiff	O
systems	O
Parallel	B-KEY
two-step	I-KEY
W-methods	I-KEY
are	O
linearly-implicit	B-KEY
integration	I-KEY
methods	I-KEY
where	O
the	O
s	O
stage	O
values	O
can	O
be	O
computed	O
in	O
parallel	O
.	O
We	O
construct	O
methods	O
of	O
stage	B-KEY
order	I-KEY
q	O
=	O
s	O
and	O
order	O
p	O
=	O
s	O
with	O
favourable	O
stability	B-KEY
properties	O
.	O
Generalizations	O
for	O
the	O
concepts	O
of	O
A	O
-	O
and	O
L-stability	O
are	O
proposed	O
and	O
conditions	O
for	O
stiff	O
accuracy	O
are	O
given	O
.	O
Numerical	O
comparisons	O
on	O
a	O
shared	B-KEY
memory	I-KEY
computer	I-KEY
show	O
the	O
efficiency	O
of	O
the	O
methods	O
,	O
especially	O
in	O
combination	O
with	O
Krylov-techniques	B-KEY
for	O
large	B-KEY
stiff	I-KEY
systems	I-KEY

Planning	O
linear	B-KEY
construction	I-KEY
projects	I-KEY
:	O
automated	O
method	O
for	O
the	O
generation	O
of	O
earthwork	B-KEY
activities	I-KEY
Earthworks	O
planning	O
for	O
road	B-KEY
construction	I-KEY
projects	I-KEY
is	O
a	O
complex	O
operation	O
and	O
the	O
planning	B-KEY
rules	I-KEY
used	O
are	O
usually	O
intuitive	O
and	O
not	O
well	O
defined	O
.	O
An	O
approach	O
to	O
automate	O
the	O
earthworks	B-KEY
planning	I-KEY
process	I-KEY
is	O
described	O
and	O
the	O
basic	O
techniques	O
that	O
are	O
used	O
are	O
outlined	O
.	O
A	O
computer-based	B-KEY
system	I-KEY
has	O
been	O
developed	O
,	O
initially	O
to	O
help	O
planners	O
use	O
existing	O
techniques	O
more	O
efficiently	O
.	O
With	O
their	O
input	O
,	O
the	O
system	O
has	O
been	O
extended	O
to	O
incorporate	O
a	O
knowledge	B-KEY
base	I-KEY
and	O
a	O
simulation	O
of	O
the	O
earthworks	O
processes	O
.	O
As	O
well	O
as	O
creating	O
activity	O
sets	O
in	O
a	O
much	O
shorter	O
time	O
,	O
the	O
system	O
has	O
shown	O
that	O
for	O
a	O
real	O
project	O
,	O
the	O
model	O
is	O
able	O
to	O
generate	O
activity	O
sets	O
that	O
are	O
comparable	O
to	O
those	O
generated	O
by	O
a	O
project	O
planner	O

Metaschemas	B-KEY
for	O
ER	O
,	O
ORM	B-KEY
and	O
UML	B-KEY
data	B-KEY
models	I-KEY
:	O
a	O
comparison	O
This	O
paper	O
provides	O
metaschemas	B-KEY
for	O
some	O
of	O
the	O
main	O
database	B-KEY
modeling	I-KEY
notations	I-KEY
used	O
in	O
industry	O
.	O
Two	O
Entity	O
Relationship	O
-LRB-	O
ER	O
-RRB-	O
notations	O
-LRB-	O
Information	B-KEY
Engineering	I-KEY
and	O
Barker	O
-RRB-	O
are	O
examined	O
in	O
detail	O
,	O
as	O
well	O
as	O
Object	B-KEY
Role	I-KEY
Modeling	I-KEY
-LRB-	O
ORM	B-KEY
-RRB-	O
conceptual	B-KEY
schema	I-KEY
diagrams	I-KEY
.	O
The	O
discussion	O
of	O
optionality	B-KEY
,	O
cardinality	B-KEY
and	O
multiplicity	B-KEY
is	O
widened	O
to	O
include	O
Unified	B-KEY
Modeling	I-KEY
Language	I-KEY
-LRB-	O
UML	B-KEY
-RRB-	O
class	B-KEY
diagrams	I-KEY
.	O
Issues	O
addressed	O
in	O
the	O
metamodel	O
analysis	O
include	O
the	O
normalization	B-KEY
impact	O
of	O
non-derived	O
constraints	O
on	O
derived	O
associations	O
,	O
the	O
influence	O
of	O
orthogonality	B-KEY
on	O
language	B-KEY
transparency	I-KEY
,	O
and	O
trade-offs	O
between	O
simplicity	O
and	O
expressibility	O
.	O
To	O
facilitate	O
comparison	O
,	O
the	O
same	O
modeling	O
notation	O
is	O
used	O
to	O
display	O
each	O
metaschema	B-KEY
.	O
For	O
this	O
purpose	O
,	O
ORM	B-KEY
is	O
used	O
because	O
of	O
its	O
greater	O
expressibility	O
and	O
clarity	O

Flexible	O
air-jet	O
tooling	O
for	O
vibratory	B-KEY
bowl	I-KEY
feeder	I-KEY
systems	O
Vibratory	B-KEY
bowl	I-KEY
feeders	I-KEY
-LRB-	O
VBFs	O
-RRB-	O
are	O
machines	O
that	O
feed	O
various	O
small	O
parts	O
in	O
large	O
volume	O
automatic	B-KEY
assembly	I-KEY
systems	I-KEY
.	O
Their	O
shortcomings	O
,	O
like	O
inflexibility	O
and	O
the	O
propensity	O
to	O
jam	O
,	O
stem	O
from	O
the	O
use	O
of	O
mechanical	O
orienting	B-KEY
devices	I-KEY
.	O
Air	O
jet	O
based	O
orienting	B-KEY
devices	I-KEY
can	O
be	O
implemented	O
to	O
overcome	O
these	O
limitations	O
.	O
Applications	O
of	O
passive	O
and	O
active	B-KEY
air	I-KEY
jet	I-KEY
based	O
orienting	B-KEY
devices	I-KEY
that	O
replace	O
conventional	O
devices	O
for	O
the	O
VBF	O
are	O
discussed	O
.	O
Passive	O
devices	O
,	O
which	O
reject	O
incorrectly	O
oriented	O
parts	O
,	O
are	O
discussed	O
first	O
.	O
Active	B-KEY
air	I-KEY
jet	I-KEY
based	O
orienting	B-KEY
devices	I-KEY
are	O
then	O
introduced	O
to	O
further	O
improve	O
the	O
flexibility	O
of	O
VBFs	O
.	O
Since	O
active	O
devices	O
reorient	O
parts	O
into	O
a	O
desired	O
orientation	O
,	O
the	O
part	O
motion	O
under	O
their	O
influence	O
is	O
analyzed	O
.	O
A	O
number	O
of	O
tests	O
demonstrate	O
the	O
feasibility	O
and	O
advantages	O
of	O
these	O
new	O
orienting	B-KEY
devices	I-KEY

Intelligent	B-KEY
optimal	I-KEY
sieving	I-KEY
method	O
for	O
FACTS	O
device	O
control	O
in	O
multi-machine	O
systems	O
A	O
multi-target	B-KEY
oriented	I-KEY
optimal	I-KEY
control	I-KEY
strategy	I-KEY
for	O
FACTS	B-KEY
devices	O
installed	O
in	O
multi-machine	O
power	O
systems	O
is	O
presented	O
in	O
this	O
paper	O
,	O
which	O
is	O
named	O
the	O
intelligent	B-KEY
optimal	I-KEY
sieving	I-KEY
control	O
-LRB-	O
IOSC	O
-RRB-	O
method	O
.	O
This	O
new	O
method	O
divides	O
the	O
FACTS	B-KEY
device	O
output	O
region	O
into	O
several	O
parts	O
and	O
selects	O
one	O
typical	O
value	O
from	O
each	O
part	O
,	O
which	O
is	O
called	O
output	O
candidate	O
.	O
Then	O
,	O
an	O
intelligent	B-KEY
optimal	I-KEY
sieve	I-KEY
is	O
constructed	O
,	O
which	O
predicts	O
the	O
impacts	O
of	O
each	O
output	O
candidate	O
on	O
a	O
power	O
system	O
and	O
sieves	O
out	O
an	O
optimal	O
output	O
from	O
all	O
of	O
the	O
candidates	O
.	O
The	O
artificial	B-KEY
neural	I-KEY
network	I-KEY
technologies	I-KEY
and	O
fuzzy	B-KEY
methods	I-KEY
are	O
applied	O
to	O
build	O
the	O
intelligent	O
sieve	O
.	O
Finally	O
,	O
the	O
real	O
control	B-KEY
signal	I-KEY
of	O
FACTS	B-KEY
devices	O
is	O
calculated	O
according	O
to	O
the	O
selected	B-KEY
optimal	I-KEY
output	I-KEY
through	O
inverse	B-KEY
system	I-KEY
method	I-KEY
.	O
Simulation	O
has	O
been	O
done	O
on	O
a	O
three-machine	B-KEY
power	I-KEY
system	I-KEY
and	O
the	O
results	O
show	O
that	O
the	O
proposed	O
IOSC	O
controller	O
can	O
effectively	O
attenuate	O
system	O
oscillations	O
and	O
enhance	O
the	O
power	O
system	O
transient	O
stability	O

Bounded	B-KEY
model	I-KEY
checking	I-KEY
for	O
the	O
universal	O
fragment	O
of	O
CTL	O
Bounded	B-KEY
Model	I-KEY
Checking	I-KEY
-LRB-	O
BMC	O
-RRB-	O
has	O
been	O
recently	O
introduced	O
as	O
an	O
efficient	O
verification	O
method	O
for	O
reactive	O
systems	O
.	O
BMC	O
based	O
on	O
SAT	B-KEY
methods	I-KEY
consists	O
in	O
searching	O
for	O
a	O
counterexample	O
of	O
a	O
particular	O
length	O
and	O
generating	O
a	O
propositional	B-KEY
formula	I-KEY
that	O
is	O
satisfiable	O
iff	O
such	O
a	O
counterexample-exists	O
.	O
This	O
new	O
technique	O
has	O
been	O
introduced	O
by	O
E.	O
Clarke	O
et	O
al.	O
for	O
model	B-KEY
checking	I-KEY
of	O
linear	B-KEY
time	I-KEY
temporal	I-KEY
logic	I-KEY
-LRB-	O
LTL	O
-RRB-	O
.	O
Our	O
paper	O
shows	O
how	O
the	O
concept	O
of	O
bounded	B-KEY
model	I-KEY
checking	I-KEY
can	O
be	O
extended	O
to	O
ACTL	O
-LRB-	O
the	O
universal	O
fragment	O
of	O
CTL	O
-RRB-	O
.	O
The	O
implementation	O
of	O
the	O
algorithm	O
for	O
Elementary	B-KEY
Net	I-KEY
Systems	I-KEY
is	O
described	O
together	O
with	O
the	O
experimental	O
results	O

Record	O
makers	O
-LSB-	O
UK	B-KEY
health	I-KEY
records	I-KEY
-RSB-	O
Plans	O
for	O
a	O
massive	O
cradle-to-grave	O
electronic	B-KEY
records	I-KEY
project	I-KEY
have	O
been	O
revealed	O
by	O
the	O
government	O
.	O
Is	O
the	O
scheme	O
really	O
viable	O
?	O

Enterprise	B-KEY
content	I-KEY
integration	I-KEY
III	O
:	O
Agari	O
Mediaware	O
's	O
Media	O
Star	O
Since	O
we	O
introduced	O
the	O
term	O
Enterprise	B-KEY
Content	I-KEY
Integration	I-KEY
-LRB-	O
ECI	O
-RRB-	O
in	O
January	O
,	O
the	O
concept	O
has	O
gained	O
momentum	O
in	O
the	O
market	O
.	O
In	O
addition	O
to	O
Context	O
Media	O
's	O
Interchange	O
Platform	O
and	O
Savantech	O
's	O
Photon	O
Commerce	O
,	O
Agari	O
Mediaware	O
's	O
Media	O
Star	O
is	O
in	O
the	O
fray	O
.	O
It	O
is	O
a	O
middleware	B-KEY
platform	O
that	O
allows	O
large	O
media	O
companies	O
to	O
integrate	O
their	O
digital	O
systems	O
with	O
great	O
flexibility	O

A	O
hybrid-neural	O
network	O
and	O
population	B-KEY
learning	I-KEY
algorithm	I-KEY
approach	O
to	O
solving	O
reliability	B-KEY
optimization	I-KEY
problem	I-KEY
Proposes	O
a	O
hybrid	B-KEY
approach	I-KEY
integrating	O
a	O
dedicated	B-KEY
artificial	I-KEY
neural	I-KEY
network	I-KEY
and	O
population	B-KEY
learning	I-KEY
algorithm	I-KEY
applied	O
to	O
maximising	O
system	B-KEY
reliability	I-KEY
under	O
cost	O
and	O
technical	B-KEY
feasibility	I-KEY
constraints	I-KEY
.	O
The	O
paper	O
includes	O
a	O
formulation	O
of	O
the	O
system	B-KEY
reliability	I-KEY
optimisation	O
-LRB-	O
SRO	O
-RRB-	O
problem	O
and	O
a	O
description	O
of	O
the	O
dedicated	O
neural	O
network	O
trained	O
by	O
applying	O
the	O
population	B-KEY
learning	I-KEY
algorithm	I-KEY
.	O
A	O
solution	O
to	O
the	O
example	O
SRO	O
problem	O
is	O
shown	O
and	O
results	O
of	O
the	O
computational	O
experiment	O
are	O
presented	O
and	O
discussed	O

Fabrication	O
of	O
polymeric	O
microlens	O
of	O
hemispherical	O
shape	O
using	O
micromolding	B-KEY
Polymeric	B-KEY
microlenses	I-KEY
play	O
an	O
important	O
role	O
in	O
reducing	O
the	O
size	B-KEY
,	O
weight	B-KEY
,	O
and	O
cost	B-KEY
of	O
optical	B-KEY
data	I-KEY
storage	I-KEY
and	O
optical	B-KEY
communication	I-KEY
systems	I-KEY
.	O
We	O
fabricate	O
polymeric	B-KEY
microlenses	I-KEY
using	O
the	O
microcompression	B-KEY
molding	I-KEY
process	I-KEY
.	O
The	O
design	O
and	O
fabrication	B-KEY
procedures	I-KEY
for	O
mold	B-KEY
insertion	I-KEY
is	O
simplified	O
using	O
silicon	B-KEY
instead	O
of	O
metal	O
.	O
PMMA	B-KEY
powder	I-KEY
is	O
used	O
as	O
the	O
molding	B-KEY
material	I-KEY
.	O
Governed	O
by	O
process	B-KEY
parameters	I-KEY
such	O
as	O
temperature	B-KEY
and	O
pressure	B-KEY
histories	O
,	O
the	O
micromolding	B-KEY
process	O
is	O
controlled	O
to	O
minimize	O
various	O
defects	O
that	O
develop	O
during	O
the	O
molding	O
process	O
.	O
The	O
radius	O
of	O
curvature	O
and	O
magnification	B-KEY
ratio	I-KEY
of	O
fabricated	O
microlens	O
are	O
measured	O
as	O
150	O
mu	O
m	O
and	O
over	O
3.0	O
,	O
respectively	O

Implementing	O
equals	O
for	O
mixed-type	B-KEY
comparison	I-KEY
The	O
idea	O
of	O
comparing	O
objects	O
of	O
different	O
types	O
is	O
not	O
entirely	O
off	O
base	O
,	O
in	O
particular	O
for	O
classes	O
from	O
the	O
same	O
class	O
hierarchy	O
.	O
After	O
all	O
,	O
objects	O
from	O
the	O
same	O
class	O
hierarchy	O
-LRB-	O
and	O
by	O
class	O
hierarchy	O
we	O
mean	O
all	O
classes	O
derived	O
from	O
a	O
common	O
superclass	B-KEY
other	O
than	O
Object	O
-RRB-	O
have	O
something	O
in	O
common	O
,	O
namely	O
at	O
least	O
the	O
superclass	B-KEY
part	O
.	O
As	O
we	O
demonstrated	O
in	O
a	O
previous	O
paper	O
-LRB-	O
2002	O
-RRB-	O
,	O
providing	O
a	O
correct	O
implementation	O
of	O
a	O
mixed-type	B-KEY
comparison	I-KEY
is	O
a	O
non-trivial	O
task	O
.	O
In	O
this	O
article	O
,	O
we	O
will	O
show	O
one	O
way	O
of	O
implementing	O
a	O
mixed-type	B-KEY
comparison	I-KEY
of	O
objects	O
from	O
the	O
same	O
class	O
hierarchy	O
that	O
meets	O
the	O
requirements	O
of	O
the	O
equals	B-KEY
contract	I-KEY

Polarization	O
of	O
the	O
RF	O
field	O
in	O
a	O
human	O
head	O
at	O
high	O
field	O
:	O
a	O
study	O
with	O
a	O
quadrature	B-KEY
surface	I-KEY
coil	I-KEY
at	O
7.0	B-KEY
T	I-KEY
The	O
RF	B-KEY
field	I-KEY
intensity	I-KEY
distribution	I-KEY
in	O
the	O
human	B-KEY
brain	I-KEY
becomes	O
inhomogeneous	O
due	O
to	O
wave	O
behavior	O
at	O
high	O
field	O
.	O
This	O
is	O
further	O
complicated	O
by	O
the	O
spatial	B-KEY
distribution	I-KEY
of	O
RF	B-KEY
field	I-KEY
polarization	I-KEY
that	O
must	O
be	O
considered	O
to	O
predict	O
image	B-KEY
intensity	I-KEY
distribution	I-KEY
.	O
An	O
additional	O
layer	O
of	O
complexity	O
is	O
involved	O
when	O
a	O
quadrature	O
coil	O
is	O
used	O
for	O
transmission	O
and	O
reception	O
.	O
To	O
study	O
such	O
complicated	O
RF	O
field	O
behavior	O
,	O
a	O
computer	B-KEY
modeling	I-KEY
method	O
was	O
employed	O
to	O
investigate	O
the	O
RF	O
field	O
of	O
a	O
quadrature	B-KEY
surface	I-KEY
coil	I-KEY
at	O
300	B-KEY
MHz	I-KEY
.	O
Theoretical	O
and	O
experimental	O
results	O
for	O
a	O
phantom	O
and	O
the	O
human	O
head	O
at	O
7.0	B-KEY
T	I-KEY
are	O
presented	O
.	O
The	O
results	O
are	O
theoretically	O
important	O
and	O
practically	O
useful	O
for	O
high-field	O
quadrature	O
coil	O
design	O
and	O
application	O

Assignment	O
of	O
periods	O
and	O
priorities	O
of	O
messages	O
and	O
tasks	O
in	O
distributed	B-KEY
control	I-KEY
systems	I-KEY
Presents	O
a	O
task	O
and	O
message-based	B-KEY
scheduling	I-KEY
method	I-KEY
to	O
guarantee	O
the	O
given	O
end-to-end	B-KEY
constraints	I-KEY
including	O
precedence	B-KEY
constraints	I-KEY
,	O
time	B-KEY
constraints	I-KEY
,	O
and	O
period	O
and	O
priority	O
of	O
task	O
and	O
message	O
.	O
The	O
method	O
is	O
an	O
integrated	O
one	O
considering	O
both	O
tasks	O
executed	O
in	O
each	O
node	O
and	O
messages	O
transmitted	O
via	O
the	O
network	O
and	O
is	O
designed	O
to	O
apply	O
to	O
a	O
general	O
distributed	B-KEY
control	I-KEY
system	I-KEY
that	O
has	O
multiple	O
loops	O
and	O
a	O
single	O
loop	O
has	O
sensor	O
nodes	O
with	O
multiple	O
sensors	O
,	O
actuator	O
nodes	O
with	O
multiple	O
actuators	O
,	O
controller	O
nodes	O
with	O
multiple	O
tasks	O
,	O
and	O
several	O
types	O
of	O
constraints	O
.	O
The	O
assigning	O
method	O
of	O
the	O
optimal	O
period	O
and	O
priority	O
of	O
task	O
and	O
message	O
is	O
proposed	O
,	O
using	O
the	O
presented	O
task	O
and	O
message-based	B-KEY
scheduling	I-KEY
method	I-KEY

Optical	B-KEY
setup	I-KEY
and	O
analysis	O
of	O
disk-type	B-KEY
photopolymer	I-KEY
high-density	I-KEY
holographic	I-KEY
storage	I-KEY
A	O
relatively	O
simple	O
scheme	O
for	O
disk-type	B-KEY
photopolymer	I-KEY
high-density	I-KEY
holographic	I-KEY
storage	I-KEY
based	O
on	O
angular	O
and	O
spatial	B-KEY
multiplexing	I-KEY
is	O
described	O
.	O
The	O
effects	O
of	O
the	O
optical	B-KEY
setup	I-KEY
on	O
the	O
recording	B-KEY
capacity	I-KEY
and	O
density	O
are	O
studied	O
.	O
Calculations	O
and	O
analysis	O
show	O
that	O
this	O
scheme	O
is	O
more	O
effective	O
than	O
a	O
scheme	O
based	O
on	O
the	O
spatioangular	O
multiplexing	O
for	O
disk-type	B-KEY
photopolymer	I-KEY
high-density	I-KEY
holographic	I-KEY
storage	I-KEY
,	O
which	O
has	O
a	O
limited	B-KEY
medium	I-KEY
thickness	I-KEY
.	O
Also	O
an	O
optimal	B-KEY
beam	I-KEY
recording	I-KEY
angle	I-KEY
exists	O
to	O
achieve	O
maximum	B-KEY
recording	I-KEY
capacity	I-KEY
and	O
density	O

Raising	O
the	O
standard	O
of	O
management	O
education	O
for	O
electronic	B-KEY
commerce	I-KEY
professionals	I-KEY
The	O
teaching	O
of	O
electronic	O
commerce	O
in	O
universities	B-KEY
has	O
become	O
a	O
growth	O
industry	O
in	O
itself	O
.	O
The	O
rapid	O
expansion	O
of	O
electronic	O
commerce	O
programmes	O
raises	O
the	O
question	O
of	O
what	O
actually	O
is	O
being	O
taught	O
.	O
The	O
association	O
of	O
electronic	O
commerce	O
as	O
primarily	O
a	O
technical	O
or	O
information	B-KEY
technology	I-KEY
-LRB-	O
IT	B-KEY
-RRB-	O
phenomenon	O
has	O
not	O
been	O
sufficient	O
to	O
constrain	O
it	B-KEY
to	O
IT	B-KEY
and	O
information	B-KEY
systems	I-KEY
departments	O
.	O
Business	B-KEY
schools	I-KEY
have	O
been	O
keen	O
entrants	O
into	O
the	O
electronic	O
commerce	O
coursework	O
race	O
and	O
they	O
are	O
developing	O
electronic	O
commerce	O
programmes	O
in	O
an	O
environment	O
where	O
there	O
is	O
no	O
agreed	O
definition	O
of	O
the	O
term	O
.	O
This	O
paper	O
draws	O
on	O
the	O
work	O
of	O
Kenneth	B-KEY
Boulding	I-KEY
who	O
argued	O
that	O
the	O
dynamics	O
of	O
change	O
in	O
society	O
are	O
largely	O
a	O
product	O
of	O
changing	O
skills	O
and	O
the	O
way	O
these	O
skills	O
are	O
arranged	O
into	O
roles	O
at	O
the	O
organizational	B-KEY
level	I-KEY
.	O
It	B-KEY
is	O
argued	O
that	O
an	O
overly	O
technical	O
interpretation	O
of	O
electronic	O
commerce	O
narrows	O
the	O
skills	O
being	O
acquired	O
as	O
part	O
of	O
formal	B-KEY
education	I-KEY
.	O
Universities	B-KEY
,	O
under	O
pressure	O
from	O
the	O
market	O
and	O
technological	O
change	O
,	O
are	O
changing	O
their	O
roles	O
resulting	O
in	O
a	O
further	O
narrowing	O
of	O
the	O
breadth	O
of	O
issues	O
that	O
is	O
seen	O
as	O
legitimate	O
to	O
be	O
included	O
as	O
electronic	O
commerce	O
.	O
The	O
outcome	O
is	O
that	O
aspiring	O
electronic	B-KEY
commerce	I-KEY
professionals	I-KEY
are	O
not	O
being	O
exposed	O
to	O
a	O
wide	O
enough	O
agenda	O
of	O
ideas	O
and	O
concepts	O
that	O
will	O
assist	O
them	O
to	O
make	O
better	O
business	O
decisions	O

Perfusion	B-KEY
quantification	I-KEY
using	O
Gaussian	B-KEY
process	I-KEY
deconvolution	I-KEY
The	O
quantification	O
of	O
perfusion	O
using	O
dynamic	B-KEY
susceptibility	I-KEY
contrast	I-KEY
MRI	I-KEY
-LRB-	O
DSC-MRI	O
-RRB-	O
requires	O
deconvolution	O
to	O
obtain	O
the	O
residual	B-KEY
impulse	I-KEY
response	I-KEY
function	I-KEY
-LRB-	O
IRF	O
-RRB-	O
.	O
In	O
this	O
work	O
,	O
a	O
method	O
using	O
the	O
Gaussian	O
process	O
for	O
deconvolution	O
-LRB-	O
GPD	O
-RRB-	O
is	O
proposed	O
.	O
The	O
fact	O
that	O
the	O
IRF	O
is	O
smooth	O
is	O
incorporated	O
as	O
a	O
constraint	O
in	O
the	O
method	O
.	O
The	O
GPD	O
method	O
,	O
which	O
automatically	O
estimates	O
the	O
noise	B-KEY
level	I-KEY
in	O
each	O
voxel	O
,	O
has	O
the	O
advantage	O
that	O
model	O
parameters	O
are	O
optimized	O
automatically	O
.	O
The	O
GPD	O
is	O
compared	O
to	O
singular	B-KEY
value	I-KEY
decomposition	I-KEY
-LRB-	O
SVD	O
-RRB-	O
using	O
a	O
common	O
threshold	O
for	O
the	O
singular	O
values	O
,	O
and	O
to	O
SVD	O
using	O
a	O
threshold	O
optimized	O
according	O
to	O
the	O
noise	B-KEY
level	I-KEY
in	O
each	O
voxel	O
.	O
The	O
comparison	O
is	O
carried	O
out	O
using	O
artificial	O
data	O
as	O
well	O
as	O
data	O
from	O
healthy	O
volunteers	O
.	O
It	O
is	O
shown	O
that	O
GPD	O
is	O
comparable	O
to	O
SVD	O
with	O
a	O
variable	O
optimized	O
threshold	O
when	O
determining	O
the	O
maximum	O
of	O
the	O
IRF	O
,	O
which	O
is	O
directly	O
related	O
to	O
the	O
perfusion	O
.	O
GPD	O
provides	O
a	O
better	O
estimate	O
of	O
the	O
entire	O
IRF	O
.	O
As	O
the	O
signal-to-noise	O
ratio	O
-LRB-	O
SNR	O
-RRB-	O
increases	O
or	O
the	O
time	O
resolution	O
of	O
the	O
measurements	O
increases	O
,	O
GPD	O
is	O
shown	O
to	O
be	O
superior	O
to	O
SVD	O
.	O
This	O
is	O
also	O
found	O
for	O
large	O
distribution	O
volumes	O

AGC	O
for	O
autonomous	B-KEY
power	I-KEY
system	I-KEY
using	O
combined	B-KEY
intelligent	I-KEY
techniques	I-KEY
In	O
the	O
present	O
work	O
two	O
intelligent	O
load	O
frequency	B-KEY
controllers	I-KEY
have	O
been	O
developed	O
to	O
regulate	O
the	O
power	O
output	O
and	O
system	O
frequency	O
by	O
controlling	O
the	O
speed	O
of	O
the	O
generator	O
with	O
the	O
help	O
of	O
fuel	B-KEY
rack	I-KEY
position	I-KEY
control	I-KEY
.	O
The	O
first	O
controller	O
is	O
obtained	O
using	O
fuzzy	B-KEY
logic	I-KEY
-LRB-	O
FL	O
-RRB-	O
only	O
,	O
whereas	O
the	O
second	O
one	O
by	O
using	O
a	O
combination	O
of	O
FL	O
,	O
genetic	B-KEY
algorithms	I-KEY
and	O
neural	B-KEY
networks	I-KEY
.	O
The	O
aim	O
of	O
the	O
proposed	O
controller	O
-LRB-	O
s	O
-RRB-	O
is	O
to	O
restore	O
in	O
a	O
very	O
smooth	O
way	O
the	O
frequency	O
to	O
its	O
nominal	O
value	O
in	O
the	O
shortest	O
time	O
possible	O
whenever	O
there	O
is	O
any	O
change	O
in	O
the	O
load	B-KEY
demand	I-KEY
etc.	O
.	O
The	O
action	O
of	O
these	O
controller	O
-LRB-	O
s	O
-RRB-	O
provides	O
a	O
satisfactory	O
balance	O
between	O
frequency	B-KEY
overshoot	I-KEY
and	O
transient	B-KEY
oscillations	I-KEY
with	O
zero	B-KEY
steady-state	I-KEY
error	I-KEY
.	O
The	O
design	O
and	O
performance	B-KEY
evaluation	I-KEY
of	O
the	O
proposed	O
controller	O
-LRB-	O
s	O
-RRB-	O
structure	O
are	O
illustrated	O
with	O
the	O
help	O
of	O
case	O
studies	O
applied	O
-LRB-	O
without	O
loss	O
of	O
generality	O
-RRB-	O
to	O
a	O
typical	O
single-area	B-KEY
power	I-KEY
system	I-KEY
.	O
It	O
is	O
found	O
that	O
the	O
proposed	O
controllers	O
exhibit	O
satisfactory	O
overall	B-KEY
dynamic	I-KEY
performance	I-KEY
and	O
overcome	O
the	O
possible	O
drawbacks	O
associated	O
with	O
other	O
competing	B-KEY
techniques	I-KEY

Modeling	O
dynamic	O
objects	O
in	O
distributed	B-KEY
systems	I-KEY
with	O
nested	B-KEY
Petri	I-KEY
nets	I-KEY
Nested	B-KEY
Petri	I-KEY
nets	I-KEY
-LRB-	O
NP-nets	O
-RRB-	O
is	O
a	O
Petri	O
net	O
extension	O
,	O
allowing	O
tokens	O
in	O
a	O
net	O
marking	O
to	O
be	O
represented	O
by	O
marked	O
nets	O
themselves	O
.	O
The	O
paper	O
discusses	O
applicability	O
of	O
NP-nets	O
for	O
modeling	O
task	O
planning	O
systems	O
,	O
multi-agent	B-KEY
systems	I-KEY
and	O
recursive-parallel	B-KEY
systems	I-KEY
.	O
A	O
comparison	O
of	O
NP-nets	O
with	O
some	O
other	O
formalisms	O
,	O
such	O
as	O
OPNs	O
of	O
R.	O
Valk	O
-LRB-	O
2000	O
-RRB-	O
,	O
recursive	O
parallel	O
programs	O
of	O
O.	O
Kushnarenko	O
and	O
Ph.	O
Schnoebelen	O
-LRB-	O
1997	O
-RRB-	O
and	O
process	B-KEY
algebras	I-KEY
is	O
given	O
.	O
Some	O
aspects	O
of	O
decidability	B-KEY
for	O
object-oriented	B-KEY
Petri	I-KEY
net	I-KEY
extensions	O
are	O
also	O
discussed	O

Hardware	O
and	O
software	B-KEY
platform	I-KEY
for	O
real-time	B-KEY
processing	I-KEY
and	O
visualization	O
of	O
echographic	B-KEY
radiofrequency	I-KEY
signals	I-KEY
In	O
this	O
paper	O
the	O
architecture	O
of	O
a	O
hardware	O
and	O
software	B-KEY
platform	I-KEY
,	O
for	O
ultrasonic	O
investigation	O
is	O
presented	O
.	O
The	O
platform	O
,	O
used	O
in	O
conjunction	O
with	O
an	O
analog	O
front-end	O
hardware	O
for	O
driving	O
the	O
ultrasonic	O
transducers	O
of	O
any	O
commercial	O
echograph	O
,	O
having	O
the	O
radiofrequency	O
echo	O
signal	O
access	O
,	O
make	O
it	O
possible	O
to	O
dispose	O
of	O
a	O
powerful	O
echographic	O
system	O
for	O
experimenting	O
any	O
processing	O
technique	O
,	O
also	O
in	O
a	O
clinical	O
environment	O
in	O
which	O
real-time	O
operation	O
mode	O
is	O
an	O
essential	O
prerequisite	O
.	O
The	O
platform	O
transforms	O
any	O
echograph	O
into	O
a	O
test-system	O
for	O
evaluating	O
the	O
diagnostic	O
effectiveness	O
of	O
new	O
investigation	O
techniques	O
.	O
A	O
particular	O
user	B-KEY
interface	I-KEY
was	O
designed	O
in	O
order	O
to	O
allow	O
a	O
real-time	O
and	O
simultaneous	O
visualization	O
of	O
the	O
results	O
produced	O
in	O
the	O
different	O
stages	O
of	O
the	O
chosen	O
processing	O
procedure	O
.	O
This	O
is	O
aimed	O
at	O
obtaining	O
a	O
better	O
optimization	O
of	O
the	O
processing	O
algorithm	O
.	O
The	O
most	O
important	O
platform	O
aspect	O
,	O
which	O
also	O
constitutes	O
the	O
basic	O
differentiation	O
with	O
respect	O
to	O
similar	O
systems	O
,	O
is	O
the	O
direct	O
processing	O
of	O
the	O
radiofrequency	O
echo	O
signal	O
,	O
which	O
is	O
essential	O
for	O
a	O
complete	O
analysis	O
of	O
the	O
particular	O
ultrasound-media	O
interaction	O
phenomenon	O
.	O
The	O
platform	O
completely	O
integrates	O
the	O
architecture	O
of	O
a	O
personal	B-KEY
computer	I-KEY
-LRB-	O
PC	O
-RRB-	O
giving	O
rise	O
to	O
several	O
benefits	O
,	O
such	O
as	O
the	O
quick	O
technological	O
evolution	O
in	O
the	O
PC	O
field	O
and	O
an	O
extreme	O
degree	O
of	O
programmability	O
for	O
different	O
applications	O
.	O
The	O
PC	O
also	O
constitutes	O
the	O
user	B-KEY
interface	I-KEY
,	O
as	O
a	O
flexible	O
and	O
intuitive	O
visualization	O
support	O
,	O
and	O
performs	O
some	O
software	O
signal	O
processing	O
,	O
by	O
custom	O
algorithms	O
and	O
commercial	O
libraries	O
.	O
The	O
realized	O
close	O
synergy	O
between	O
hardware	O
and	O
software	O
allows	O
the	O
acquisition	O
and	O
real-time	B-KEY
processing	I-KEY
of	O
the	O
echographic	O
radiofrequency	O
-LRB-	O
RF	O
-RRB-	O
signal	O
with	O
fast	O
data	O
representation	O

Strobbe	B-KEY
Graphics	I-KEY
'	O
next	O
frontier	O
:	O
CTP	O
for	O
commercial	O
printers	O
Strobbe	O
is	O
one	O
of	O
the	O
more	O
successful	O
makers	O
of	O
newspaper	O
platesetters	B-KEY
,	O
which	O
are	O
sold	O
by	O
Agfa	B-KEY
under	O
the	O
Polaris	B-KEY
name	O
.	O
But	O
the	O
company	O
also	O
has	O
a	O
growing	O
presence	O
in	O
commercial	B-KEY
printing	I-KEY
markets	O
,	O
where	O
it	O
sells	O
under	O
its	O
own	O
name	O

Successive	O
expansion	O
method	O
of	O
network	O
planning	O
applying	O
symbolic	B-KEY
analysis	I-KEY
method	O
The	O
conventional	O
power	B-KEY
system	I-KEY
network	I-KEY
successive	I-KEY
expansion	I-KEY
planning	I-KEY
method	O
is	O
discussed	O
in	O
the	O
context	O
of	O
the	O
new	O
paradigm	O
of	O
competitive	O
electric	O
power	O
,	O
energy	O
and	O
service	O
market	O
.	O
In	O
sequel	O
,	O
the	O
paper	O
presents	O
an	O
application	O
of	O
the	O
conceptually	O
new	O
computer	B-KEY
program	I-KEY
,	O
based	O
on	O
the	O
symbolic	B-KEY
analysis	I-KEY
of	O
load	B-KEY
flows	I-KEY
in	O
power	O
system	O
networks	O
.	O
The	O
network	O
parameters	O
and	O
variables	O
are	O
defined	O
as	O
symbols	O
.	O
The	O
symbolic	B-KEY
analyzer	I-KEY
,	O
which	O
models	O
analytically	O
the	O
power	O
system	O
DC	O
load	B-KEY
flows	I-KEY
,	O
enables	O
the	O
sensitivity	O
analysis	O
of	O
the	O
power	O
system	O
to	O
parameter	O
and	O
variable	O
variations	O
-LRB-	O
costs	O
,	O
transfers	O
,	O
injections	O
-RRB-	O
,	O
a	O
valuable	O
tool	O
for	O
the	O
expansion	O
planning	O
analysis	O
.	O
That	O
virtue	O
could	O
not	O
be	O
found	O
within	O
the	O
conventional	O
approach	O
,	O
relying	O
on	O
compensation	B-KEY
methods	I-KEY
,	O
precalculated	B-KEY
distribution	I-KEY
factors	I-KEY
,	O
and	O
so	O
on	O
.	O
This	O
novel	O
application	O
sheds	O
some	O
light	O
on	O
the	O
traditional	O
power	B-KEY
system	I-KEY
network	I-KEY
expansion	I-KEY
planning	I-KEY
method	I-KEY
,	O
as	O
well	O
as	O
on	O
its	O
possible	O
application	O
within	O
the	O
system	O
network	O
expansion	O
planning	O
in	O
the	O
new	O
environment	O
assuming	O
the	O
competitive	B-KEY
electric	I-KEY
power	I-KEY
market	I-KEY

A	O
framework	O
for	O
rapid	B-KEY
local	I-KEY
area	I-KEY
modeling	I-KEY
for	O
construction	B-KEY
automation	I-KEY
Rapid	B-KEY
3D	I-KEY
positioning	I-KEY
and	O
modeling	O
in	O
construction	O
can	O
be	O
used	O
to	O
more	O
effectively	O
plan	O
,	O
visualize	O
,	O
and	O
communicate	O
operations	O
before	O
execution	O
.	O
It	O
can	O
also	O
help	O
to	O
optimize	O
equipment	B-KEY
operations	I-KEY
,	O
significantly	O
improve	O
safety	O
,	O
and	O
enhance	O
a	O
remote	O
operator	O
's	O
spatial	B-KEY
perception	I-KEY
of	O
the	O
workspace	O
.	O
A	O
new	O
framework	O
for	O
rapid	B-KEY
local	I-KEY
area	I-KEY
sensing	I-KEY
and	O
3D	B-KEY
modeling	I-KEY
for	O
better	O
planning	O
and	O
control	O
of	O
construction	O
equipment	B-KEY
operation	I-KEY
is	O
described	O
and	O
demonstrated	O
.	O
By	O
combining	O
human-assisted	B-KEY
graphical	I-KEY
workspace	I-KEY
modeling	I-KEY
with	O
pre-stored	O
Computer-Aided	O
Design	O
-LRB-	O
CAD	O
-RRB-	O
models	O
and	O
simple	O
sensors	O
-LRB-	O
such	O
as	O
single-axis	B-KEY
laser	I-KEY
rangefinders	I-KEY
and	O
remote	B-KEY
video	I-KEY
cameras	I-KEY
-RRB-	O
,	O
modeling	O
time	O
can	O
be	O
significantly	O
reduced	O
while	O
potentially	O
increasing	O
modeling	O
accuracy	O

On	O
Implicit	O
Euler	O
for	O
high-order	O
high-index	O
DAEs	O
The	O
Implicit	B-KEY
Euler	I-KEY
method	I-KEY
is	O
seldom	O
used	O
to	O
solve	O
differential-algebraic	B-KEY
equations	I-KEY
-LRB-	O
DAEs	O
-RRB-	O
of	O
differential	B-KEY
index	I-KEY
r	O
>	O
or	O
=	O
3	O
,	O
since	O
the	O
method	O
in	O
general	O
fails	O
to	O
converge	B-KEY
in	O
the	O
first	O
r	O
-	O
2	O
steps	O
after	O
a	O
change	O
of	O
stepsize	O
.	O
However	O
,	O
if	O
the	O
differential	O
equation	O
is	O
of	O
order	O
d	O
=	O
r	O
-	O
1	O
>	O
or	O
=	O
1	O
,	O
an	O
alternative	O
variable-step	O
version	O
of	O
the	O
Euler	O
method	O
can	O
be	O
shown	O
uniformly	O
convergent	B-KEY
.	O
For	O
d	O
=	O
r	O
-	O
1	O
,	O
this	O
variable-step	B-KEY
method	I-KEY
is	O
equivalent	O
to	O
the	O
Implicit	O
Euler	O
except	O
for	O
the	O
first	O
r	O
-	O
2	O
steps	O
after	O
a	O
change	O
of	O
stepsize	O
.	O
Generalization	O
to	O
DAEs	O
with	O
differential	O
equations	O
of	O
order	O
d	O
>	O
r	O
-	O
1	O
>	O
or	O
=	O
1	O
,	O
and	O
to	O
variable-order	B-KEY
formulas	I-KEY
is	O
discussed	O

Production	B-KEY
capacity	I-KEY
of	O
flexible	B-KEY
manufacturing	I-KEY
systems	I-KEY
with	O
fixed	B-KEY
production	I-KEY
ratios	I-KEY
Determining	O
the	O
production	B-KEY
capacity	I-KEY
of	O
flexible	B-KEY
manufacturing	I-KEY
systems	I-KEY
is	O
a	O
very	O
important	O
issue	O
in	O
the	O
design	O
of	O
such	O
systems	O
.	O
We	O
propose	O
an	O
approach	O
for	O
determining	O
the	O
production	B-KEY
capacity	I-KEY
-LRB-	O
i.e.	O
the	O
maximum	B-KEY
production	I-KEY
rate	I-KEY
-RRB-	O
of	O
a	O
flexible	B-KEY
manufacturing	I-KEY
system	I-KEY
with	O
several	O
part	O
types	O
,	O
dedicated	B-KEY
pallets	I-KEY
,	O
and	O
fixed	B-KEY
production	I-KEY
ratios	I-KEY
among	O
the	O
different	O
part	O
types	O
.	O
We	O
show	O
that	O
the	O
problem	O
reduces	O
to	O
the	O
determination	O
of	O
a	O
single	O
parameter	O
for	O
which	O
we	O
propose	O
an	O
iterative	B-KEY
procedure	I-KEY
.	O
Simulation	B-KEY
or	O
approximate	B-KEY
analytical	I-KEY
techniques	I-KEY
can	O
be	O
used	O
as	O
the	O
building	B-KEY
block	I-KEY
performance	I-KEY
evaluation	I-KEY
technique	I-KEY
in	O
the	O
iterative	B-KEY
procedure	I-KEY

Regression	O
testing	O
of	O
database	B-KEY
applications	I-KEY
Database	B-KEY
applications	I-KEY
features	O
such	O
as	O
Structured	B-KEY
Query	I-KEY
Language	I-KEY
or	O
SQL	B-KEY
,	O
exception	B-KEY
programming	I-KEY
,	O
integrity	B-KEY
constraints	I-KEY
,	O
and	O
table	B-KEY
triggers	I-KEY
pose	O
difficulties	O
for	O
maintenance	O
activities	O
;	O
especially	O
for	O
regression	O
testing	O
that	O
follows	O
modifications	O
to	O
database	B-KEY
applications	I-KEY
.	O
In	O
this	O
work	O
,	O
we	O
address	O
these	O
difficulties	O
and	O
propose	O
a	O
two	B-KEY
phase	I-KEY
regression	I-KEY
testing	I-KEY
methodology	I-KEY
.	O
In	O
phase	O
1	O
,	O
we	O
explore	O
control	O
flow	O
and	O
data	B-KEY
flow	I-KEY
analysis	I-KEY
issues	O
of	O
database	B-KEY
applications	I-KEY
.	O
Then	O
,	O
we	O
propose	O
an	O
impact	B-KEY
analysis	I-KEY
technique	O
that	O
is	O
based	O
on	O
dependencies	O
that	O
exist	O
among	O
the	O
components	O
of	O
database	B-KEY
applications	I-KEY
.	O
This	O
analysis	O
leads	O
to	O
selecting	O
test	O
cases	O
from	O
the	O
initial	O
test	O
suite	O
for	O
regression	O
testing	O
the	O
modified	O
application	O
.	O
In	O
phase	O
2	O
,	O
further	O
reduction	O
in	O
the	O
regression	O
test	O
cases	O
is	O
performed	O
by	O
using	O
reduction	B-KEY
algorithms	I-KEY
.	O
We	O
present	O
two	O
such	O
algorithms	O
.	O
The	O
Graph	B-KEY
Walk	I-KEY
algorithm	I-KEY
walks	O
through	O
the	O
control	B-KEY
flow	I-KEY
graph	I-KEY
of	O
database	O
modules	O
and	O
selects	O
a	O
safe	O
set	O
of	O
test	O
cases	O
to	O
retest	O
.	O
The	O
Call	B-KEY
Graph	I-KEY
Firewall	I-KEY
algorithm	I-KEY
uses	O
a	O
firewall	O
for	O
the	O
inter	O
procedural	O
level	O
.	O
Finally	O
,	O
a	O
maintenance	O
environment	O
for	O
database	B-KEY
applications	I-KEY
is	O
described	O
.	O
Our	O
experience	O
with	O
this	O
regression	O
testing	O
methodology	O
shows	O
that	O
the	O
impact	B-KEY
analysis	I-KEY
technique	O
is	O
adequate	O
for	O
selecting	O
regression	O
tests	O
and	O
that	O
phase	O
2	O
techniques	O
can	O
be	O
used	O
for	O
further	O
reduction	O
in	O
the	O
number	O
of	O
theses	O
tests	O

Mining	O
the	O
optimal	O
class	O
association	O
rule	O
set	O
We	O
define	O
an	O
optimal	O
class	O
association	O
rule	O
set	O
to	O
be	O
the	O
minimum	B-KEY
rule	I-KEY
set	I-KEY
with	O
the	O
same	O
predictive	B-KEY
power	I-KEY
of	O
the	O
complete	O
class	O
association	O
rule	O
set	O
.	O
Using	O
this	O
rule	O
set	O
instead	O
of	O
the	O
complete	O
class	O
association	O
rule	O
set	O
we	O
can	O
avoid	O
redundant	B-KEY
computation	I-KEY
that	O
would	O
otherwise	O
be	O
required	O
for	O
mining	O
predictive	B-KEY
association	I-KEY
rules	I-KEY
and	O
hence	O
improve	O
the	O
efficiency	O
of	O
the	O
mining	O
process	O
significantly	O
.	O
We	O
present	O
an	O
efficient	O
algorithm	O
for	O
mining	O
the	O
optimal	O
class	O
association	O
rule	O
set	O
using	O
an	O
upward	B-KEY
closure	I-KEY
property	I-KEY
of	O
pruning	O
weak	O
rules	O
before	O
they	O
are	O
actually	O
generated	O
.	O
We	O
have	O
implemented	O
the	O
algorithm	O
and	O
our	O
experimental	B-KEY
results	I-KEY
show	O
that	O
our	O
algorithm	O
generates	O
the	O
optimal	O
class	O
association	O
rule	O
set	O
,	O
whose	O
size	O
is	O
smaller	O
than	O
1/17	O
of	O
the	O
complete	O
class	O
association	O
rule	O
set	O
on	O
average	O
,	O
in	O
significantly	O
less	O
time	O
than	O
generating	O
the	O
complete	O
class	O
association	O
rule	O
set	O
.	O
Our	O
proposed	O
criterion	O
has	O
been	O
shown	O
very	O
effective	O
for	O
pruning	O
weak	O
rules	O
in	O
dense	B-KEY
databases	I-KEY

Experimental	O
design	O
methodology	O
and	O
data	B-KEY
analysis	I-KEY
technique	I-KEY
applied	O
to	O
optimise	O
an	O
organic	B-KEY
synthesis	I-KEY
The	O
study	O
was	O
aimed	O
at	O
maximising	O
the	O
yield	O
of	O
a	O
Michaelis-Becker	B-KEY
dibromoalkane	I-KEY
monophosphorylation	I-KEY
reaction	I-KEY
.	O
In	O
order	O
to	O
save	O
time	O
and	O
money	O
,	O
we	O
first	O
applied	O
a	O
full	B-KEY
factorial	I-KEY
experimental	I-KEY
design	I-KEY
to	O
search	O
for	O
the	O
optimum	B-KEY
conditions	I-KEY
while	O
performing	O
a	O
small	O
number	O
of	O
experiments	O
.	O
We	O
then	O
used	O
the	O
principal	B-KEY
component	I-KEY
analysis	I-KEY
-LRB-	O
PCA	O
-RRB-	O
technique	O
to	O
evidence	O
two	O
uncontrolled	B-KEY
factors	I-KEY
.	O
Lastly	O
,	O
a	O
special	O
experimental	O
design	O
that	O
took	O
into	O
account	O
all	O
the	O
influential	O
factors	O
allowed	O
us	O
to	O
determine	O
the	O
maximum-yield	B-KEY
experimental	I-KEY
conditions	I-KEY
.	O
This	O
study	O
also	O
evidenced	O
the	O
complementary	O
nature	O
of	O
experimental	O
design	O
methodology	O
and	O
data	B-KEY
analysis	I-KEY
techniques	I-KEY

The	O
set	O
of	O
just-in-time	B-KEY
management	I-KEY
strategies	I-KEY
:	O
an	O
assessment	O
of	O
their	O
impact	O
on	O
plant-level	B-KEY
productivity	I-KEY
and	O
input-factor	B-KEY
substitutability	I-KEY
using	O
variable	B-KEY
cost	I-KEY
function	I-KEY
estimates	I-KEY
Many	O
manufacturers	O
in	O
the	O
automobile	B-KEY
industry	I-KEY
around	O
the	O
world	O
have	O
adopted	O
the	O
just-in-time	O
-LRB-	O
JIT	B-KEY
-RRB-	O
set	O
of	O
management	O
strategies	O
in	O
an	O
effort	O
to	O
improve	O
productivity	O
,	O
efficiency	B-KEY
and	O
product	B-KEY
quality	I-KEY
.	O
The	O
paper	O
provides	O
empirical	O
evidence	O
that	O
supports	O
the	O
idea	O
that	O
JIT	B-KEY
manufacturing	O
environments	O
are	O
,	O
in	O
fact	O
,	O
more	O
productive	O
than	O
their	O
non-JIT	O
counterparts	O
.	O
Plant-level	O
cross-sectional	O
data	O
from	O
auto-parts	B-KEY
manufacturing	I-KEY
firms	I-KEY
are	O
used	O
to	O
estimate	O
variable	O
cost	O
functions	O
for	O
a	O
JIT	B-KEY
group	O
as	O
well	O
as	O
for	O
a	O
non-JIT	O
group	O
of	O
plants	O
.	O
Differences	O
in	O
cost	O
function	O
characteristics	O
between	O
the	O
two	O
groups	O
are	O
examined	O
and	O
discussed	O

A	O
comparison	O
of	O
high-power	O
converter	O
topologies	O
for	O
the	O
implementation	O
of	O
FACTS	B-KEY
controllers	I-KEY
This	O
paper	O
compares	O
four	O
power	O
converter	O
topologies	O
for	O
the	O
implementation	O
of	O
flexible	O
AC	O
transmission	O
system	O
-LRB-	O
FACTS	O
-RRB-	O
controllers	O
:	O
three	O
multilevel	B-KEY
topologies	I-KEY
-LRB-	O
multipoint	O
clamped	O
-LRB-	O
MPC	O
-RRB-	O
,	O
chain	O
,	O
and	O
nested	O
cell	O
-RRB-	O
and	O
the	O
well-established	O
multipulse	B-KEY
topology	I-KEY
.	O
In	O
keeping	O
with	O
the	O
need	O
to	O
implement	O
very-high-power	O
inverters	B-KEY
,	O
switching	B-KEY
frequency	I-KEY
is	O
restricted	O
to	O
line	O
frequency	O
.	O
The	O
study	O
addresses	O
device	B-KEY
count	I-KEY
,	O
DC	B-KEY
filter	I-KEY
ratings	I-KEY
,	O
restrictions	O
on	O
voltage	O
control	O
,	O
active	O
power	O
transfer	O
through	O
the	O
DC	O
link	O
,	O
and	O
balancing	O
of	O
DC-link	O
voltages	O
.	O
Emphasis	O
is	O
placed	O
on	O
capacitor	O
sizing	O
because	O
of	O
its	O
impact	O
on	O
the	O
cost	O
and	O
size	O
of	O
the	O
FACTS	B-KEY
controller	I-KEY
.	O
A	O
method	O
for	O
the	O
dimensioning	O
the	O
DC	O
capacitor	O
filter	O
is	O
presented	O
.	O
It	O
is	O
found	O
that	O
the	O
chain	O
converter	O
is	O
attractive	O
for	O
the	O
implementation	O
of	O
a	O
static	B-KEY
compensator	I-KEY
or	O
a	O
static	B-KEY
synchronous	I-KEY
series	I-KEY
compensator	I-KEY
.	O
The	O
MPC	O
converter	O
is	O
attractive	O
for	O
the	O
implementation	O
of	O
a	O
unified	B-KEY
power	I-KEY
flow	I-KEY
controller	I-KEY
or	O
an	O
interline	O
power	O
flow	O
controller	O
,	O
but	O
a	O
special	O
arrangement	O
is	O
required	O
to	O
overcome	O
the	O
limitations	O
on	O
voltage	O
control	O

Bigger	O
is	O
better	O
:	O
the	O
influence	O
of	O
physical	O
size	O
on	O
aesthetic	B-KEY
preference	I-KEY
judgments	I-KEY
The	O
hypothesis	O
that	O
the	O
physical	O
size	O
of	O
an	O
object	O
can	O
influence	O
aesthetic	O
preferences	O
was	O
investigated	O
.	O
In	O
a	O
series	O
of	O
four	O
experiments	O
,	O
participants	O
were	O
presented	O
with	O
pairs	O
of	O
abstract	B-KEY
stimuli	I-KEY
and	O
asked	O
to	O
indicate	O
which	O
member	O
of	O
each	O
pair	O
they	O
preferred	O
.	O
A	O
preference	O
for	O
larger	O
stimuli	O
was	O
found	O
on	O
the	O
majority	O
of	O
trials	O
using	O
various	O
types	O
of	O
stimuli	O
,	O
stimuli	O
of	O
various	O
sizes	O
,	O
and	O
with	O
both	O
adult	O
and	O
3-year-old	O
participants	O
.	O
This	O
preference	B-KEY
pattern	I-KEY
was	O
disrupted	O
only	O
when	O
participants	O
had	O
both	O
stimuli	O
that	O
provided	O
a	O
readily	O
accessible	O
alternative	O
source	O
of	O
preference-evoking	B-KEY
information	I-KEY
and	O
sufficient	O
attentional	B-KEY
resources	I-KEY
to	O
make	O
their	O
preference	O
judgments	O

Quick	O
media	B-KEY
response	I-KEY
averts	O
PR	O
disaster	O
Sometimes	O
it	O
's	O
not	O
what	O
you	O
do	O
,	O
but	O
how	O
you	O
do	O
it	O
.	O
After	O
hackers	O
broke	O
the	O
blocking	O
code	O
on	O
the	O
home	O
version	O
of	O
its	O
popular	O
Cyber	B-KEY
Patrol	I-KEY
Internet	I-KEY
filtering	I-KEY
software	I-KEY
and	O
posted	O
it	O
on	O
the	O
Internet	O
,	O
marketers	O
at	O
Microsystems	B-KEY
Software	I-KEY
pulled	O
out	O
a	O
playbook	O
of	O
standard	O
crisis	B-KEY
management	I-KEY
and	O
PR	O
techniques	O
.	O
But	O
the	O
Cyber	O
Patrol	O
PR	O
team	O
including	O
outside	O
PR	O
counsel	O
and	O
the	O
company	O
's	O
outside	O
law	O
firm	O
,	O
used	O
those	O
tools	O
aggressively	O
in	O
order	O
to	O
turn	O
the	O
tide	O
of	O
public	O
and	O
media	O
opinion	O
away	O
from	O
the	O
hackers	O
,	O
who	O
initially	O
were	O
hailed	O
as	O
folk	O
heroes	O
,	O
and	O
in	O
favor	O
of	O
the	O
company	O
's	O
interests	O
,	O
to	O
save	O
the	O
product	O
's	O
and	O
the	O
company	O
's	O
reputations	O
and	O
inherent	O
value	O
.	O
And	O
the	O
entire	O
team	O
managed	O
to	O
move	O
at	O
Internet	O
speed	O
:	O
The	O
crisis	O
was	O
essentially	O
over	O
in	O
about	O
three	O
weeks	O

PKI	B-KEY
:	O
coming	O
to	O
an	O
enterprise	O
near	O
you	O
?	O
For	O
many	O
years	O
public	B-KEY
key	I-KEY
infrastructure	I-KEY
-LRB-	O
PKI	B-KEY
-RRB-	O
deployments	O
were	O
the	O
provenance	O
of	O
governments	O
and	O
large	O
,	O
security-conscious	O
corporations	O
and	O
financial	O
institutions	O
.	O
These	O
organizations	O
have	O
the	O
financial	O
and	O
human	O
resources	O
necessary	O
to	O
successfully	O
manage	O
the	O
complexities	O
of	O
a	O
public	O
key	O
system	O
.	O
Lately	O
however	O
,	O
several	O
forces	O
have	O
converged	O
to	O
encourage	O
a	O
broader	O
base	O
of	O
enterprises	O
to	O
take	O
a	O
closer	O
look	O
at	O
PKI	B-KEY
.	O
These	O
forces	O
are	O
discussed	O
.	O
PKI	B-KEY
vendors	O
are	O
now	O
demonstrating	O
to	O
customers	O
how	O
they	O
can	O
make	O
essential	O
business	O
applications	O
faster	O
and	O
more	O
efficient	O
by	O
moving	O
them	O
to	O
the	O
Internet-without	O
sacrificing	O
security	O
.	O
Those	O
applications	O
usually	O
include	O
secure	B-KEY
remote	O
access	O
,	O
secure	O
messaging	O
,	O
electronic	O
document	O
exchange	O
,	O
transaction	O
validation	O
,	O
and	O
network	O
authentication	O
.	O
After	O
a	O
brief	O
discussion	O
of	O
PKI	B-KEY
basics	O
the	O
author	O
reviews	O
various	O
products	O
available	O
on	O
the	O
market	O

A	O
new	O
result	O
on	O
the	O
global	O
convergence	O
of	O
Hopfield	B-KEY
neural	I-KEY
networks	I-KEY
In	O
this	O
work	O
,	O
we	O
discuss	O
Hopfield	B-KEY
neural	I-KEY
networks	I-KEY
,	O
investigating	O
their	O
global	B-KEY
stability	I-KEY
.	O
Some	O
sufficient	B-KEY
conditions	I-KEY
for	O
a	O
class	O
of	O
Hopfield	B-KEY
neural	I-KEY
networks	I-KEY
to	O
be	O
globally	O
stable	O
and	O
globally	O
exponentially	O
stable	O
are	O
given	O

New	O
approach	O
to	O
standing	O
phase	O
angle	O
reduction	O
for	O
power	B-KEY
system	I-KEY
restoration	I-KEY
During	O
power	B-KEY
system	I-KEY
restoration	I-KEY
,	O
it	O
is	O
necessary	O
to	O
check	O
the	O
phase	O
angle	O
between	O
two	O
buses	O
before	O
closing	O
circuit	O
breakers	O
to	O
connect	O
a	O
line	O
between	O
them	O
.	O
These	O
angles	O
may	O
occur	O
across	O
a	O
tie	O
line	O
between	O
two	O
systems	O
or	O
between	O
two	O
connected	O
subsystems	O
within	O
a	O
system	O
.	O
In	O
case	O
of	O
large	O
standing	O
phase	O
angle	O
-LRB-	O
SPA	O
-RRB-	O
difference	O
the	O
synchro-check	O
relay	O
does	O
not	O
allow	O
closing	O
of	O
the	O
breaker	O
for	O
this	O
line	O
.	O
Therefore	O
,	O
this	O
excessive	O
SPA	O
has	O
to	O
be	O
reduced	O
before	O
attempting	O
to	O
connect	O
the	O
line	O
.	O
In	O
this	O
paper	O
,	O
a	O
new	O
and	O
fast	O
method	O
for	O
reducing	O
SPA	O
is	O
presented	O
.	O
For	O
this	O
purpose	O
,	O
the	O
standing	O
phase	O
angle	O
difference	O
between	O
two	O
specific	O
buses	O
is	O
represented	O
in	O
terms	O
of	O
sensitivity	B-KEY
factors	I-KEY
associated	O
with	O
the	O
change	O
in	O
active	O
power	O
generations	O
and	O
consumption	O
at	O
the	O
buses	O
.	O
Then	O
,	O
the	O
proposed	O
method	O
reschedule	O
generation	O
of	O
selected	O
units	O
or	O
shed	O
load	O
of	O
selected	O
buses	O
to	O
reduce	O
excessive	O
SPA	O
difference	O
between	O
two	O
buses	O
based	O
on	O
sensitivity	B-KEY
factors	I-KEY

IT	B-KEY
security	I-KEY
issues	O
:	O
the	O
need	O
for	O
end	B-KEY
user	I-KEY
oriented	I-KEY
research	I-KEY
Considerable	O
attention	O
has	O
been	O
given	O
to	O
the	O
technical	O
and	O
policy	O
issues	O
involved	O
with	O
IT	B-KEY
security	I-KEY
issues	O
in	O
recent	O
years	O
.	O
The	O
growth	O
of	O
e-commerce	B-KEY
and	O
the	O
Internet	B-KEY
,	O
as	O
well	O
as	O
widely	O
publicized	O
hacker	B-KEY
attacks	I-KEY
,	O
have	O
brought	O
IT	B-KEY
security	I-KEY
into	O
prominent	O
focus	O
and	O
routine	O
corporate	O
attention	O
.	O
Yet	O
,	O
much	O
more	O
research	O
is	O
needed	O
from	O
the	O
end	O
user	O
-LRB-	O
EU	O
-RRB-	O
perspective	O
.	O
This	O
position	O
paper	O
is	O
a	O
call	O
for	O
such	O
research	O
and	O
outlines	O
some	O
possible	O
directions	O
of	O
interest	O

Basin	B-KEY
configuration	I-KEY
of	O
a	O
six-dimensional	B-KEY
model	I-KEY
of	O
an	O
electric	B-KEY
power	I-KEY
system	I-KEY
As	O
part	O
of	O
an	O
ongoing	O
project	O
on	O
the	O
stability	O
of	O
massively	O
complex	O
electrical	B-KEY
power	I-KEY
systems	I-KEY
,	O
we	O
discuss	O
the	O
global	O
geometric	O
structure	O
of	O
contacts	O
among	O
the	O
basins	O
of	O
attraction	O
of	O
a	O
six-dimensional	O
dynamical	O
system	O
.	O
This	O
system	O
represents	O
a	O
simple	O
model	O
of	O
an	O
electrical	B-KEY
power	I-KEY
system	I-KEY
involving	O
three	O
machines	O
and	O
an	O
infinite	B-KEY
bus	I-KEY
.	O
Apart	O
from	O
the	O
possible	O
occurrence	O
of	O
attractors	B-KEY
representing	O
pathological	B-KEY
states	I-KEY
,	O
the	O
contacts	O
between	O
the	O
basins	O
have	O
a	O
practical	O
importance	O
,	O
from	O
the	O
point	O
of	O
view	O
of	O
the	O
operation	O
of	O
a	O
real	O
electrical	B-KEY
power	I-KEY
system	I-KEY
.	O
With	O
the	O
aid	O
of	O
a	O
global	B-KEY
map	I-KEY
of	O
basins	O
,	O
one	O
could	O
hope	O
to	O
design	O
an	O
intervention	O
strategy	O
to	O
boot	O
the	O
power	O
system	O
back	O
into	O
its	O
normal	O
state	O
.	O
Our	O
method	O
involves	O
taking	O
two-dimensional	O
sections	O
of	O
the	O
six-dimensional	O
state	B-KEY
space	I-KEY
,	O
and	O
then	O
determining	O
the	O
basins	O
directly	O
by	O
numerical	O
simulation	O
from	O
a	O
dense	O
grid	O
of	O
initial	O
conditions	O
.	O
The	O
relations	O
among	O
all	O
the	O
basins	O
are	O
given	O
for	O
a	O
specific	O
numerical	O
example	O
,	O
that	O
is	O
,	O
choosing	O
particular	O
values	O
for	O
the	O
parameters	O
in	O
our	O
model	O

Application-layer	B-KEY
multicasting	I-KEY
with	O
Delaunay	B-KEY
triangulation	I-KEY
overlays	I-KEY
Application-layer	B-KEY
multicast	I-KEY
supports	O
group	B-KEY
applications	I-KEY
without	O
the	O
need	O
for	O
a	O
network-layer	B-KEY
multicast	I-KEY
protocol	I-KEY
.	O
Here	O
,	O
applications	O
arrange	O
themselves	O
in	O
a	O
logical	B-KEY
overlay	I-KEY
network	I-KEY
and	O
transfer	O
data	O
within	O
the	O
overlay	O
.	O
We	O
present	O
an	O
application-layer	B-KEY
multicast	I-KEY
solution	O
that	O
uses	O
a	O
Delaunay	O
triangulation	O
as	O
an	O
overlay	B-KEY
network	I-KEY
topology	O
.	O
An	O
advantage	O
of	O
using	O
a	O
Delaunay	O
triangulation	O
is	O
that	O
it	O
allows	O
each	O
application	O
to	O
locally	O
derive	O
next-hop	B-KEY
routing	I-KEY
information	I-KEY
without	O
requiring	O
a	O
routing	O
protocol	O
in	O
the	O
overlay	O
.	O
A	O
disadvantage	O
of	O
using	O
a	O
Delaunay	O
triangulation	O
is	O
that	O
the	O
mapping	O
of	O
the	O
overlay	O
to	O
the	O
network	O
topology	O
at	O
the	O
network	O
and	O
data	B-KEY
link	I-KEY
layer	I-KEY
may	O
be	O
suboptimal	O
.	O
We	O
present	O
a	O
protocol	O
,	O
called	O
Delaunay	O
triangulation	O
-LRB-	O
DT	B-KEY
protocol	I-KEY
-RRB-	O
,	O
which	O
constructs	O
Delaunay	B-KEY
triangulation	I-KEY
overlay	I-KEY
networks	O
.	O
We	O
present	O
measurement	B-KEY
experiments	I-KEY
of	O
the	O
DT	B-KEY
protocol	I-KEY
for	O
overlay	B-KEY
networks	I-KEY
with	O
up	O
to	O
10	O
000	O
members	O
,	O
that	O
are	O
running	O
on	O
a	O
local	B-KEY
PC	I-KEY
cluster	I-KEY
with	O
100	O
Linux	B-KEY
PCs	I-KEY
.	O
The	O
results	O
show	O
that	O
the	O
protocol	O
stabilizes	O
quickly	O
,	O
e.g.	O
,	O
an	O
overlay	B-KEY
network	I-KEY
with	O
10	O
000	O
nodes	O
can	O
be	O
built	O
in	O
just	O
over	O
30	O
s	O
.	O
The	O
traffic	B-KEY
measurements	I-KEY
indicate	O
that	O
the	O
average	B-KEY
overhead	I-KEY
of	O
a	O
node	O
is	O
only	O
a	O
few	O
kilobits	O
per	O
second	O
if	O
the	O
overlay	B-KEY
network	I-KEY
is	O
in	O
a	O
steady	O
state	O
.	O
Results	O
of	O
throughput	B-KEY
experiments	I-KEY
of	O
multicast	B-KEY
transmissions	I-KEY
-LRB-	O
using	O
TCP	B-KEY
unicast	I-KEY
connections	I-KEY
between	O
neighbors	O
in	O
the	O
overlay	B-KEY
network	I-KEY
-RRB-	O
show	O
an	O
achievable	O
throughput	O
of	O
approximately	O
15	O
Mb/s	O
in	O
an	O
overlay	O
with	O
100	O
nodes	O
and	O
2	O
Mb/s	O
in	O
an	O
overlay	O
with	O
1000	O
nodes	O

Achieving	O
performance	O
under	O
OpenMP	B-KEY
on	O
ccNUMA	O
and	O
software	O
distributed	O
shared	O
memory	O
systems	O
OpenMP	B-KEY
is	O
emerging	O
as	O
a	O
viable	O
high-level	O
programming	B-KEY
model	I-KEY
for	O
shared	B-KEY
memory	I-KEY
parallel	I-KEY
systems	I-KEY
.	O
It	O
was	O
conceived	O
to	O
enable	O
easy	O
,	O
portable	O
application	O
development	O
on	O
this	O
range	O
of	O
systems	O
,	O
and	O
it	O
has	O
also	O
been	O
implemented	O
on	O
cache-coherent	B-KEY
Non-Uniform	I-KEY
Memory	I-KEY
Access	I-KEY
-LRB-	O
ccNUMA	O
-RRB-	O
architectures	O
.	O
Unfortunately	O
,	O
it	O
is	O
hard	O
to	O
obtain	O
high	O
performance	O
on	O
the	O
latter	O
architecture	O
,	O
particularly	O
when	O
large	O
numbers	O
of	O
threads	O
are	O
involved	O
.	O
In	O
this	O
paper	O
,	O
we	O
discuss	O
the	O
difficulties	O
faced	O
when	O
writing	O
OpenMP	B-KEY
programs	O
for	O
ccNUMA	O
systems	O
,	O
and	O
explain	O
how	O
the	O
vendors	O
have	O
attempted	O
to	O
overcome	O
them	O
.	O
We	O
focus	O
on	O
one	O
such	O
system	O
,	O
the	O
SGI	O
Origin	O
2000	O
,	O
and	O
perform	O
a	O
variety	O
of	O
experiments	O
designed	O
to	O
illustrate	O
the	O
impact	O
of	O
the	O
vendor	O
's	O
efforts	O
.	O
We	O
compare	O
codes	O
written	O
in	O
a	O
standard	O
,	O
loop-level	O
parallel	O
style	O
under	O
OpenMP	B-KEY
with	O
alternative	O
versions	O
written	O
in	O
a	O
Single	B-KEY
Program	I-KEY
Multiple	I-KEY
Data	I-KEY
-LRB-	O
SPMD	O
-RRB-	O
fashion	O
,	O
also	O
realized	O
via	O
OpenMP	B-KEY
,	O
and	O
show	O
that	O
the	O
latter	O
consistently	O
provides	O
superior	O
performance	O
.	O
A	O
carefully	O
chosen	O
set	O
of	O
language	O
extensions	O
can	O
help	O
us	O
translate	O
programs	O
from	O
the	O
former	O
style	O
to	O
the	O
latter	O
-LRB-	O
or	O
to	O
compile	O
directly	O
,	O
but	O
in	O
a	O
similar	O
manner	O
-RRB-	O
.	O
Syntax	O
for	O
these	O
extensions	O
can	O
be	O
borrowed	O
from	O
HPF	B-KEY
,	O
and	O
some	O
aspects	O
of	O
HPF	B-KEY
compiler	O
technology	O
can	O
help	O
the	O
translation	O
process	O
.	O
It	O
is	O
our	O
expectation	O
that	O
an	O
extended	O
language	O
,	O
if	O
well	O
compiled	O
,	O
would	O
improve	O
the	O
attractiveness	O
of	O
OpenMP	B-KEY
as	O
a	O
language	O
for	O
high-performance	O
computation	O
on	O
an	O
important	O
class	O
of	O
modern	O
architectures	O

Embeddings	B-KEY
of	O
planar	B-KEY
graphs	I-KEY
that	O
minimize	O
the	O
number	O
of	O
long-face	B-KEY
cycles	I-KEY
We	O
consider	O
the	O
problem	O
of	O
finding	O
embeddings	B-KEY
of	O
planar	B-KEY
graphs	I-KEY
that	O
minimize	O
the	O
number	O
of	O
long-face	B-KEY
cycles	I-KEY
.	O
We	O
prove	O
that	O
for	O
any	O
k	O
>	O
or	O
=	O
4	O
,	O
it	O
is	O
NP-complete	O
to	O
find	O
an	O
embedding	B-KEY
that	O
minimizes	O
the	O
number	O
of	O
face	O
cycles	O
of	O
length	O
at	O
least	O
k	O

Design	B-KEY
methodology	I-KEY
for	O
diagnostic	B-KEY
strategies	I-KEY
for	O
industrial	B-KEY
systems	I-KEY
This	O
paper	O
presents	O
a	O
method	O
for	O
the	O
construction	O
of	O
diagnostic	O
systems	O
for	O
complex	O
industrial	O
applications	O
.	O
The	O
approach	O
has	O
been	O
explicitely	O
developed	O
to	O
shorten	O
the	O
design	O
cycle	O
and	O
meet	O
some	O
specific	O
requirements	O
,	O
such	O
as	O
modularity	B-KEY
,	O
flexibility	O
,	O
and	O
the	O
possibility	O
of	O
merging	O
many	O
different	O
sources	O
of	O
information	O
.	O
The	O
method	O
allows	O
one	O
to	O
consider	O
multiple	O
simultaneous	O
failures	O
and	O
is	O
specifically	O
designed	O
to	O
make	O
easier	O
the	O
coordination	O
and	O
simplification	O
of	O
local	B-KEY
diagnostic	I-KEY
algorithms	I-KEY
developed	O
by	O
different	O
teams	O

Use	O
of	O
natural	B-KEY
language	I-KEY
processing	I-KEY
to	O
translate	O
clinical	O
information	O
from	O
a	O
database	O
of	O
889,921	O
chest	O
radiographic	O
reports	O
The	O
aim	O
was	O
to	O
evaluate	O
translation	O
of	O
chest	O
radiographic	O
reports	O
using	O
natural	B-KEY
language	I-KEY
processing	I-KEY
and	O
to	O
compare	O
the	O
findings	O
with	O
those	O
in	O
the	O
literature	O
.	O
A	O
natural	O
language	O
processor	O
coded	O
10	O
years	O
of	O
narrative	O
chest	O
radiographic	O
reports	O
from	O
an	O
urban	B-KEY
academic	I-KEY
medical	I-KEY
center	I-KEY
.	O
Coding	O
for	O
150	O
reports	O
was	O
compared	O
with	O
manual	O
coding	O
.	O
Frequencies	O
and	O
cooccurrences	O
of	O
24	O
clinical	O
conditions	O
-LRB-	O
diseases	O
,	O
abnormalities	O
,	O
and	O
clinical	O
states	O
-RRB-	O
were	O
estimated	O
.	O
The	O
ratio	O
of	O
right	O
to	O
left	O
lung	O
mass	O
,	O
association	O
of	O
pleural	B-KEY
effusion	I-KEY
with	O
other	O
conditions	O
,	O
and	O
frequency	O
of	O
bullet	O
and	O
stab	B-KEY
wounds	I-KEY
were	O
compared	O
with	O
independent	O
observations	O
.	O
The	O
sensitivity	O
and	O
specificity	O
of	O
the	O
system	O
's	O
pneumothorax	B-KEY
coding	I-KEY
were	O
compared	O
with	O
those	O
of	O
manual	O
financial	O
coding	O
.	O
Internal	O
and	O
external	O
validation	O
in	O
this	O
study	O
confirmed	O
the	O
accuracy	O
of	O
natural	B-KEY
language	I-KEY
processing	I-KEY
for	O
translating	O
chest	O
radiographic	O
narrative	O
reports	O
into	O
a	O
large	O
database	O
of	O
information	O

Image	O
reconstruction	O
from	O
fan-beam	B-KEY
projections	I-KEY
on	O
less	O
than	O
a	O
short	O
scan	O
This	O
work	O
is	O
concerned	O
with	O
2D	B-KEY
image	I-KEY
reconstruction	I-KEY
from	O
fan-beam	B-KEY
projections	I-KEY
.	O
It	O
is	O
shown	O
that	O
exact	O
and	O
stable	O
reconstruction	O
of	O
a	O
given	O
region-of-interest	B-KEY
in	O
the	O
object	O
does	O
not	O
require	O
all	O
lines	O
passing	O
through	O
the	O
object	O
to	O
be	O
measured	O
.	O
Complete	O
-LRB-	O
non-truncated	O
-RRB-	O
fan-beam	B-KEY
projections	I-KEY
provide	O
sufficient	O
information	O
for	O
reconstruction	O
when	O
`	O
every	O
line	O
passing	O
through	O
the	O
region-of-interest	B-KEY
intersects	O
the	O
vertex	B-KEY
path	I-KEY
in	O
a	O
non-tangential	O
way	O
'	O
.	O
The	O
practical	O
implications	O
of	O
this	O
condition	O
are	O
discussed	O
and	O
a	O
new	O
filtered-backprojection	B-KEY
algorithm	I-KEY
is	O
derived	O
for	O
reconstruction	O
.	O
Experiments	O
with	O
computer-simulated	O
data	O
are	O
performed	O
to	O
support	O
the	O
mathematical	O
results	O

Super	O
high	O
definition	O
image	O
-LRB-	O
WHD	O
:	O
Wide/Double	O
HD	O
-RRB-	O
transmission	O
system	O
This	O
paper	O
describes	O
a	O
WHD	B-KEY
image	I-KEY
transmission	I-KEY
system	I-KEY
constructed	O
from	O
a	O
display	O
projector	O
,	O
CODECs	B-KEY
,	O
and	O
a	O
camera	B-KEY
system	I-KEY
imaging	I-KEY
a	O
super	O
high	O
definition	O
image	O
-LRB-	O
WHD	O
:	O
Wide/Double	O
HD	O
-RRB-	O
corresponding	O
to	O
two	O
screen	O
portions	O
of	O
common	O
high-vision	O
images	O
.	O
This	O
system	O
was	O
developed	O
as	O
a	O
transmission	O
system	O
to	O
communicate	O
with	O
or	O
transmit	O
information	O
giving	O
a	O
reality-enhanced	B-KEY
feeling	I-KEY
to	O
a	O
remote	O
location	O
by	O
using	O
images	O
of	O
super	O
high	O
definition	O
.	O
In	O
addition	O
,	O
the	O
correction	O
processing	O
for	O
the	O
distortions	O
of	O
images	O
occurring	O
due	O
to	O
the	O
structure	O
of	O
the	O
camera	O
system	O
,	O
an	O
outline	O
of	O
the	O
transmission	O
experiments	O
using	O
the	O
proposed	O
system	O
,	O
and	O
subjective	O
evaluation	O
experiments	O
using	O
WHD	O
images	O
are	O
presented	O

Taiwan	B-KEY
power	I-KEY
company	I-KEY
phases	O
into	O
AM/FM	B-KEY
To	O
face	O
the	O
challenges	O
and	O
impact	O
of	O
the	O
inevitable	O
trend	O
toward	O
privatization	B-KEY
and	O
deregulation	B-KEY
,	O
the	O
Taiwan	O
Power	O
Co.	O
-LRB-	O
TPC	O
-RRB-	O
devised	O
short	O
-	O
and	O
long-term	O
strategic	O
computerization	O
development	O
plans	O
.	O
These	O
development	O
efforts	O
created	O
a	O
master	O
plan	O
that	O
included	O
building	O
an	O
Automated	B-KEY
Mapping	I-KEY
and	I-KEY
Facilities	I-KEY
Management	I-KEY
-LRB-	O
AM	B-KEY
/	I-KEY
FM	I-KEY
-RRB-	O
system	O
for	O
Taipei	B-KEY
City	I-KEY
District	I-KEY
Office	I-KEY
-LRB-	O
TCDO	O
-RRB-	O
.	O
This	O
project	O
included	O
a	O
pilot	B-KEY
project	I-KEY
followed	O
by	O
evaluation	O
before	O
the	O
roll	O
out	O
to	O
the	O
complete	B-KEY
service	I-KEY
territory	I-KEY
of	O
TCDO	O
.	O
The	O
pilot	B-KEY
project	I-KEY
took	O
three	O
years	O
to	O
install	O
,	O
commission	O
and-via	O
the	O
evaluation	O
process-reach	O
the	O
conclusion	O
that	O
AM/FM	B-KEY
was	O
technologically	O
feasible	O

Verona	B-KEY
Lastre	I-KEY
:	O
consolidation	O
provides	O
opening	O
for	O
a	O
new	O
plate	O
vendor	O
Fewer	O
companies	O
than	O
ever	O
are	O
manufacturing	O
CTP	B-KEY
plates	I-KEY
.	O
The	O
market	O
has	O
become	O
globalized	O
,	O
with	O
just	O
four	O
big	O
firms	O
dominating	O
the	O
picture	O
.	O
To	O
the	O
Samor	O
Group	O
,	O
however	O
,	O
globalization	O
looked	O
like	O
an	O
opportunity	O
;	O
it	O
reasoned	O
that	O
many	O
a	O
national	O
and	O
local	O
distributor	O
would	O
welcome	O
a	O
small	O
,	O
competitive	O
,	O
regional	O
manufacturer	O
.	O
A	O
couple	O
of	O
years	O
ago	O
it	O
formed	O
a	O
company	O
,	O
Verona	B-KEY
Lastre	I-KEY
,	O
to	O
exploit	O
that	O
opportunity	O
.	O
Now	O
Vela	B-KEY
,	O
as	O
it	O
's	O
familiarly	O
called	O
,	O
has	O
launched	O
its	O
line	O
of	O
high-quality	O
thermal	O
plates	O
and	O
is	O
busily	O
lining	O
up	O
dealers	O
in	O
Europe	O
and	O
the	O
Americas	O

Local	O
satellite	O
Consumer	O
based	O
mobile	B-KEY
satellite	I-KEY
phone	I-KEY
services	I-KEY
went	O
from	O
boom	O
to	O
burn	O
up	O
in	O
twelve	O
months	O
despite	O
original	O
forecasts	O
predicting	O
10	O
million	O
to	O
40	O
million	O
users	O
by	O
2005	O
.	O
Julian	O
Bright	O
wonders	O
what	O
prospects	O
the	O
technology	O
has	O
now	O
and	O
if	O
going	O
regional	O
might	O
be	O
one	O
answer	O

Integrated	B-KEY
optical	I-KEY
metrology	I-KEY
controls	O
post	O
etch	O
CDs	O
Control	O
of	O
the	O
transistor	B-KEY
gate	I-KEY
critical	I-KEY
dimension	I-KEY
-LRB-	O
CD	O
-RRB-	O
on	O
the	O
order	O
of	O
a	O
few	O
nanometers	O
is	O
a	O
top	O
priority	O
in	O
many	O
advanced	O
IC	B-KEY
fabs	I-KEY
.	O
Each	O
nanometer	O
deviation	O
from	O
the	O
target	B-KEY
gate	I-KEY
length	I-KEY
translates	O
directly	O
into	O
the	O
operational	B-KEY
speed	I-KEY
of	O
these	O
devices	O
.	O
However	O
,	O
using	O
in-line	B-KEY
process	I-KEY
control	I-KEY
by	O
linking	O
the	O
lithography	O
and	O
etch	O
tools	O
can	O
improve	O
CD	B-KEY
performance	I-KEY
beyond	O
what	O
each	O
individual	O
tool	O
can	O
achieve	O
.	O
The	O
integration	O
of	O
optical	B-KEY
CD	I-KEY
metrology	I-KEY
tools	I-KEY
to	O
etch	B-KEY
mainframes	I-KEY
can	O
result	O
in	O
excellent	O
etcher	B-KEY
stability	I-KEY
and	O
better	O
control	O
of	O
post-etch	O
CDs	O

A	O
scalable	O
and	O
lightweight	O
QoS	B-KEY
monitoring	I-KEY
technique	O
combining	O
passive	O
and	O
active	O
approaches	O
:	O
on	O
the	O
mathematical	O
formulation	O
of	O
CoMPACT	B-KEY
monitor	I-KEY
To	O
make	O
a	O
scalable	O
and	O
lightweight	O
QoS	B-KEY
monitoring	I-KEY
system	O
,	O
we	O
-LRB-	O
2002	O
-RRB-	O
have	O
proposed	O
a	O
new	O
QoS	B-KEY
monitoring	I-KEY
technique	O
,	O
called	O
the	O
change-of-measure	B-KEY
based	O
passive/active	O
monitoring	O
-LRB-	O
CoMPACT	B-KEY
Monitor	I-KEY
-RRB-	O
,	O
which	O
is	O
based	O
on	O
the	O
change-of-measure	B-KEY
framework	O
and	O
is	O
an	O
active	O
measurement	O
transformed	O
by	O
using	O
passively	B-KEY
monitored	I-KEY
data	O
.	O
This	O
technique	O
enables	O
us	O
to	O
measure	O
detailed	O
QoS	O
information	O
for	O
individual	O
users	O
,	O
applications	O
and	O
organizations	O
,	O
in	O
a	O
scalable	O
and	O
lightweight	O
manner	O
.	O
In	O
this	O
paper	O
,	O
we	O
present	O
the	O
mathematical	O
foundation	O
of	O
CoMPACT	B-KEY
Monitor	I-KEY
.	O
In	O
addition	O
,	O
we	O
show	O
its	O
characteristics	O
through	O
simulations	O
in	O
terms	O
of	O
typical	O
implementation	O
issues	O
for	O
inferring	O
the	O
delay	B-KEY
distributions	I-KEY
.	O
The	O
results	O
show	O
that	O
CoMPACT	B-KEY
Monitor	I-KEY
gives	O
accurate	O
QoS	O
estimations	O
with	O
only	O
a	O
small	O
amount	O
of	O
extra	O
traffic	O
for	O
active	O
measurement	O

A	O
min-max	B-KEY
theorem	I-KEY
on	O
feedback	B-KEY
vertex	I-KEY
sets	I-KEY
We	O
establish	O
a	O
necessary	O
and	O
sufficient	O
condition	O
for	O
the	O
linear	B-KEY
system	I-KEY
-LCB-	O
x	O
:	O
Hx	O
>	O
or	O
=	O
e	O
,	O
x	O
>	O
or	O
=	O
0	O
-RCB-	O
associated	O
with	O
a	O
bipartite	B-KEY
tournament	I-KEY
to	O
be	O
totally	O
dual	O
integral	O
,	O
where	O
H	O
is	O
the	O
cycle-vertex	B-KEY
incidence	I-KEY
matrix	I-KEY
and	O
e	O
is	O
the	O
all-one	B-KEY
vector	I-KEY
.	O
The	O
consequence	O
is	O
a	O
min-max	O
relation	O
on	O
packing	O
and	O
covering	B-KEY
cycles	I-KEY
,	O
together	O
with	O
strongly	B-KEY
polynomial	I-KEY
time	I-KEY
algorithms	I-KEY
for	O
the	O
feedback	B-KEY
vertex	I-KEY
set	I-KEY
problem	O
and	O
the	O
cycle	O
packing	O
problem	O
on	O
the	O
corresponding	O
bipartite	O
tournaments	O
.	O
In	O
addition	O
,	O
we	O
show	O
that	O
the	O
feedback	B-KEY
vertex	I-KEY
set	I-KEY
problem	O
on	O
general	O
bipartite	O
tournaments	O
is	O
NP-complete	O
and	O
approximable	O
within	O
3.5	O
based	O
on	O
the	O
min-max	O
theorem	O

FinancialContent	O
.	O
Credibility	O
is	O
king	O
If	O
you	O
went	O
to	O
a	O
site	O
named	O
Financialcontent.com	B-KEY
,	O
you	O
'd	O
probably	O
expect	O
to	O
find	O
,	O
well	O
,	O
financial	O
content	O
.	O
Maybe	O
stock	O
prices	O
or	O
company	O
earnings	O
or	O
market	O
charts	O
or	O
economic	O
statistics	O
or	O
corporate	O
news	O
reports	O
.	O
Well	O
,	O
you	O
'd	O
be	O
partially	O
correct	O
.	O
Financialcontent.com	B-KEY
does	O
deal	O
in	O
financial	B-KEY
information	I-KEY
,	O
but	O
its	O
main	O
objective	O
is	O
not	O
to	O
distribute	O
its	O
financial	O
content	O
to	O
individual	O
investors	O
,	O
but	O
to	O
distribute	O
it	O
through	O
other	O
Web	B-KEY
sites	I-KEY
.	O
In	O
other	O
words	O
,	O
FinancialContent	O
is	O
a	O
wholesaler	O
,	O
not	O
a	O
retailer	O
.	O
As	O
an	O
aggregator	B-KEY
,	O
FinancialContent	O
provides	O
partner	B-KEY
sites	I-KEY
with	O
financial	B-KEY
information	I-KEY
that	O
is	O
tailored	O
to	O
that	O
individual	O
Web	B-KEY
site	I-KEY

Inhibiting	O
decoherence	B-KEY
via	O
ancilla	B-KEY
processes	I-KEY
General	B-KEY
conditions	I-KEY
are	O
derived	O
for	O
preventing	O
the	O
decoherence	B-KEY
of	O
a	O
single	B-KEY
two-state	I-KEY
quantum	I-KEY
system	I-KEY
-LRB-	O
qubit	B-KEY
-RRB-	O
in	O
a	O
thermal	B-KEY
bath	I-KEY
.	O
The	O
employed	O
auxiliary	B-KEY
systems	I-KEY
required	O
for	O
this	O
purpose	O
are	O
merely	O
assumed	O
to	O
be	O
weak	O
for	O
the	O
general	B-KEY
condition	I-KEY
while	O
various	O
examples	O
such	O
as	O
extra	B-KEY
qubits	I-KEY
and	O
extra	O
classical	O
fields	O
are	O
studied	O
for	O
applications	O
in	O
quantum	O
information	O
processing	O
.	O
The	O
general	B-KEY
condition	I-KEY
is	O
confirmed	O
by	O
well	O
known	O
approaches	O
toward	O
inhibiting	O
decoherence	B-KEY
.	O
An	O
approach	O
to	O
decoherence-free	B-KEY
quantum	I-KEY
memories	I-KEY
and	O
quantum	B-KEY
operations	I-KEY
is	O
presented	O
by	O
placing	O
the	O
qubit	B-KEY
into	O
the	O
center	O
of	O
a	O
sphere	O
with	O
extra	B-KEY
qubits	I-KEY
on	O
its	O
surface	O

Micro-optical	O
realization	O
of	O
arrays	O
of	O
selectively	O
addressable	O
dipole	O
traps	O
:	O
a	O
scalable	B-KEY
configuration	I-KEY
for	O
quantum	B-KEY
computation	I-KEY
with	O
atomic	B-KEY
qubits	I-KEY
We	O
experimentally	O
demonstrate	O
novel	O
structures	O
for	O
the	O
realization	O
of	O
registers	B-KEY
of	O
atomic	B-KEY
qubits	I-KEY
:	O
We	O
trap	O
neutral	B-KEY
atoms	I-KEY
in	O
one	O
-	O
and	O
two-dimensional	O
arrays	O
of	O
far-detuned	B-KEY
dipole	I-KEY
traps	I-KEY
obtained	O
by	O
focusing	O
a	O
red-detuned	B-KEY
laser	I-KEY
beam	I-KEY
with	O
a	O
microfabricated	B-KEY
array	I-KEY
of	O
microlenses	B-KEY
.	O
We	O
are	O
able	O
to	O
selectively	O
address	O
individual	O
trap	O
sites	O
due	O
to	O
their	O
large	O
lateral	O
separation	O
of	O
125	O
mu	O
m	O
.	O
We	O
initialize	O
and	O
read	O
out	O
different	O
internal	B-KEY
states	I-KEY
for	O
the	O
individual	O
sites	O
.	O
We	O
also	O
create	O
two	O
interleaved	O
sets	O
of	O
trap	O
arrays	O
with	O
adjustable	O
separation	O
,	O
as	O
required	O
for	O
many	O
proposed	O
implementations	O
of	O
quantum	B-KEY
gate	I-KEY
operations	I-KEY

A	O
partial	B-KEY
converse	I-KEY
to	O
Hadamard	O
's	O
theorem	O
on	O
homeomorphisms	B-KEY
A	O
theorem	O
by	O
Hadamard	O
gives	O
a	O
two-part	O
condition	O
under	O
which	O
a	O
map	O
from	O
one	O
Banach	B-KEY
space	I-KEY
to	O
another	O
is	O
a	O
homeomorphism	B-KEY
.	O
The	O
theorem	O
,	O
while	O
often	O
very	O
useful	O
,	O
is	O
incomplete	O
in	O
the	O
sense	O
that	O
it	O
does	O
not	O
explicitly	O
specify	O
the	O
family	O
of	O
maps	O
for	O
which	O
the	O
condition	O
is	O
met	O
.	O
Here	O
,	O
under	O
a	O
typically	O
weak	O
additional	O
assumption	O
on	O
the	O
map	O
,	O
we	O
show	O
that	O
Hadamard	O
's	O
condition	O
is	O
met	O
if	O
,	O
and	O
only	O
if	O
,	O
the	O
map	O
is	O
a	O
homeomorphism	B-KEY
with	O
a	O
Lipschitz	B-KEY
continuous	I-KEY
inverse	I-KEY
.	O
An	O
application	O
is	O
given	O
concerning	O
the	O
relation	O
between	O
the	O
stability	O
of	O
a	O
nonlinear	O
system	O
and	O
the	O
stability	O
of	O
related	O
linear	B-KEY
systems	O

Playing	O
for	O
time	O
-LSB-	O
3G	B-KEY
networks	I-KEY
-RSB-	O
The	O
delays	B-KEY
in	O
rolling	O
out	O
3G	B-KEY
networks	I-KEY
across	O
Europe	B-KEY
should	O
not	O
always	O
be	O
seen	O
with	O
a	O
negative	O
slant	O

Knowledge	B-KEY
acquisition	I-KEY
for	O
expert	B-KEY
systems	I-KEY
in	O
accounting	B-KEY
and	O
financial	O
problem	O
domains	O
Since	O
the	O
mid-1980s	O
,	O
expert	B-KEY
systems	I-KEY
have	O
been	O
developed	O
for	O
a	O
variety	O
of	O
problems	O
in	O
accounting	B-KEY
and	O
finance	B-KEY
.	O
The	O
most	O
commonly	O
cited	O
problems	O
in	O
developing	O
these	O
systems	O
are	O
the	O
unavailability	O
of	O
the	O
experts	O
and	O
knowledge	O
engineers	O
and	O
difficulties	O
with	O
the	O
rule	B-KEY
extraction	I-KEY
process	I-KEY
.	O
Within	O
the	O
field	O
of	O
artificial	B-KEY
intelligence	I-KEY
,	O
this	O
has	O
been	O
called	O
the	O
`	O
knowledge	B-KEY
acquisition	I-KEY
'	O
-LRB-	O
KA	O
-RRB-	O
problem	O
and	O
has	O
been	O
identified	O
as	O
a	O
major	O
bottleneck	O
in	O
the	O
expert	B-KEY
system	I-KEY
development	O
process	O
.	O
Recent	O
empirical	O
research	O
reveals	O
that	O
certain	O
KA	O
techniques	O
are	O
significantly	O
more	O
efficient	O
than	O
others	O
in	O
helping	O
to	O
extract	O
certain	O
types	O
of	O
knowledge	O
within	O
specific	O
problem	O
domains	O
.	O
This	O
paper	O
presents	O
a	O
mapping	O
between	O
these	O
empirical	O
studies	O
and	O
a	O
generic	O
taxonomy	O
of	O
expert	B-KEY
system	I-KEY
problem	O
domains	O
.	O
To	O
accomplish	O
this	O
,	O
we	O
first	O
examine	O
the	O
range	O
of	O
problem	O
domains	O
and	O
suggest	O
a	O
mapping	O
of	O
accounting	B-KEY
and	O
finance	B-KEY
tasks	O
to	O
a	O
generic	O
problem	B-KEY
domain	I-KEY
taxonomy	I-KEY
.	O
We	O
then	O
identify	O
and	O
describe	O
the	O
most	O
prominent	O
KA	O
techniques	O
employed	O
in	O
developing	O
expert	B-KEY
systems	I-KEY
in	O
accounting	B-KEY
and	O
finance	B-KEY
.	O
After	O
examining	O
and	O
summarizing	O
the	O
existing	O
empirical	O
KA	O
work	O
,	O
we	O
conclude	O
by	O
showing	O
how	O
the	O
empirical	O
KA	O
research	O
in	O
the	O
various	O
problem	O
domains	O
can	O
be	O
used	O
to	O
provide	O
guidance	O
to	O
developers	O
of	O
expert	B-KEY
systems	I-KEY
in	O
the	O
fields	O
of	O
accounting	B-KEY
and	O
finance	B-KEY

Embedding	O
the	O
outer	O
automorphism	O
group	O
Out	O
-LRB-	O
F/sub	O
n	O
/	O
-RRB-	O
of	O
a	O
free	B-KEY
group	I-KEY
of	O
rank	O
n	O
in	O
the	O
group	O
Out	O
-LRB-	O
F/sub	O
m	O
/	O
-RRB-	O
for	O
m	O
>	O
n	O
It	O
is	O
proved	O
that	O
for	O
every	O
n	O
>	O
or	O
=	O
1	O
,	O
the	O
group	O
Out	O
-LRB-	O
F/sub	O
n	O
/	O
-RRB-	O
is	O
embedded	O
in	O
the	O
group	O
Out	O
-LRB-	O
F/sub	O
m	O
/	O
-RRB-	O
with	O
m	O
=	O
1	O
+	O
-LRB-	O
n	O
-	O
1	O
-RRB-	O
k/sup	O
n	O
/	O
,	O
where	O
k	O
is	O
an	O
arbitrary	B-KEY
natural	I-KEY
number	I-KEY
coprime	I-KEY
to	O
n	O
-	O
1	O

Encouraging	O
women	O
in	O
computer	O
science	O
At	O
a	O
cost	O
to	O
both	O
their	O
own	O
opportunities	O
and	O
society	O
's	O
ability	O
to	O
produce	O
people	O
with	O
much-needed	O
technical	B-KEY
skills	I-KEY
,	O
women	O
continue	O
to	O
be	O
underrepresented	O
in	O
computer	B-KEY
science	I-KEY
degree	I-KEY
programs	I-KEY
at	O
both	O
the	O
undergraduate	O
and	O
graduate	B-KEY
level	I-KEY
.	O
Although	O
some	O
of	O
the	O
barriers	O
that	O
women	O
face	O
have	O
their	O
foundations	O
in	O
cultural	B-KEY
expectations	I-KEY
established	O
well	O
before	O
the	O
college	O
level	O
,	O
we	O
believe	O
that	O
departments	O
can	O
take	O
effective	O
steps	O
to	O
increase	O
recruitment	O
and	O
retention	O
of	O
women	O
students	O
.	O
This	O
paper	O
describes	O
several	O
strategies	O
we	O
have	O
adopted	O
at	O
Stanford	O
over	O
the	O
past	O
decade	O

A	O
meteorological	B-KEY
fuzzy	I-KEY
expert	I-KEY
system	I-KEY
incorporating	O
subjective	B-KEY
user	I-KEY
input	I-KEY
We	O
present	O
a	O
fuzzy	O
expert	O
system	O
,	O
MEDEX	B-KEY
,	O
for	O
forecasting	O
gale-force	O
winds	O
in	O
the	O
Mediterranean	B-KEY
basin	I-KEY
.	O
The	O
most	O
successful	O
local	O
wind	O
forecasting	O
in	O
this	O
region	O
is	O
achieved	O
by	O
an	O
expert	O
human	O
forecaster	O
with	O
access	O
to	O
numerical	B-KEY
weather	I-KEY
prediction	I-KEY
products	I-KEY
.	O
That	O
forecaster	O
's	O
knowledge	O
is	O
expressed	O
as	O
a	O
set	O
of	O
`	O
rules-of-thumb	B-KEY
'	O
.	O
Fuzzy	O
set	O
methodologies	O
have	O
proved	O
well	O
suited	O
for	O
encoding	O
the	O
forecaster	O
's	O
knowledge	O
,	O
and	O
for	O
accommodating	O
the	O
uncertainty	B-KEY
inherent	O
in	O
the	O
specification	O
of	O
rules	O
,	O
as	O
well	O
as	O
in	O
subjective	O
and	O
objective	O
input	O
.	O
MEDEX	B-KEY
uses	O
fuzzy	B-KEY
set	I-KEY
theory	I-KEY
in	O
two	O
ways	O
:	O
as	O
a	O
fuzzy	B-KEY
rule	I-KEY
base	I-KEY
in	O
the	O
expert	O
system	O
,	O
and	O
for	O
fuzzy	B-KEY
pattern	I-KEY
matching	I-KEY
to	O
select	O
dominant	O
wind	B-KEY
circulation	I-KEY
patterns	I-KEY
as	O
one	O
input	O
to	O
the	O
expert	O
system	O
.	O
The	O
system	O
was	O
developed	O
,	O
tuned	O
,	O
and	O
verified	O
over	O
a	O
two-year	O
period	O
,	O
during	O
which	O
the	O
weather	O
conditions	O
from	O
539	O
days	O
were	O
individually	O
analyzed	O
.	O
Evaluations	O
of	O
MEDEX	B-KEY
performance	O
for	O
both	O
the	O
onset	O
and	O
cessation	O
of	O
winter	O
and	O
summer	O
winds	O
are	O
presented	O
,	O
and	O
demonstrate	O
that	O
MEDEX	B-KEY
has	O
forecasting	O
skill	O
competitive	O
with	O
the	O
US	O
Navy	O
's	O
regional	O
forecasting	O
center	O
in	O
Rota	O
,	O
Spain	O

Simple	O
nonlinear	B-KEY
dual-window	I-KEY
operator	I-KEY
for	O
edge	B-KEY
detection	I-KEY
We	O
propose	O
a	O
nonlinear	O
edge	B-KEY
detection	I-KEY
technique	O
based	O
on	O
a	O
two-concentric-circular-window	O
operator	O
.	O
We	O
perform	O
a	O
preliminary	O
selection	O
of	O
edge	O
candidates	O
using	O
a	O
standard	B-KEY
gradient	I-KEY
and	O
use	O
the	O
dual-window	O
operator	O
to	O
reveal	O
edges	O
as	O
zero-crossing	B-KEY
points	I-KEY
of	O
a	O
simple	O
difference	B-KEY
function	I-KEY
depending	O
only	O
on	O
the	O
minimum	O
and	O
maximum	B-KEY
values	I-KEY
in	O
the	O
two	O
windows	O
.	O
Comparisons	O
with	O
other	O
well-established	O
techniques	O
are	O
reported	O
in	O
terms	O
of	O
visual	O
appearance	O
and	O
computational	B-KEY
efficiency	I-KEY
.	O
They	O
show	O
that	O
detected	B-KEY
edges	I-KEY
are	O
surely	O
comparable	O
with	O
Canny	O
's	O
and	O
Laplacian	O
of	O
Gaussian	B-KEY
algorithms	I-KEY
,	O
with	O
a	O
noteworthy	O
reduction	O
in	O
terms	O
of	O
computational	B-KEY
load	I-KEY

CRM	B-KEY
:	O
approaching	O
zenith	O
Looks	O
at	O
how	O
manufacturers	B-KEY
are	O
starting	O
to	O
warm	O
up	O
to	O
the	O
concept	O
of	O
customer	B-KEY
relationship	I-KEY
management	I-KEY
.	O
CRM	B-KEY
has	O
matured	O
into	O
what	O
is	O
expected	O
to	O
be	O
big	O
business	O
.	O
As	O
CRM	B-KEY
software	O
evolves	O
to	O
its	O
second	O
,	O
some	O
say	O
third	O
,	O
generation	O
,	O
it	O
's	O
likely	O
to	O
be	O
more	O
valuable	O
to	O
holdouts	O
in	O
manufacturing	B-KEY
and	O
other	O
sectors	O

Failures	O
and	O
successes	O
:	O
notes	O
on	O
the	O
development	O
of	O
electronic	B-KEY
cash	I-KEY
Between	O
1997	O
and	O
2001	O
,	O
two	O
mid-sized	O
communities	O
in	O
Canada	B-KEY
hosted	O
North	O
America	O
's	O
most	O
comprehensive	O
experiment	O
to	O
introduce	O
electronic	B-KEY
cash	I-KEY
and	O
,	O
in	O
the	O
process	O
,	O
replace	O
physical	O
cash	O
for	O
casual	O
,	O
low-value	B-KEY
payments	I-KEY
.	O
The	O
technology	O
used	O
was	O
Mondex	B-KEY
,	O
and	O
its	O
implementation	O
was	O
supported	O
by	O
all	O
the	O
country	O
's	O
major	B-KEY
banks	I-KEY
.	O
It	O
was	O
launched	O
with	O
an	O
extensive	O
publicity	B-KEY
campaign	I-KEY
to	O
promote	O
Mondex	B-KEY
not	O
only	O
in	O
the	O
domestic	O
but	O
also	O
in	O
the	O
global	B-KEY
market	I-KEY
,	O
for	O
which	O
the	O
Canadian	B-KEY
implementation	I-KEY
was	O
to	O
serve	O
as	O
a	O
``	O
showcase	O
.	O
''	O
However	O
,	O
soon	O
after	O
the	O
start	O
of	O
the	O
first	O
field	O
test	O
it	O
became	O
apparent	O
that	O
the	O
new	O
technology	O
did	O
not	O
work	O
smoothly	O
.	O
On	O
the	O
contrary	O
,	O
it	O
created	O
a	O
host	O
of	O
controversies	O
,	O
in	O
areas	O
as	O
varied	O
as	O
computer	B-KEY
security	I-KEY
,	O
consumer	B-KEY
privacy	I-KEY
,	O
and	O
monetary	B-KEY
policy	I-KEY
.	O
In	O
the	O
following	O
years	O
,	O
few	O
of	O
these	O
controversies	O
could	O
be	O
resolved	O
and	O
Mondex	B-KEY
could	O
not	O
be	O
established	O
as	O
a	O
widely	O
used	O
payment	B-KEY
mechanism	I-KEY
.	O
In	O
2001	O
,	O
the	O
experiment	O
was	O
finally	O
terminated	O
.	O
Using	O
the	O
concepts	O
developed	O
in	O
recent	O
science	B-KEY
and	I-KEY
technology	I-KEY
studies	I-KEY
-LRB-	O
STS	O
-RRB-	O
,	O
the	O
article	O
analyzes	O
these	O
controversies	O
as	O
resulting	O
from	O
the	O
difficulties	O
of	O
fitting	O
electronic	B-KEY
cash	I-KEY
,	O
a	O
new	O
sociotechnical	B-KEY
system	I-KEY
,	O
into	O
the	O
complex	O
setting	O
of	O
the	O
existing	O
payment	O
system	O
.	O
The	O
story	O
of	O
Mondex	B-KEY
not	O
only	O
offers	O
lessons	O
on	O
why	O
technologies	O
fail	O
,	O
but	O
also	O
offers	O
insight	O
into	O
how	O
short-term	B-KEY
failures	I-KEY
can	O
contribute	O
to	O
long-term	B-KEY
transformations	I-KEY
.	O
This	O
suggests	O
the	O
need	O
to	O
rethink	O
the	O
dichotomy	O
of	O
success	O
and	O
failure	O

Female	B-KEY
computer	I-KEY
science	I-KEY
doctorates	I-KEY
:	O
what	O
does	O
the	O
survey	B-KEY
of	I-KEY
earned	I-KEY
doctorates	I-KEY
reveal	O
?	O
Based	O
on	O
the	O
National	O
Center	O
for	O
Education	O
Statistics	O
-LRB-	O
2000	O
-RRB-	O
,	O
in	O
the	O
1997-1998	O
academic	O
year	O
26.7	O
%	O
of	O
earned	O
bachelors	O
'	O
degrees	O
,	O
29.0	O
%	O
of	O
earned	O
masters	O
'	O
degrees	O
and	O
16.3	O
%	O
of	O
earned	O
doctorates	O
'	O
degrees	O
in	O
computer	O
science	O
were	O
awarded	O
to	O
women	O
.	O
As	O
these	O
percentages	O
suggest	O
,	O
women	O
are	O
underrepresented	O
at	O
all	O
academic	O
levels	O
in	O
computer	O
science	O
-LRB-	O
Camp	O
,	O
1997	O
-RRB-	O
.	O
The	O
most	O
severe	O
shortage	O
occurs	O
at	O
the	O
top	O
level-the	O
doctorate	O
in	O
computer	O
science	O
.	O
We	O
know	O
very	O
little	O
about	O
the	O
women	O
who	O
persist	O
to	O
the	O
top	O
level	O
of	O
academic	O
achievement	O
in	O
computer	O
science	O
.	O
This	O
paper	O
examines	O
a	O
subset	O
of	O
data	O
collected	O
through	O
the	O
Survey	B-KEY
of	I-KEY
Earned	I-KEY
Doctorates	I-KEY
-LRB-	O
SED	O
-RRB-	O
.	O
The	O
specific	O
focus	O
of	O
this	O
paper	O
is	O
to	O
identify	O
trends	O
that	O
have	O
emerged	O
from	O
the	O
SED	O
with	O
respect	O
to	O
females	O
completing	O
doctorates	O
in	O
computer	O
science	O
between	O
the	O
academic	O
years	O
1990-1991	O
and	O
1999-2000	O
.	O
Although	O
computer	O
science	O
doctorates	O
include	O
doctorates	O
in	O
information	B-KEY
science	I-KEY
,	O
prior	O
research	O
-LRB-	O
Camp	O
,	O
1997	O
-RRB-	O
suggests	O
that	O
the	O
percentage	O
of	O
women	O
completing	O
doctorates	O
in	O
information	B-KEY
science	I-KEY
as	O
compared	O
to	O
computer	O
science	O
is	O
low	O
.	O
The	O
specific	O
research	O
questions	O
are	O
:	O
1	O
.	O
How	O
does	O
the	O
percentage	O
of	O
women	O
who	O
complete	O
doctorates	O
in	O
computer	O
science	O
compare	O
to	O
those	O
that	O
complete	O
doctorates	O
in	O
other	O
fields	O
?	O
2	O
.	O
How	O
does	O
the	O
length	O
of	O
time	O
in	O
school	O
and	O
the	O
sources	O
of	O
funding	O
differ	O
for	O
females	O
as	O
compared	O
to	O
males	O
who	O
complete	O
doctorates	O
in	O
computer	O
science	O
?	O
3	O
.	O
Where	O
do	O
women	O
go	O
after	O
completing	O
doctorates	O
in	O
computer	O
science	O
and	O
what	O
positions	O
do	O
they	O
acquire	O
?	O
How	O
do	O
these	O
experiences	O
differ	O
from	O
their	O
male	O
peers	O
?	O

If	O
the	O
RedBoot	B-KEY
fits	O
-LSB-	O
open-source	B-KEY
ROM	I-KEY
monitor	I-KEY
-RSB-	O
Many	O
embedded	O
developers	O
today	O
use	O
a	O
ROM	O
-	O
or	O
flash-resident	B-KEY
software	I-KEY
program	I-KEY
that	O
provides	O
functionality	O
such	O
as	O
loading	O
and	O
running	O
application	O
software	O
,	O
scripting	B-KEY
,	O
read/write	O
access	O
to	O
processor	O
registers	O
,	O
and	O
memory	B-KEY
dumps	I-KEY
.	O
A	O
ROM	O
monitor	O
,	O
as	O
it	O
is	O
often	O
called	O
,	O
can	O
be	O
a	O
useful	O
and	O
far	O
less	O
expensive	O
debugging	B-KEY
tool	I-KEY
than	O
an	O
in-circuit	O
emulator	O
.	O
This	O
article	O
describes	O
the	O
RedBoot	B-KEY
ROM	O
monitor	O
.	O
It	O
takes	O
a	O
look	O
at	O
the	O
features	O
offered	O
by	O
the	O
RedBoot	B-KEY
ROM	O
monitor	O
and	O
sees	O
how	O
it	O
can	O
be	O
configured	O
.	O
It	O
also	O
walks	O
through	O
the	O
steps	O
of	O
rebuilding	O
and	O
installing	O
a	O
new	O
RedBoot	B-KEY
image	O
on	O
a	O
target	O
platform	O
.	O
Finally	O
,	O
it	O
looks	O
at	O
future	O
enhancements	O
that	O
are	O
coming	O
in	O
new	O
releases	O
and	O
how	O
to	O
get	O
support	O
and	O
additional	O
information	O
when	O
using	O
RedBoot	B-KEY
.	O
Although	O
RedBoot	B-KEY
uses	O
software	O
modules	O
from	O
the	O
eCos	B-KEY
real-time	B-KEY
operating	I-KEY
system	I-KEY
-LRB-	O
RTOS	O
-RRB-	O
and	O
is	O
often	O
used	O
in	O
systems	O
running	O
embedded	B-KEY
Linux	I-KEY
,	O
it	O
is	O
completely	O
independent	O
of	O
both	O
operating	O
systems	O
.	O
RedBoot	B-KEY
can	O
be	O
used	O
with	O
any	O
operating	O
system	O
or	O
RTOS	O
,	O
or	O
even	O
without	O
one	O

Upper	B-KEY
bound	I-KEY
analysis	I-KEY
of	O
oblique	B-KEY
cutting	I-KEY
with	O
nose	B-KEY
radius	I-KEY
tools	I-KEY
A	O
generalized	O
upper	O
bound	O
model	O
for	O
calculating	O
the	O
chip	B-KEY
flow	I-KEY
angle	I-KEY
in	O
oblique	B-KEY
cutting	I-KEY
using	O
flat-faced	O
nose	B-KEY
radius	I-KEY
tools	I-KEY
is	O
described	O
.	O
The	O
projection	O
of	O
the	O
uncut	B-KEY
chip	I-KEY
area	I-KEY
on	O
the	O
rake	O
face	O
is	O
divided	O
into	O
a	O
number	O
of	O
elements	O
parallel	O
to	O
an	O
assumed	O
chip	O
flow	O
direction	O
.	O
The	O
length	O
of	O
each	O
of	O
these	O
elements	O
is	O
used	O
to	O
find	O
the	O
length	O
of	O
the	O
corresponding	O
element	O
on	O
the	O
shear	B-KEY
surface	I-KEY
using	O
the	O
ratio	O
of	O
the	O
shear	B-KEY
velocity	I-KEY
to	O
the	O
chip	B-KEY
velocity	I-KEY
.	O
The	O
area	O
of	O
each	O
element	O
is	O
found	O
as	O
the	O
cross	O
product	O
of	O
the	O
length	O
and	O
its	O
width	O
along	O
the	O
cutting	O
edge	O
.	O
Summing	O
up	O
the	O
area	O
of	O
the	O
elements	O
along	O
the	O
shear	B-KEY
surface	I-KEY
,	O
the	O
total	O
shear	B-KEY
surface	I-KEY
area	O
is	O
obtained	O
.	O
The	O
friction	B-KEY
area	I-KEY
is	O
calculated	O
using	O
the	O
similarity	O
between	O
orthogonal	O
and	O
oblique	B-KEY
cutting	I-KEY
in	O
the	O
`	O
equivalent	O
'	O
plane	O
that	O
includes	O
both	O
the	O
cutting	O
velocity	O
and	O
chip	B-KEY
velocity	I-KEY
.	O
The	O
cutting	O
power	O
is	O
obtained	O
by	O
summing	O
the	O
shear	O
power	O
and	O
the	O
friction	O
power	O
.	O
The	O
actual	O
chip	B-KEY
flow	I-KEY
angle	I-KEY
and	O
chip	B-KEY
velocity	I-KEY
are	O
obtained	O
by	O
minimizing	O
the	O
cutting	O
power	O
with	O
respect	O
to	O
both	O
these	O
variables	O
.	O
The	O
shape	O
of	O
the	O
curved	O
shear	B-KEY
surface	I-KEY
,	O
the	O
chip	O
cross	O
section	O
and	O
the	O
cutting	O
force	O
obtained	O
from	O
this	O
model	O
are	O
presented	O

H/sub	O
2	O
/	O
optimization	O
of	O
the	O
three-element	B-KEY
type	I-KEY
dynamic	I-KEY
vibration	I-KEY
absorbers	I-KEY
The	O
dynamic	O
vibration	O
absorber	O
-LRB-	O
DVA	O
-RRB-	O
is	O
a	O
passive	B-KEY
vibration	I-KEY
control	I-KEY
device	O
which	O
is	O
attached	O
to	O
a	O
vibrating	O
body	O
-LRB-	O
called	O
a	O
primary	O
system	O
-RRB-	O
subjected	O
to	O
exciting	O
force	O
or	O
motion	O
.	O
In	O
this	O
paper	O
,	O
we	O
will	O
discuss	O
an	O
optimization	O
problem	O
of	O
the	O
three-element	O
type	O
DVA	O
on	O
the	O
basis	O
of	O
the	O
H/sub	O
2	O
/	O
optimization	O
criterion	O
.	O
The	O
objective	O
of	O
the	O
H/sub	O
2	O
/	O
optimization	O
is	O
to	O
reduce	O
the	O
total	O
vibration	O
energy	O
of	O
the	O
system	O
for	O
overall	O
frequencies	O
;	O
the	O
total	O
area	O
under	O
the	O
power	B-KEY
spectrum	I-KEY
response	I-KEY
curve	O
is	O
minimized	O
in	O
this	O
criterion	O
.	O
If	O
the	O
system	O
is	O
subjected	O
to	O
random	O
excitation	O
instead	O
of	O
sinusoidal	O
excitation	O
,	O
then	O
the	O
H/sub	O
2	O
/	O
optimization	O
is	O
probably	O
more	O
desirable	O
than	O
the	O
popular	O
H/sub	O
infinity	O
/	O
optimization	O
.	O
In	O
the	O
past	O
decade	O
there	O
has	O
been	O
increasing	O
interest	O
in	O
the	O
three-element	O
type	O
DVA	O
.	O
However	O
,	O
most	O
previous	O
studies	O
on	O
this	O
type	O
of	O
DVA	O
were	O
based	O
on	O
the	O
H/sub	O
infinity	O
/	O
optimization	O
design	O
,	O
and	O
no	O
one	O
has	O
been	O
able	O
to	O
find	O
the	O
algebraic	O
solution	O
as	O
of	O
yet	O
.	O
We	O
found	O
a	O
closed-form	O
exact	O
solution	O
for	O
a	O
special	O
case	O
where	O
the	O
primary	O
system	O
has	O
no	O
damping	O
.	O
Furthermore	O
,	O
the	O
general	O
case	O
solution	O
including	O
the	O
damped	O
primary	O
system	O
is	O
presented	O
in	O
the	O
form	O
of	O
a	O
numerical	O
solution	O
.	O
The	O
optimum	O
parameters	O
obtained	O
here	O
are	O
compared	O
to	O
those	O
of	O
the	O
conventional	O
Voigt	O
type	O
DVA	O
.	O
They	O
are	O
also	O
compared	O
to	O
other	O
optimum	O
parameters	O
based	O
on	O
the	O
H/sub	O
infinity	O
/	O
criterion	O

Modeling	B-KEY
and	O
simulation	O
of	O
an	O
ABR	B-KEY
flow	I-KEY
control	I-KEY
algorithm	I-KEY
using	O
a	O
virtual	B-KEY
source/virtual	I-KEY
destination	I-KEY
switch	I-KEY
The	O
available	O
bit	O
rate	O
-LRB-	O
ABR	O
-RRB-	O
service	O
class	O
of	O
asynchronous	O
transfer	O
mode	O
networks	O
uses	O
a	O
feedback	B-KEY
control	I-KEY
mechanism	I-KEY
to	O
adapt	O
to	O
varying	O
link	B-KEY
capacities	I-KEY
.	O
The	O
virtual	O
source/virtual	O
destination	O
-LRB-	O
VS/VD	O
-RRB-	O
technique	O
offers	O
the	O
possibility	O
of	O
segmenting	O
the	O
otherwise	O
end-to-end	O
ABR	O
control	B-KEY
loop	I-KEY
into	O
separate	O
loops	O
.	O
The	O
improved	O
feedback	B-KEY
delay	I-KEY
and	O
control	O
of	O
ABR	O
traffic	O
inside	O
closed	B-KEY
segments	I-KEY
provide	O
a	O
better	O
performance	O
for	O
ABR	O
connections	O
.	O
This	O
article	O
presents	O
the	O
use	O
of	O
classical	B-KEY
linear	I-KEY
control	I-KEY
theory	I-KEY
to	O
model	B-KEY
and	O
develop	O
an	O
ABR	O
VS/VD	O
flow	O
control	O
algorithm	O
.	O
Discrete	B-KEY
event	I-KEY
simulations	I-KEY
are	O
used	O
to	O
analyze	O
the	O
behavior	O
of	O
the	O
algorithm	O
with	O
respect	O
to	O
transient	B-KEY
behavior	I-KEY
and	O
correctness	O
of	O
the	O
control	B-KEY
model	I-KEY
.	O
Linear	O
control	O
theory	O
offers	O
the	O
means	O
to	O
derive	O
correct	O
choices	O
of	O
parameters	O
and	O
to	O
assess	O
performance	B-KEY
issues	I-KEY
,	O
such	O
as	O
stability	B-KEY
of	O
the	O
system	O
,	O
during	O
the	O
design	O
phase	O
.	O
The	O
performance	O
goals	O
are	O
high	B-KEY
link	I-KEY
utilization	I-KEY
,	O
fair	B-KEY
bandwidth	I-KEY
distribution	I-KEY
,	O
and	O
robust	B-KEY
operation	I-KEY
in	O
various	O
environments	O
,	O
which	O
are	O
verified	O
by	O
discrete	B-KEY
event	I-KEY
simulations	I-KEY
.	O
The	O
major	O
contribution	O
of	O
this	O
work	O
is	O
the	O
use	O
of	O
analytic	O
methods	O
-LRB-	O
linear	O
control	O
theory	O
-RRB-	O
to	O
model	B-KEY
and	O
design	O
an	O
ABR	B-KEY
flow	I-KEY
control	I-KEY
algorithm	I-KEY
tailored	O
for	O
the	O
special	O
layout	O
of	O
a	O
VS/VD	O
switch	O
,	O
and	O
the	O
use	O
of	O
simulation	O
techniques	O
to	O
verify	O
the	O
result	O

Domesticating	O
computers	O
and	O
the	O
Internet	B-KEY
The	O
people	O
who	O
use	O
computers	O
and	O
the	O
ways	O
they	O
use	O
them	O
have	O
changed	O
substantially	O
over	O
the	O
past	O
25	O
years	O
.	O
In	O
the	O
beginning	O
highly	B-KEY
educated	I-KEY
people	I-KEY
,	O
mostly	O
men	O
,	O
in	O
technical	B-KEY
professions	I-KEY
used	O
computers	O
for	O
work	O
,	O
but	O
over	O
time	O
a	O
much	O
broader	O
range	O
of	O
people	O
are	O
using	O
computers	O
for	O
personal	O
and	O
domestic	B-KEY
purposes	I-KEY
.	O
This	O
trend	O
is	O
still	O
continuing	O
,	O
and	O
over	O
a	O
shorter	O
time	O
scale	O
has	O
been	O
replicated	O
with	O
the	O
use	O
of	O
the	O
Internet	B-KEY
.	O
The	O
paper	O
uses	O
data	O
from	O
four	O
national	B-KEY
surveys	I-KEY
to	O
document	O
how	O
personal	B-KEY
computers	I-KEY
and	O
the	O
Internet	B-KEY
have	O
become	O
increasingly	O
domesticated	O
since	O
1995	O
and	O
to	O
explore	O
the	O
mechanisms	O
for	O
this	O
shift	O
.	O
Now	O
people	O
log	O
on	O
more	O
often	O
from	O
home	O
than	O
from	O
places	O
of	O
employment	O
and	O
do	O
so	O
for	O
pleasure	O
and	O
for	O
personal	O
purposes	O
rather	O
than	O
for	O
their	O
jobs	O
.	O
Analyses	O
comparing	O
veteran	O
Internet	B-KEY
users	O
to	O
novices	O
in	O
1998	O
and	O
2000	O
and	O
analyses	O
comparing	O
the	O
change	O
in	O
use	O
within	O
a	O
single	O
sample	O
between	O
1995	O
and	O
1996	O
support	O
two	O
complementary	O
explanations	O
for	O
how	O
these	O
technologies	O
have	O
become	O
domesticated	O
.	O
Women	B-KEY
,	O
children	B-KEY
,	O
and	O
less	O
well-educated	O
individuals	O
are	O
increasingly	O
using	O
computers	O
and	O
the	O
Internet	B-KEY
and	O
have	O
a	O
more	O
personal	O
set	O
of	O
motives	O
than	O
well-educated	O
men	O
.	O
In	O
addition	O
,	O
the	O
widespread	O
diffusion	O
of	O
the	O
PC	O
and	O
the	O
Internet	B-KEY
and	O
the	O
response	O
of	O
the	O
computing	B-KEY
industry	I-KEY
to	O
the	O
diversity	O
in	O
consumers	O
has	O
led	O
to	O
a	O
rich	O
set	O
of	O
personal	O
and	O
domestic	B-KEY
services	I-KEY

Quadratic	B-KEY
Newton	I-KEY
iteration	I-KEY
for	O
systems	B-KEY
with	I-KEY
multiplicity	I-KEY
Newton	O
's	O
iterator	O
is	O
one	O
of	O
the	O
most	O
popular	O
components	O
of	O
polynomial	B-KEY
equation	I-KEY
system	I-KEY
solvers	I-KEY
,	O
either	O
from	O
the	O
numeric	O
or	O
symbolic	O
point	O
of	O
view	O
.	O
This	O
iterator	O
usually	O
handles	O
smooth	O
situations	O
only	O
-LRB-	O
when	O
the	O
Jacobian	B-KEY
matrix	I-KEY
associated	O
to	O
the	O
system	O
is	O
invertible	O
-RRB-	O
.	O
This	O
is	O
often	O
a	O
restrictive	O
factor	O
.	O
Generalizing	O
Newton	O
's	O
iterator	O
is	O
still	O
an	O
open	O
problem	O
:	O
How	O
to	O
design	O
an	O
efficient	O
iterator	O
with	O
a	O
quadratic	B-KEY
convergence	I-KEY
even	O
in	O
degenerate	O
cases	O
?	O
We	O
propose	O
an	O
answer	O
for	O
an	O
m-adic	B-KEY
topology	I-KEY
when	O
the	O
ideal	O
m	O
can	O
be	O
chosen	O
generic	O
enough	O
:	O
compared	O
to	O
a	O
smooth	O
case	O
we	O
prove	O
quadratic	B-KEY
convergence	I-KEY
with	O
a	O
small	O
overhead	O
that	O
grows	O
with	O
the	O
square	O
of	O
the	O
multiplicity	O
of	O
the	O
root	O

A	O
mechanism	O
for	O
inferring	O
approximate	B-KEY
solutions	I-KEY
under	O
incomplete	B-KEY
knowledge	I-KEY
based	O
on	O
rule	B-KEY
similarity	I-KEY
This	O
paper	O
proposes	O
an	O
inference	B-KEY
method	I-KEY
which	O
can	O
obtain	O
an	O
approximate	B-KEY
solution	I-KEY
even	O
if	O
the	O
knowledge	O
stored	O
in	O
the	O
problem-solving	O
system	O
is	O
incomplete	O
.	O
When	O
a	O
rule	O
needed	O
for	O
solving	O
the	O
problem	O
does	O
not	O
exist	O
,	O
the	O
problem	O
can	O
be	O
solved	O
by	O
using	O
rules	B-KEY
similar	I-KEY
to	O
the	O
existing	O
rules	O
.	O
In	O
an	O
implementation	O
using	O
the	O
SLD	B-KEY
procedure	I-KEY
,	O
a	O
resolution	O
is	O
executed	O
between	O
a	O
subgoal	O
and	O
a	O
rule	O
if	O
an	O
atom	O
of	O
the	O
subgoal	O
is	O
similar	O
to	O
the	O
consequence	B-KEY
atom	I-KEY
of	O
the	O
rule	O
.	O
Similarities	O
between	O
atoms	O
are	O
calculated	O
using	O
a	O
knowledge	O
base	O
of	O
words	O
with	O
account	O
of	O
the	O
reasoning	B-KEY
situation	O
,	O
and	O
the	O
reliability	B-KEY
of	O
the	O
derived	O
solution	O
is	O
calculated	O
based	O
on	O
these	O
similarities	O
.	O
If	O
many	O
solutions	O
are	O
obtained	O
,	O
they	O
are	O
grouped	O
into	O
classes	O
of	O
similar	O
solutions	O
and	O
a	O
representative	B-KEY
solution	I-KEY
is	O
then	O
selected	O
for	O
each	O
class	O
.	O
The	O
proposed	O
method	O
was	O
verified	O
experimentally	O
by	O
solving	O
simple	O
problems	O

Repeated	B-KEY
games	I-KEY
with	O
lack	O
of	O
information	O
on	O
one	O
side	O
:	O
the	O
dual	O
differential	O
approach	O
We	O
introduce	O
the	O
dual	B-KEY
differential	I-KEY
game	I-KEY
of	O
a	O
repeated	B-KEY
game	I-KEY
with	O
lack	O
of	O
information	O
on	O
one	O
side	O
as	O
the	O
natural	O
continuous	O
time	O
version	O
of	O
the	O
dual	O
game	O
introduced	O
by	O
De	O
Meyer	O
-LRB-	O
1996	O
-RRB-	O
.	O
A	O
traditional	O
way	O
to	O
study	O
the	O
value	O
of	O
differential	O
games	O
is	O
through	O
discrete	B-KEY
time	I-KEY
approximations	O
.	O
Here	O
,	O
we	O
follow	O
the	O
opposite	O
approach	O
:	O
We	O
identify	O
the	O
limit	B-KEY
value	I-KEY
of	O
a	O
repeated	B-KEY
game	I-KEY
in	O
discrete	B-KEY
time	I-KEY
as	O
the	O
value	O
of	O
a	O
differential	O
game	O
.	O
Namely	O
,	O
we	O
use	O
the	O
recursive	O
structure	O
for	O
the	O
finitely	O
repeated	O
version	O
of	O
the	O
dual	O
game	O
to	O
construct	O
a	O
differential	O
game	O
for	O
which	O
the	O
upper	O
values	O
of	O
the	O
uniform	O
discretization	O
satisfy	O
precisely	O
the	O
same	O
property	O
.	O
The	O
value	O
of	O
the	O
dual	B-KEY
differential	I-KEY
game	I-KEY
exists	O
and	O
is	O
the	O
unique	O
viscosity	B-KEY
solution	I-KEY
of	O
a	O
first-order	O
derivative	O
equation	O
with	O
a	O
limit	B-KEY
condition	I-KEY
.	O
We	O
identify	O
the	O
solution	O
by	O
translating	O
viscosity	O
properties	O
in	O
the	O
primal	O

Revisiting	O
Hardy	O
's	O
paradox	O
:	O
Counterfactual	B-KEY
statements	I-KEY
,	O
real	B-KEY
measurements	I-KEY
,	O
entanglement	B-KEY
and	O
weak	B-KEY
values	I-KEY
Hardy	O
's	O
-LRB-	O
1992	O
-RRB-	O
paradox	O
is	O
revisited	O
.	O
Usually	O
the	O
paradox	O
is	O
dismissed	O
on	O
grounds	O
of	O
counterfactuality	O
,	O
i.e.	O
,	O
because	O
the	O
paradoxical	B-KEY
effects	I-KEY
appear	O
only	O
when	O
one	O
considers	O
results	O
of	O
experiments	O
which	O
do	O
not	O
actually	O
take	O
place	O
.	O
We	O
suggest	O
a	O
new	O
set	O
of	O
measurements	O
in	O
connection	O
with	O
Hardy	O
's	O
scheme	O
,	O
and	O
show	O
that	O
when	O
they	O
are	O
actually	O
performed	O
,	O
they	O
yield	O
strange	O
and	O
surprising	O
outcomes	O
.	O
More	O
generally	O
,	O
we	O
claim	O
that	O
counterfactual	O
paradoxes	O
point	O
to	O
a	O
deeper	O
structure	O
inherent	O
to	O
quantum	B-KEY
mechanics	I-KEY

The	O
treatment	O
of	O
fear	O
of	O
flying	O
:	O
a	O
controlled	O
study	O
of	O
imaginal	O
and	O
virtual	B-KEY
reality	I-KEY
graded	I-KEY
exposure	I-KEY
therapy	I-KEY
The	O
goal	O
of	O
this	O
study	O
was	O
to	O
determine	O
if	O
virtual	B-KEY
reality	I-KEY
graded	I-KEY
exposure	I-KEY
therapy	I-KEY
-LRB-	O
VRGET	O
-RRB-	O
was	O
equally	O
efficacious	O
,	O
more	O
efficacious	O
,	O
or	O
less	O
efficacious	O
,	O
than	O
imaginal	B-KEY
exposure	I-KEY
therapy	I-KEY
in	O
the	O
treatment	O
of	O
fear	O
of	O
flying	O
.	O
Thirty	O
participants	O
-LRB-	O
Age	O
=	O
39.8	O
+	O
or-9	O
.7	O
-RRB-	O
with	O
confirmed	O
DSM-IV	O
diagnosis	O
of	O
specific	O
phobia	B-KEY
fear	O
of	O
flying	O
were	O
randomly	O
assigned	O
to	O
one	O
of	O
three	O
groups	O
:	O
VRGET	O
with	O
no	O
physiological	B-KEY
feedback	O
-LRB-	O
VRGETno	O
-RRB-	O
,	O
VRGET	O
with	O
physiological	O
feedback	O
-LRB-	O
VRGETpm	O
-RRB-	O
,	O
or	O
systematic	O
desensitization	O
with	O
imaginal	O
exposure	O
therapy	O
-LRB-	O
IET	O
-RRB-	O
.	O
Eight	O
sessions	O
were	O
conducted	O
once	O
a	O
week	O
.	O
During	O
each	O
session	O
,	O
physiology	B-KEY
was	O
measured	O
to	O
give	O
an	O
objective	O
measurement	O
of	O
improvement	O
over	O
the	O
course	O
of	O
exposure	O
therapy	O
.	O
In	O
addition	O
,	O
self-report	O
questionnaires	B-KEY
,	O
subjective	B-KEY
ratings	I-KEY
of	I-KEY
anxiety	I-KEY
-LRB-	O
SUDs	O
-RRB-	O
,	O
and	O
behavioral	B-KEY
observations	I-KEY
-LRB-	O
included	O
here	O
as	O
flying	O
behavior	O
before	O
beginning	O
treatment	O
and	O
at	O
a	O
three-month	O
posttreatment	O
followup	O
-RRB-	O
were	O
included	O
.	O
In	O
the	O
analysis	O
of	O
results	O
,	O
the	O
Chi-square	B-KEY
test	I-KEY
of	O
behavioral	B-KEY
observations	I-KEY
based	O
on	O
a	O
three-month	O
posttreatment	O
followup	O
revealed	O
a	O
statistically	O
significant	O
difference	O
in	O
flying	O
behavior	O
between	O
the	O
groups	O
-LSB-	O
chi	O
/	O
sup	O
2	O
/	O
-LRB-	O
4	O
-RRB-	O
=	O
19.41	O
,	O
p	O
<	O
0.001	O
-RSB-	O
.	O
Only	O
one	O
participant	O
-LRB-	O
10	O
%	O
-RRB-	O
who	O
received	O
IET	O
,	O
eight	O
of	O
the	O
ten	O
participants	O
-LRB-	O
80	O
%	O
-RRB-	O
who	O
received	O
VRGETno	O
,	O
and	O
ten	O
out	O
of	O
the	O
ten	O
participants	O
-LRB-	O
100	O
%	O
-RRB-	O
who	O
received	O
VRGETpm	O
reported	O
an	O
ability	O
to	O
fly	O
without	O
medication	O
or	O
alcohol	O
at	O
three-month	O
followup	O
.	O
Although	O
this	O
study	O
included	O
small	O
sample	O
sizes	O
for	O
the	O
three	O
groups	O
,	O
the	O
results	O
showed	O
VRGET	O
was	O
more	O
effective	O
than	O
IET	O
in	O
the	O
treatment	O
of	O
flying	O
.	O
It	O
also	O
suggests	O
that	O
physiological	B-KEY
feedback	O
may	O
add	O
to	O
the	O
efficacy	O
of	O
VR	O
treatment	O

Integration	O
,	O
the	O
Web	O
are	O
key	O
this	O
season	O
-LSB-	O
tax	O
-RSB-	O
Integration	O
and	O
the	O
Web	O
are	O
driving	O
many	O
of	O
the	O
enhancements	O
planned	O
by	O
tax	O
preparation	O
software	O
vendors	O
for	O
this	O
coming	O
season	O

Caring	O
for	O
your	O
new	B-KEY
lawyers	I-KEY
In	O
any	O
given	O
year	O
,	O
a	O
striking	O
number	O
of	O
lawyers	O
are	O
in	O
a	O
state	O
of	O
flux	O
,	O
from	O
newly	O
minted	O
law	O
school	O
graduates	O
looking	O
for	O
their	O
first	O
job	O
,	O
to	O
senior	O
litigators	O
migrating	O
to	O
new	O
challenges	O
with	O
new	O
firms	O
.	O
The	O
one	O
certainty	O
is	O
that	O
lawyers	O
new	O
to	O
any	O
firm	O
need	O
care	O
and	O
feeding	O
in	O
myriad	O
ways	O
.	O
All	O
of	O
them	O
need	O
to	O
know	O
and	O
understand	O
three	O
things	O
:	O
-LRB-	O
1	O
-RRB-	O
the	O
firm	O
's	O
culture	O
;	O
-LRB-	O
2	O
-RRB-	O
the	O
resources	B-KEY
available	O
to	O
help	O
them	O
develop	O
their	O
practices	O
;	O
and	O
-LRB-	O
3	O
-RRB-	O
where	O
to	O
get	O
help	O
and	O
guidance	O
for	O
research	O
and	O
practice	O
purposes	O
.	O
Obtaining	O
a	O
thorough	O
understanding	O
of	O
a	O
new	O
firm	O
's	O
workings	O
may	O
be	O
the	O
greatest	O
research	O
project	O
lawyers	O
face	O
.	O
How	O
can	O
a	O
firm	O
help	O
its	O
new	B-KEY
lawyers	I-KEY
learn	O
what	O
they	O
need	O
to	O
know	O
?	O
To	O
offer	O
an	O
example	O
,	O
here	O
are	O
programs	O
in	O
place	O
at	O
my	O
firm	O

Fidelity	B-KEY
of	O
quantum	B-KEY
teleportation	I-KEY
through	O
noisy	O
channels	O
We	O
investigate	O
quantum	B-KEY
teleportation	I-KEY
through	O
noisy	B-KEY
quantum	I-KEY
channels	I-KEY
by	O
solving	O
analytically	O
and	O
numerically	O
a	O
master	O
equation	O
in	O
the	O
Lindblad	O
form	O
.	O
We	O
calculate	O
the	O
fidelity	B-KEY
as	O
a	O
function	O
of	O
decoherence	O
rates	O
and	O
angles	O
of	O
a	O
state	O
to	O
be	O
teleported	O
.	O
It	O
is	O
found	O
that	O
the	O
average	O
fidelity	B-KEY
and	O
the	O
range	O
of	O
states	O
to	O
be	O
accurately	O
teleported	O
depend	O
on	O
types	O
of	O
noises	O
acting	O
on	O
quantum	B-KEY
channels	I-KEY
.	O
If	O
the	O
quantum	B-KEY
channels	I-KEY
are	O
subject	O
to	O
isotropic	B-KEY
noise	I-KEY
,	O
the	O
average	O
fidelity	B-KEY
decays	O
to	O
1/2	O
,	O
which	O
is	O
smaller	O
than	O
the	O
best	O
possible	O
value	O
of	O
2/3	O
obtained	O
only	O
by	O
the	O
classical	B-KEY
communication	I-KEY
.	O
On	O
the	O
other	O
hand	O
,	O
if	O
the	O
noisy	B-KEY
quantum	I-KEY
channel	I-KEY
is	O
modeled	O
by	O
a	O
single	O
Lindblad	O
operator	O
,	O
the	O
average	O
fidelity	O
is	O
always	O
greater	O
than	O
2/3	O

To	O
classify	O
or	O
not	O
to	O
classify	O
,	O
that	O
is	O
the	O
question	O
?	O
In	O
addressing	O
classification	O
issues	O
,	O
the	O
librarian	O
needs	O
to	O
decide	O
what	O
best	O
suits	O
the	O
purpose	O
and	O
requirements	O
of	O
the	O
user	O
group	O
and	O
the	O
organisation	O
they	O
work	O
in	O
.	O
The	O
author	O
has	O
used	O
the	O
well-established	O
Moys	B-KEY
Classification	I-KEY
Scheme	I-KEY
.	O
This	O
gives	O
the	O
level	O
of	O
detail	O
required	O
for	O
current	O
stock	O
and	O
allows	O
for	O
the	O
incorporation	O
of	O
new	O
material	O
as	O
the	O
firm	O
's	O
specialisations	O
develop	O
.	O
The	O
scheme	O
is	O
widely	O
used	O
in	O
other	O
firms	O
as	O
well	O
as	O
in	O
the	O
local	O
law	B-KEY
society	I-KEY
library	I-KEY
,	O
so	O
it	O
will	O
be	O
familiar	O
to	O
many	O
users	O

Individual	B-KEY
decision	I-KEY
making	I-KEY
using	O
fuzzy	B-KEY
set	I-KEY
theory	I-KEY
The	O
paper	O
shows	O
the	O
importance	O
of	O
decision	O
making	O
by	O
an	O
individual	O
and	O
highlights	O
the	O
prime	O
domain	O
of	O
decision	O
making	O
where	O
fuzzy	B-KEY
set	I-KEY
theory	I-KEY
can	O
be	O
used	O
as	O
a	O
tool	O
.	O
Fuzzy	B-KEY
set	I-KEY
theory	I-KEY
has	O
been	O
used	O
on	O
rational	O
model	O
of	O
decision	O
making	O
to	O
arrive	O
at	O
the	O
desired	O
conclusion	O

Place/Transition	B-KEY
Petri	I-KEY
net	I-KEY
evolutions	I-KEY
:	O
recording	O
ways	O
,	O
analysis	O
and	O
synthesis	O
Four	O
semantic	B-KEY
domains	I-KEY
for	O
Place/Transition	O
Petri	O
nets	O
and	O
their	O
relationships	O
are	O
considered	O
.	O
They	O
are	O
monoids	B-KEY
of	O
respectively	O
:	O
firing	B-KEY
sequences	I-KEY
,	O
processes	O
,	O
traces	O
and	O
dependence	B-KEY
graphs	I-KEY
.	O
For	O
each	O
of	O
them	O
the	O
analysis	O
and	O
synthesis	O
problem	O
is	O
stated	O
and	O
solved	O
.	O
The	O
monoid	B-KEY
of	O
processes	O
is	O
defined	O
in	O
a	O
non-standard	O
way	O
,	O
Nets	O
under	O
consideration	O
involve	O
weights	O
of	O
arrows	O
and	O
capacities	O
-LRB-	O
finite	O
or	O
infinite	O
-RRB-	O
of	O
places	O
.	O
However	O
,	O
the	O
analysis	O
and	O
synthesis	O
tasks	O
require	O
nets	O
to	O
be	O
pure	O
,	O
i.e.	O
each	O
of	O
their	O
transition	O
must	O
have	O
the	O
pre-set	O
and	O
post-set	B-KEY
disjoint	I-KEY

Development	O
of	O
visual	B-KEY
design	I-KEY
steering	I-KEY
as	O
an	O
aid	O
in	O
large-scale	B-KEY
multidisciplinary	I-KEY
design	I-KEY
optimization	I-KEY
.	O
I.	O
Method	O
development	O
A	O
modified	O
paradigm	O
of	O
computational	B-KEY
steering	I-KEY
-LRB-	O
CS	O
-RRB-	O
,	O
termed	O
visual	B-KEY
design	I-KEY
steering	I-KEY
-LRB-	O
VDS	O
-RRB-	O
,	O
is	O
developed	O
in	O
this	O
paper	O
.	O
The	O
VDS	O
paradigm	O
is	O
applied	O
to	O
optimal	B-KEY
design	I-KEY
problems	I-KEY
to	O
provide	O
a	O
means	O
for	O
capturing	O
and	O
enabling	O
designer	O
insights	O
.	O
VDS	O
allows	O
a	O
designer	O
to	O
make	O
decisions	O
before	O
,	O
during	O
or	O
after	O
an	O
analysis	O
or	O
optimization	O
via	O
a	O
visual	O
environment	O
,	O
in	O
order	O
to	O
effectively	O
steer	O
the	O
solution	O
process	O
.	O
The	O
objective	O
of	O
VDS	O
is	O
to	O
obtain	O
a	O
better	O
solution	O
in	O
less	O
time	O
through	O
the	O
use	O
of	O
designer	O
knowledge	O
and	O
expertise	O
.	O
Using	O
visual	B-KEY
representations	I-KEY
of	O
complex	B-KEY
systems	O
in	O
this	O
manner	O
enables	O
human	O
experience	O
and	O
judgement	O
to	O
be	O
incorporated	O
into	O
the	O
optimal	O
design	O
process	O
at	O
appropriate	O
steps	O
,	O
rather	O
than	O
having	O
traditional	O
black	O
box	O
solvers	O
return	O
solutions	O
from	O
a	O
prescribed	O
input	O
set	O
.	O
Part	O
I	O
of	O
this	O
paper	O
focuses	O
on	O
the	O
research	O
issues	O
pertaining	O
to	O
the	O
Graph	B-KEY
Morphing	I-KEY
visualization	I-KEY
method	I-KEY
created	O
to	O
represent	O
an	O
n-dimensional	B-KEY
optimization	I-KEY
problem	O
using	O
2-dimensional	O
and	O
3-dimensional	O
visualizations	O
.	O
Part	O
II	O
investigates	O
the	O
implementation	O
of	O
the	O
VDS	O
paradigm	O
,	O
using	O
the	O
graph	O
morphing	O
approach	O
,	O
to	O
improve	O
an	O
optimal	O
design	O
process	O
.	O
Specifically	O
,	O
the	O
following	O
issues	O
are	O
addressed	O
:	O
impact	O
of	O
design	B-KEY
variable	I-KEY
changes	I-KEY
on	O
the	O
optimal	O
design	O
space	O
;	O
identification	O
of	O
possible	O
constraint	B-KEY
redundancies	I-KEY
;	O
impact	O
of	O
constraint	B-KEY
tolerances	I-KEY
on	O
the	O
optimal	O
solution	O
:	O
and	O
smoothness	O
of	O
the	O
objective	O
function	O
contours	O
.	O
It	O
is	O
demonstrated	O
that	O
graph	O
morphing	O
can	O
effectively	O
reduce	O
the	O
complexity	B-KEY
and	O
computational	B-KEY
time	I-KEY
associated	O
with	O
some	O
optimization	O
problems	O

Mathematical	O
properties	O
of	O
dominant	B-KEY
AHP	I-KEY
and	O
concurrent	B-KEY
convergence	I-KEY
method	I-KEY
This	O
study	O
discusses	O
the	O
mathematical	O
structure	O
of	O
the	O
dominant	B-KEY
AHP	I-KEY
and	O
the	O
concurrent	B-KEY
convergence	I-KEY
method	I-KEY
which	O
were	O
originally	O
developed	O
by	O
Kinoshita	O
and	O
Nakanishi	O
.	O
They	O
introduced	O
a	O
new	O
concept	O
of	O
a	O
regulating	O
alternative	O
into	O
an	O
analyzing	O
tool	O
for	O
a	O
simple	O
evaluation	O
problem	O
with	O
a	O
criterion	O
set	O
and	O
an	O
alternative	O
set	O
.	O
Although	O
the	O
original	O
idea	O
of	O
the	O
dominant	B-KEY
AHP	I-KEY
and	O
the	O
concurrent	B-KEY
convergence	I-KEY
method	I-KEY
is	O
unique	O
,	O
the	O
dominant	B-KEY
AHP	I-KEY
and	O
the	O
concurrent	B-KEY
convergence	I-KEY
method	I-KEY
are	O
not	O
sufficiently	O
analyzed	O
in	O
mathematical	O
theory	O
.	O
This	O
study	O
shows	O
that	O
the	O
dominant	B-KEY
AHP	I-KEY
consists	O
of	O
a	O
pair	O
of	O
evaluation	O
rules	O
satisfying	O
a	O
certain	O
property	O
of	O
overall	B-KEY
evaluation	I-KEY
vectors	I-KEY
.	O
This	O
study	O
also	O
shows	O
that	O
the	O
convergence	O
of	O
concurrent	B-KEY
convergence	I-KEY
method	I-KEY
is	O
guaranteed	O
theoretically	O

Applying	O
BGL	O
to	O
computational	B-KEY
geometry	I-KEY
The	O
author	O
applies	O
Boost	B-KEY
Graph	I-KEY
Library	I-KEY
to	O
the	O
domain	O
of	O
computational	B-KEY
geometry	I-KEY
.	O
First	O
,	O
he	O
formulates	O
a	O
concrete	O
problem	O
in	O
graph	O
terms	O
.	O
Second	O
,	O
he	O
develops	O
a	O
way	O
to	O
transform	O
the	O
output	O
of	O
an	O
existing	O
algorithm	O
into	O
an	O
appropriate	O
Boost	B-KEY
Graph	I-KEY
Library	I-KEY
data	O
structure	O
.	O
Finally	O
,	O
he	O
implements	O
two	O
new	O
algorithms	O
for	O
my	O
Boost	B-KEY
Graph	I-KEY
Library	I-KEY
graph	O
.	O
The	O
first	O
algorithm	O
gets	O
the	O
job	O
done	O
,	O
but	O
could	O
have	O
been	O
written	O
in	O
any	O
programming	O
language	O
.	O
The	O
second	O
algorithm	O
,	O
however	O
,	O
shows	O
the	O
power	O
of	O
Boost	B-KEY
Graph	I-KEY
Library	I-KEY
's	O
generic	O
programming	O
approach.Graphs	O
,	O
graphics	O
,	O
and	O
generic	O
programming	O
combine	O
in	O
this	O
novel	O
use	O
of	B-KEY
the	I-KEY
Boost	I-KEY
Graph	O
Library	O

Scalable	B-KEY
techniques	I-KEY
from	O
nonparametric	B-KEY
statistics	I-KEY
for	O
real	B-KEY
time	I-KEY
robot	I-KEY
learning	I-KEY
Locally	B-KEY
weighted	I-KEY
learning	I-KEY
-LRB-	O
LWL	O
-RRB-	O
is	O
a	O
class	O
of	O
techniques	O
from	O
nonparametric	B-KEY
statistics	I-KEY
that	O
provides	O
useful	O
representations	O
and	O
training	B-KEY
algorithms	I-KEY
for	O
learning	O
about	O
complex	B-KEY
phenomena	I-KEY
during	O
autonomous	B-KEY
adaptive	I-KEY
control	I-KEY
of	O
robotic	O
systems	O
.	O
The	O
paper	O
introduces	O
several	O
LWL	O
algorithms	O
that	O
have	O
been	O
tested	O
successfully	O
in	O
real-time	O
learning	O
of	O
complex	O
robot	O
tasks	O
.	O
We	O
discuss	O
two	O
major	O
classes	O
of	O
LWL	O
,	O
memory-based	O
LWL	O
and	O
purely	O
incremental	O
LWL	O
that	O
does	O
not	O
need	O
to	O
remember	O
any	O
data	O
explicitly	O
.	O
In	O
contrast	O
to	O
the	O
traditional	O
belief	O
that	O
LWL	O
methods	O
can	O
not	O
work	O
well	O
in	O
high-dimensional	O
spaces	O
,	O
we	O
provide	O
new	O
algorithms	O
that	O
have	O
been	O
tested	O
on	O
up	O
to	O
90	O
dimensional	O
learning	O
problems	O
.	O
The	O
applicability	O
of	O
our	O
LWL	O
algorithms	O
is	O
demonstrated	O
in	O
various	O
robot	O
learning	O
examples	O
,	O
including	O
the	O
learning	O
of	O
devil-sticking	B-KEY
,	O
pole-balancing	B-KEY
by	O
a	O
humanoid	B-KEY
robot	I-KEY
arm	I-KEY
,	O
and	O
inverse-dynamics	B-KEY
learning	I-KEY
for	O
a	O
seven	O
and	O
a	O
30	O
degree-of-freedom	O
robot	O
.	O
In	O
all	O
these	O
examples	O
,	O
the	O
application	O
of	O
our	O
statistical	B-KEY
neural	I-KEY
networks	I-KEY
techniques	I-KEY
allowed	O
either	O
faster	O
or	O
more	O
accurate	O
acquisition	O
of	O
motor	O
control	O
than	O
classical	O
control	O
engineering	O

Synthesis	O
of	O
the	O
control	O
systems	O
via	O
reflection	O
onto	O
auxiliary	B-KEY
surfaces	I-KEY
An	O
approach	O
to	O
robust	B-KEY
control	I-KEY
systems	I-KEY
synthesis	O
,	O
both	O
linear	O
and	O
nonlinear	O
,	O
and	O
nonstationary	O
is	O
offered	O
.	O
The	O
control	O
is	O
carried	O
out	O
,	O
providing	O
the	O
given	O
phase	O
constraints	O
varied	O
in	O
acceptable	O
limits	O
,	O
in	O
view	O
of	O
constraints	O
on	O
its	O
value	O
and	O
incompleteness	O
of	O
the	O
information	O
about	O
functioning	O
disturbances	O
.	O
The	O
approach	O
is	O
based	O
on	O
the	O
introduction	O
of	O
auxiliary	O
integral	O
surfaces	O
,	O
on	O
which	O
the	O
initial	O
moving	O
is	O
projected	O
.	O
As	O
a	O
result	O
the	O
reduced	O
equivalent	O
moving	O
is	O
formed	O
,	O
being	O
described	O
by	O
the	O
scalar	O
equation	O
which	O
in	O
many	O
important	O
cases	O
can	O
be	O
integrated	O
directly	O
.	O
On	O
the	O
basis	O
of	O
the	O
equation	O
obtained	O
solving	O
a	O
synthesis	O
task	O
is	O
carried	O
out	O
and	O
can	O
be	O
reduced	O
to	O
algebraic	O
or	O
integral	B-KEY
inequalities	I-KEY
.	O
The	O
final	O
relations	O
defined	O
for	O
linear	B-KEY
equivalent	I-KEY
moving	I-KEY
are	O
presented	O

Design	O
and	O
manufacture	O
of	O
a	O
lightweight	B-KEY
piezo-composite	I-KEY
curved	I-KEY
actuator	I-KEY
In	O
this	O
paper	O
we	O
are	O
concerned	O
with	O
the	O
design	O
,	O
manufacture	O
and	O
performance	B-KEY
test	I-KEY
of	O
a	O
lightweight	B-KEY
piezo-composite	I-KEY
curved	I-KEY
actuator	I-KEY
-LRB-	O
called	O
LIPCA	B-KEY
-RRB-	O
using	O
a	O
top	O
carbon	B-KEY
fiber	I-KEY
composite	I-KEY
layer	I-KEY
with	O
near-zero	B-KEY
coefficient	I-KEY
of	I-KEY
thermal	I-KEY
expansion	I-KEY
-LRB-	O
CTE	O
-RRB-	O
,	O
a	O
middle	O
PZT	B-KEY
ceramic	I-KEY
wafer	I-KEY
,	O
and	O
a	O
bottom	O
glass/epoxy	B-KEY
layer	I-KEY
with	O
a	O
high	O
CTE	O
.	O
The	O
main	O
point	O
of	O
the	O
design	O
for	O
LIPCA	B-KEY
is	O
to	O
replace	O
the	O
heavy	O
metal	O
layers	O
of	O
THUNDER	B-KEY
TM	O
by	O
lightweight	O
fiber	B-KEY
reinforced	I-KEY
plastic	I-KEY
layers	I-KEY
without	O
losing	O
the	O
capabilities	O
for	O
generating	O
high	O
force	O
and	O
large	O
displacement	O
.	O
It	O
is	O
possible	O
to	O
save	O
up	O
to	O
about	O
40	O
%	O
of	O
the	O
weight	O
if	O
we	O
replace	O
the	O
metallic	O
backing	O
material	O
by	O
the	O
light	O
fiber	O
composite	O
layer	O
.	O
We	O
can	O
also	O
have	O
design	O
flexibility	O
by	O
selecting	O
the	O
fiber	O
direction	O
and	O
the	O
size	O
of	O
prepreg	O
layers	O
.	O
In	O
addition	O
to	O
the	O
lightweight	O
advantage	O
and	O
design	O
flexibility	O
,	O
the	O
proposed	O
device	O
can	O
be	O
manufactured	O
without	O
adhesive	O
layers	O
when	O
we	O
use	O
an	O
epoxy	O
resin	O
prepreg	O
system	O
.	O
Glass/epoxy	O
prepregs	O
,	O
a	O
ceramic	O
wafer	O
with	O
electrode	O
surfaces	O
,	O
and	O
a	O
carbon	O
prepreg	O
were	O
simply	O
stacked	O
and	O
cured	O
at	O
an	O
elevated	O
temperature	O
-LRB-	O
177	B-KEY
degrees	I-KEY
C	I-KEY
-RRB-	O
after	O
following	O
an	O
autoclave	O
bagging	O
process	O
.	O
We	O
found	O
that	O
the	O
manufactured	O
composite	O
laminate	O
device	O
had	O
a	O
sufficient	O
curvature	O
after	O
being	O
detached	O
from	O
a	O
flat	O
mould	O
.	O
An	O
analysis	O
method	O
using	O
the	O
classical	O
lamination	O
theory	O
is	O
presented	O
to	O
predict	O
the	O
curvature	O
of	O
LIPCA	B-KEY
after	O
curing	O
at	O
an	O
elevated	O
temperature	O
.	O
The	O
predicted	B-KEY
curvatures	I-KEY
are	O
in	O
quite	O
good	O
agreement	O
with	O
the	O
experimental	O
values	O
.	O
In	O
order	O
to	O
investigate	O
the	O
merits	O
of	O
LIPCA	B-KEY
,	O
performance	B-KEY
tests	I-KEY
of	O
both	O
LIPCA	B-KEY
and	O
THUNDER	B-KEY
TM	O
have	O
been	O
conducted	O
under	O
the	O
same	O
boundary	B-KEY
conditions	I-KEY
.	O
From	O
the	O
experimental	O
actuation	O
tests	O
,	O
it	O
was	O
observed	O
that	O
the	O
developed	O
actuator	O
could	O
generate	O
larger	O
actuation	O
displacement	O
than	O
THUNDER	B-KEY
TM	O

Adjoint-based	B-KEY
optimization	I-KEY
of	O
steady	B-KEY
suction	I-KEY
for	O
disturbance	B-KEY
control	I-KEY
in	O
incompressible	B-KEY
flows	I-KEY
The	O
optimal	O
distribution	O
of	O
steady	B-KEY
suction	I-KEY
needed	O
to	O
control	O
the	O
growth	O
of	O
single	O
or	O
multiple	O
disturbances	O
in	O
quasi-three-dimensional	O
incompressible	O
boundary	O
layers	O
on	O
a	O
flat	B-KEY
plate	I-KEY
is	O
investigated	O
.	O
The	O
evolution	O
of	O
disturbances	O
is	O
analysed	O
in	O
the	O
framework	O
of	O
the	O
parabolized	B-KEY
stability	I-KEY
equations	I-KEY
-LRB-	O
PSE	O
-RRB-	O
.	O
A	O
gradient-based	B-KEY
optimization	I-KEY
procedure	I-KEY
is	O
used	O
and	O
the	O
gradients	O
are	O
evaluated	O
using	O
the	O
adjoint	O
of	O
the	O
parabolized	B-KEY
stability	I-KEY
equations	I-KEY
-LRB-	O
APSE	O
-RRB-	O
and	O
the	O
adjoint	O
of	O
the	O
boundary	O
layer	O
equations	O
-LRB-	O
ABLE	O
-RRB-	O
.	O
The	O
accuracy	O
of	O
the	O
gradient	O
is	O
increased	O
by	O
introducing	O
a	O
stabilization	B-KEY
procedure	I-KEY
for	O
the	O
PSE	O
.	O
Results	O
show	O
that	O
a	O
suction	O
peak	O
appears	O
in	O
the	O
upstream	O
part	O
of	O
the	O
suction	O
region	O
for	O
optimal	O
control	O
of	O
Tollmien-Schlichting	O
-LRB-	O
T-S	O
-RRB-	O
waves	B-KEY
,	I-KEY
steady	I-KEY
streamwise	O
streaks	O
in	O
a	O
two-dimensional	O
boundary	O
layer	B-KEY
and	I-KEY
oblique	O
waves	O
in	O
a	O
quasi-three-dimensional	O
boundary	O
layer	O
subject	O
to	B-KEY
an	I-KEY
adverse	I-KEY
pressure	O
gradient	O
.	O
The	O
mean	B-KEY
flow	I-KEY
modifications	O
due	O
to	O
suction	O
are	O
shown	O
to	O
have	O
a	O
stabilizing	O
effect	O
similar	O
to	O
that	O
of	O
a	O
favourable	O
pressure	O
gradient	O
.	O
It	O
is	O
also	O
shown	O
that	O
the	O
optimal	O
suction	O
distribution	O
for	O
the	O
disturbance	O
of	O
interest	O
reduces	O
the	O
growth	O
rate	O
of	O
other	O
perturbations	O
.	O
Results	O
for	O
control	O
of	O
a	O
steady	B-KEY
cross-flow	I-KEY
mode	I-KEY
in	O
a	O
three-dimensional	O
boundary	O
layer	O
subject	O
to	O
a	O
favourable	O
pressure	O
gradient	O
show	O
that	O
not	O
even	O
large	O
amounts	O
of	O
suction	O
can	O
completely	O
stabilize	O
the	O
disturbance	O

Effect	O
of	O
multileaf	B-KEY
collimator	I-KEY
leaf	I-KEY
width	I-KEY
on	O
physical	B-KEY
dose	I-KEY
distributions	I-KEY
in	O
the	O
treatment	O
of	O
CNS	O
and	O
head	B-KEY
and	I-KEY
neck	I-KEY
neoplasms	I-KEY
with	O
intensity	B-KEY
modulated	I-KEY
radiation	I-KEY
therapy	I-KEY
The	O
purpose	O
of	O
this	O
work	O
is	O
to	O
examine	O
physical	O
radiation	O
dose	O
differences	O
between	O
two	O
multileaf	O
collimator	O
-LRB-	O
MLC	O
-RRB-	O
leaf	O
widths	O
-LRB-	O
5	O
and	O
10	B-KEY
mm	I-KEY
-RRB-	O
in	O
the	O
treatment	O
of	O
CNS	O
and	O
head	B-KEY
and	I-KEY
neck	I-KEY
neoplasms	I-KEY
with	O
intensity	B-KEY
modulated	I-KEY
radiation	I-KEY
therapy	I-KEY
-LRB-	O
IMRT	O
-RRB-	O
.	O
Three	O
clinical	O
patients	O
with	O
CNS	B-KEY
tumors	I-KEY
were	O
planned	O
with	O
two	O
different	O
MLC	O
leaf	O
sizes	O
,	O
5	O
and	O
10	B-KEY
mm	I-KEY
,	O
representing	O
Varian-120	O
and	O
Varian-80	O
Millennium	O
multileaf	O
collimators	O
,	O
respectively	O
.	O
Two	O
sets	O
of	O
IMRT	O
treatment	B-KEY
plans	I-KEY
were	O
developed	O
.	O
The	O
goal	O
of	O
the	O
first	O
set	O
was	O
radiation	O
dose	O
conformality	O
in	O
three	O
dimensions	O
.	O
The	O
goal	O
for	O
the	O
second	O
set	O
was	O
organ	O
avoidance	O
of	O
a	O
nearby	O
critical	O
structure	O
while	O
maintaining	O
adequate	O
coverage	O
of	O
the	O
target	O
volume	O
.	O
Treatment	B-KEY
planning	I-KEY
utilized	O
the	O
CadPlan/Helios	O
system	O
-LRB-	O
Varian	O
Medical	O
Systems	O
,	O
Milpitas	O
CA	O
-RRB-	O
for	O
dynamic	O
MLC	O
treatment	O
delivery	O
.	O
All	O
beam	O
parameters	O
and	O
optimization	O
-LRB-	O
cost	O
function	O
-RRB-	O
parameters	O
were	O
identical	O
for	O
the	O
5	O
and	O
10	B-KEY
mm	I-KEY
plans	O
.	O
For	O
all	O
cases	O
the	O
number	O
of	O
beams	O
,	O
gantry	O
positions	O
,	O
and	O
table	O
positions	O
were	O
taken	O
from	O
clinically	O
treated	O
three-dimensional	O
conformal	B-KEY
radiotherapy	I-KEY
plans	O
.	O
Conformality	O
was	O
measured	O
by	O
the	O
ratio	O
of	O
the	O
planning	O
isodose	O
volume	O
to	O
the	O
target	O
volume	O
.	O
Organ	O
avoidance	O
was	O
measured	O
by	O
the	O
volume	O
of	O
the	O
critical	O
structure	O
receiving	O
greater	O
than	O
90	O
%	O
of	O
the	O
prescription	O
dose	O
-LRB-	O
V/sub	O
90	O
/	O
-RRB-	O
.	O
For	O
three	O
patients	O
with	O
squamous	O
cell	O
carcinoma	O
of	O
the	O
head	O
and	O
neck	O
-LRB-	O
T2-T4	O
N0-N2c	O
M0	O
-RRB-	O
5	O
and	O
10	B-KEY
mm	I-KEY
leaf	O
widths	O
were	O
compared	O
for	O
parotid	B-KEY
preservation	I-KEY
utilizing	O
nine	O
coplanar	O
equally	O
spaced	O
beams	O
delivering	O
a	O
simultaneous	O
integrated	O
boost	O
.	O
Because	O
modest	O
differences	O
in	O
physical	O
dose	O
to	O
the	O
parotid	O
were	O
detected	O
,	O
a	O
NTCP	O
model	O
based	O
upon	O
the	O
clinical	O
parameters	O
of	O
Eisbruch	O
et	O
al.	O
was	O
then	O
used	O
for	O
comparisons	O
.	O
The	O
conformality	O
improved	O
in	O
all	O
three	O
CNS	O
cases	O
for	O
the	O
5	B-KEY
mm	I-KEY
plans	O
compared	O
to	O
the	O
10	B-KEY
mm	I-KEY
plans	O
.	O
For	O
the	O
organ	O
avoidance	O
plans	O
,	O
V/sub	O
90	O
/	O
also	O
improved	O
in	O
two	O
of	O
the	O
three	O
cases	O
when	O
the	O
5	B-KEY
mm	I-KEY
leaf	O
width	O
was	O
utilized	O
for	O
IMRT	O
treatment	O
delivery	O
.	O
In	O
the	O
third	O
case	O
,	O
both	O
the	O
5	O
and	O
10	B-KEY
mm	I-KEY
plans	O
were	O
able	O
to	O
spare	O
the	O
critical	O
structure	O
with	O
none	O
of	O
the	O
structure	O
receiving	O
more	O
than	O
90	O
%	O
of	O
the	O
prescription	O
dose	O
,	O
but	O
in	O
the	O
moderate	O
dose	O
range	O
,	O
less	O
dose	O
was	O
delivered	O
to	O
the	O
critical	O
structure	O
with	O
the	O
5	B-KEY
mm	I-KEY
plan	O
.	O
For	O
the	O
head	O
and	O
neck	O
cases	O
both	O
the	O
5	O
and	O
10	O
*	O
2.5	O
mm	O
beamlets	O
dMLC	O
sliding	O
window	O
techniques	O
spared	O
the	O
contralateral	O
parotid	O
gland	O
while	O
maintaining	O
target	O
volume	O
coverage	O
.	O
The	O
mean	O
parotid	O
dose	O
was	O
modestly	O
lower	O
with	O
the	O
smaller	O
beamlet	B-KEY
size	I-KEY
-LRB-	O
21.04	B-KEY
Gy	I-KEY
vs	O
22.36	B-KEY
Gy	I-KEY
-RRB-	O
.	O
The	O
resulting	O
average	O
NTCP	O
values	O
were	O
13.72	O
%	O
for	O
10	B-KEY
mm	I-KEY
dMLC	O
and	O
8.24	O
%	O
for	O
5	B-KEY
mm	I-KEY
dMLC	O
.	O
In	O
conclusion	O
,	O
five	O
mm	O
leaf	O
width	O
results	O
in	O
an	O
improvement	O
in	O
physical	B-KEY
dose	I-KEY
distribution	I-KEY
over	O
10	B-KEY
mm	I-KEY
leaf	O
width	O
that	O
may	O
be	O
clinically	O
relevant	O
in	O
some	O
cases	O
.	O
These	O
differences	O
may	O
be	O
most	O
pronounced	O
for	O
single	B-KEY
fraction	I-KEY
radiosurgery	I-KEY
or	O
in	O
cases	O
where	O
the	O
tolerance	O
of	O
the	O
sensitive	O
organ	O
is	O
less	O
than	O
or	O
close	O
to	O
the	O
target	O
volume	O
prescription	O

Data	O
allocation	O
on	O
wireless	B-KEY
broadcast	I-KEY
channels	I-KEY
for	O
efficient	O
query	B-KEY
processing	I-KEY
Data	O
broadcast	O
is	O
an	O
excellent	O
method	O
for	O
efficient	O
data	O
dissemination	O
in	O
the	O
mobile	B-KEY
computing	I-KEY
environment	O
.	O
The	O
application	O
domain	O
of	O
data	O
broadcast	O
will	O
be	O
widely	O
expanded	O
in	O
the	O
near	O
future	O
,	O
where	O
the	O
client	O
is	O
expected	O
to	O
perform	O
complex	O
queries	O
or	O
transactions	O
on	O
the	O
broadcast	O
data	O
.	O
To	O
reduce	O
the	O
access	B-KEY
latency	I-KEY
for	O
processing	O
the	O
complex	O
query	O
,	O
it	O
is	O
beneficial	O
to	O
place	O
the	O
data	O
accessed	O
in	O
a	O
query	O
close	O
to	O
each	O
other	O
on	O
the	O
broadcast	O
channel	O
.	O
In	O
this	O
paper	O
,	O
we	O
propose	O
an	O
efficient	O
algorithm	O
to	O
determine	O
the	O
allocation	O
of	O
the	O
data	O
on	O
the	O
broadcast	O
channel	O
such	O
that	O
frequently	O
co-accessed	O
data	O
are	O
not	O
only	O
allocated	O
close	O
to	O
each	O
other	O
,	O
but	O
also	O
in	O
a	O
particular	O
order	O
which	O
optimizes	O
the	O
performance	O
of	O
query	B-KEY
processing	I-KEY
.	O
Our	O
mechanism	O
is	O
based	O
on	O
the	O
well-known	O
problem	O
named	O
optimal	O
linear	O
ordering	O
.	O
Experiments	O
are	O
performed	O
to	O
justify	O
the	O
benefit	O
of	O
our	O
approach	O

Extrapolation	O
in	O
Lie	B-KEY
groups	I-KEY
with	O
approximated	B-KEY
BCH-formula	I-KEY
We	O
present	O
an	O
extrapolation	O
algorithm	O
for	O
the	O
integration	O
of	O
differential	B-KEY
equations	I-KEY
in	O
Lie	B-KEY
groups	I-KEY
which	O
is	O
a	O
suitable	O
generalization	O
of	O
the	O
well-known	O
GBS-algorithm	B-KEY
for	O
ODEs	O
.	O
Sufficiently	O
accurate	O
approximations	O
to	O
the	O
BCH-formula	O
are	O
required	O
to	O
reach	O
a	O
given	O
order	O
.	O
We	O
give	O
such	O
approximations	O
with	O
a	O
minimized	O
number	O
of	O
commutators	O

Strategic	B-KEY
implementation	I-KEY
of	O
IT/IS	O
projects	O
in	O
construction	O
:	O
a	O
case	O
study	O
The	O
need	O
for	O
improved	O
implementation	O
of	O
Information	B-KEY
Technology	I-KEY
-LRB-	O
IT	O
-RRB-	O
and	O
Information	B-KEY
Systems	I-KEY
-LRB-	O
IS	O
-RRB-	O
has	O
been	O
emphasised	O
in	O
both	O
empirical	O
and	O
prescriptive	O
research	O
studies	O
.	O
This	O
problem	O
is	O
magnified	O
in	O
the	O
construction	B-KEY
industry	I-KEY
,	O
which	O
has	O
been	O
slow	O
to	O
embrace	O
and	O
utilise	O
new	O
technologies	O
with	O
negative	O
consequences	O
on	O
productivity	O
and	O
innovation	O
.	O
This	O
paper	O
presents	O
a	O
strategic	B-KEY
implementation	I-KEY
framework	O
for	O
IT/IS	O
projects	O
in	O
construction	O
.	O
The	O
framework	O
builds	O
upon	O
recent	O
published	O
works	O
and	O
encompasses	O
well-documented	O
predictors	B-KEY
for	O
effective	O
IT/IS	O
implementation	O
.	O
A	O
case	O
study	O
with	O
a	O
large	B-KEY
multi-national	I-KEY
construction	I-KEY
organisation	I-KEY
is	O
used	O
to	O
demonstrate	O
the	O
strategic	B-KEY
implementation	I-KEY
of	O
a	O
Project	B-KEY
Management	I-KEY
Information	I-KEY
System	I-KEY
-LRB-	O
PMIS	O
-RRB-	O
used	O
for	O
the	O
construction	O
of	O
a	O
mobile	O
phone	O
telecommunications	O
network	O
in	O
the	O
South	O
East	O
of	O
Queensland	O
,	O
Australia	O

Chaotic	B-KEY
phenomena	I-KEY
and	O
fractional-order	B-KEY
dynamics	I-KEY
in	O
the	O
trajectory	B-KEY
control	I-KEY
of	O
redundant	B-KEY
manipulators	I-KEY
Redundant	B-KEY
manipulators	I-KEY
have	O
some	O
advantages	O
when	O
compared	O
with	O
classical	B-KEY
arms	I-KEY
because	O
they	O
allow	O
the	O
trajectory	B-KEY
optimization	I-KEY
,	O
both	O
on	O
the	O
free	O
space	O
and	O
on	O
the	O
presence	O
of	O
obstacles	O
,	O
and	O
the	O
resolution	O
of	O
singularities	O
.	O
For	O
this	O
type	O
of	O
arms	O
the	O
proposed	O
kinematic	B-KEY
control	I-KEY
algorithms	I-KEY
adopt	O
generalized	B-KEY
inverse	I-KEY
matrices	I-KEY
but	O
,	O
in	O
general	O
,	O
the	O
corresponding	O
trajectory	B-KEY
planning	I-KEY
schemes	I-KEY
show	O
important	O
limitations	O
.	O
Motivated	O
by	O
these	O
problems	O
this	O
paper	O
studies	O
the	O
chaos	O
revealed	O
by	O
the	O
pseudoinverse-based	O
trajectory	O
planning	O
algorithms	O
,	O
using	O
the	O
theory	O
of	O
fractional	B-KEY
calculus	I-KEY

Arbortext	B-KEY
:	O
enabler	O
of	O
multichannel	B-KEY
publishing	I-KEY
A	O
company	O
has	O
a	O
document-say	O
,	O
dosage	O
instructions	O
for	O
a	O
prescription	O
drug	O
or	O
a	O
troubleshooting	O
sheet	O
for	O
a	O
DVD	O
drive	O
.	O
That	O
document	O
starts	O
its	O
life	O
in	O
a	O
predictable	O
format	O
,	O
probably	O
Microsoft	O
Word	O
or	O
WordPerfect	O
,	O
but	O
then-to	O
meet	O
the	O
needs	O
of	O
readers	O
who	O
nowadays	O
demand	O
access	O
via	O
multiple	O
devices-the	O
material	O
has	O
to	O
be	O
translated	O
into	O
many	O
more	O
formats	O
:	O
HTML	O
,	O
PageMaker	O
,	O
or	O
Quark	O
,	O
possibly	O
RTF	O
,	O
almost	O
certainly	O
PDF	O
,	O
and	O
nowadays	O
,	O
next-generation	B-KEY
devices	I-KEY
-LRB-	O
cell	O
phones	O
,	O
handheld	O
computers	O
-RRB-	O
also	O
impose	O
their	O
own	O
requirements	O
.	O
And	O
what	O
if	O
,	O
suddenly	O
,	O
the	O
dosage	O
levels	O
change	O
or	O
new	O
workarounds	O
emerge	O
to	O
handle	O
DVD	O
problems	O
?	O
That	O
's	O
when	O
a	O
company	O
should	O
put	O
in	O
a	O
call	O
to	O
Arbortext	B-KEY
,	O
a	O
20-year-old	O
Ann	O
Arbor	O
,	O
Michigan-based	O
company	O
that	O
exists	O
to	O
solve	O
a	O
single	O
problem	O
:	O
helping	O
clients	O
automate	O
multichannel	B-KEY
publishing	I-KEY

A	O
knowledge-navigation	O
system	O
for	O
dimensional	B-KEY
metrology	I-KEY
Geometric	B-KEY
dimensioning	I-KEY
and	O
tolerancing	B-KEY
-LRB-	O
GD&T	O
-RRB-	O
is	O
a	O
method	O
to	O
specify	O
the	O
dimensions	O
and	O
form	O
of	O
a	O
part	O
so	O
that	O
it	O
will	O
meet	O
its	O
design	O
intent	O
.	O
GD&T	O
is	O
difficult	O
to	O
master	O
for	O
two	O
main	O
reasons	O
.	O
First	O
,	O
it	O
is	O
based	O
on	O
complex	O
3D	O
geometric	O
entities	O
and	O
relationships	O
.	O
Second	O
,	O
the	O
geometry	O
is	O
associated	O
with	O
a	O
large	O
,	O
diverse	O
knowledge	O
base	O
of	O
dimensional	B-KEY
metrology	I-KEY
with	O
many	O
interconnections	O
.	O
This	O
paper	O
describes	O
an	O
approach	O
to	O
create	O
a	O
dimensional	B-KEY
metrology	I-KEY
knowledge	O
base	O
that	O
is	O
organized	O
around	O
a	O
set	O
of	O
key	O
concepts	O
and	O
to	O
represent	O
those	O
concepts	O
as	O
virtual	O
objects	O
that	O
can	O
be	O
navigated	O
with	O
interactive	O
,	O
computer	O
visualization	B-KEY
techniques	O
to	O
access	O
the	O
associated	O
knowledge	O
.	O
The	O
approach	O
can	O
enable	O
several	O
applications	O
.	O
First	O
is	O
the	O
application	O
to	O
convey	O
the	O
definition	O
and	O
meaning	O
of	O
GD&T	O
over	O
a	O
broad	O
range	O
of	O
tolerance	O
types	O
.	O
Second	O
is	O
the	O
application	O
to	O
provide	O
a	O
visualization	B-KEY
of	O
dimensional	B-KEY
metrology	I-KEY
knowledge	O
within	O
a	O
control	O
hierarchy	O
of	O
the	O
inspection	B-KEY
process	O
.	O
Third	O
is	O
the	O
application	O
to	O
show	O
the	O
coverage	O
of	O
interoperability	B-KEY
standards	I-KEY
to	O
enable	O
industry	O
to	O
make	O
decisions	O
on	O
standards	O
development	O
and	O
harmonization	O
efforts	O
.	O
A	O
prototype	O
system	O
has	O
been	O
implemented	O
to	O
demonstrate	O
the	O
principles	O
involved	O
in	O
the	O
approach	O

A	O
unified	O
view	O
for	O
vector	B-KEY
rotational	I-KEY
CORDIC	I-KEY
algorithms	I-KEY
and	O
architectures	O
based	O
on	O
angle	B-KEY
quantization	I-KEY
approach	O
Vector	O
rotation	O
is	O
the	O
key	O
operation	O
employed	O
extensively	O
in	O
many	O
digital	B-KEY
signal	I-KEY
processing	I-KEY
applications	I-KEY
.	O
In	O
this	O
paper	O
,	O
we	O
introduce	O
a	O
new	O
design	O
concept	O
called	O
Angle	B-KEY
Quantization	I-KEY
-LRB-	O
AQ	O
-RRB-	O
.	O
It	O
can	O
be	O
used	O
as	O
a	O
design	B-KEY
index	I-KEY
for	O
vector	B-KEY
rotational	I-KEY
operation	I-KEY
,	O
where	O
the	O
rotational	O
angle	O
is	O
known	O
in	O
advance	O
.	O
Based	O
on	O
the	O
AQ	O
process	O
,	O
we	O
establish	O
a	O
unified	B-KEY
design	I-KEY
framework	I-KEY
for	O
cost-effective	O
low-latency	B-KEY
rotational	I-KEY
algorithms	I-KEY
and	O
architectures	O
.	O
Several	O
existing	O
works	O
,	O
such	O
as	O
conventional	O
COordinate	O
Rotational	O
Digital	O
Computer	O
-LRB-	O
CORDIC	O
-RRB-	O
,	O
AR-CORDIC	O
,	O
MVR-CORDIC	O
,	O
and	O
EEAS-based	O
CORDIC	O
,	O
can	O
be	O
fitted	O
into	O
the	O
design	O
framework	O
,	O
forming	O
a	O
Vector	O
Rotational	O
CORDIC	O
Family	O
.	O
Moreover	O
,	O
we	O
address	O
four	O
searching	B-KEY
algorithms	I-KEY
to	O
solve	O
the	O
optimization	B-KEY
problem	I-KEY
encountered	O
in	O
the	O
proposed	O
vector	O
rotational	O
CORDIC	O
family	O
.	O
The	O
corresponding	O
scaling	B-KEY
operations	I-KEY
of	O
the	O
CORDIC	O
family	O
are	O
also	O
discussed	O
.	O
Based	O
on	O
the	O
new	O
design	O
framework	O
,	O
we	O
can	O
realize	O
high-speed/low-complexity	O
rotational	O
VLSI	O
circuits	O
,	O
whereas	O
without	O
degrading	O
the	O
precision	O
performance	O
in	O
fixed-point	O
implementations	O

The	O
Advanced	B-KEY
Encryption	I-KEY
Standard	I-KEY
-	O
implementation	O
and	O
transition	O
to	O
a	O
new	O
cryptographic	B-KEY
benchmark	I-KEY
Cryptography	O
is	O
the	O
science	O
of	O
coding	B-KEY
information	O
to	O
create	O
unintelligible	B-KEY
ciphers	I-KEY
that	O
conceal	O
or	O
hide	O
messages	O
.	O
The	O
process	O
that	O
achieves	O
this	O
goal	O
is	O
commonly	O
referred	O
to	O
as	O
encryption	O
.	O
Although	O
encryption	O
processes	O
of	O
various	O
forms	O
have	O
been	O
employed	O
for	O
centuries	O
to	O
protect	O
the	O
exchange	O
of	O
messages	O
,	O
the	O
advent	O
of	O
the	O
information	O
age	O
has	O
underscored	O
the	O
importance	O
of	O
strong	O
cryptography	O
as	O
a	O
process	O
to	O
secure	O
data	B-KEY
exchanged	I-KEY
through	O
electronic	O
means	O
,	O
and	O
has	O
accentuated	O
the	O
demand	O
for	O
products	O
offering	O
these	O
services	O
.	O
This	O
article	O
describes	O
the	O
process	O
that	O
has	O
led	O
to	O
the	O
development	O
of	O
the	O
latest	O
cryptographic	B-KEY
benchmark	I-KEY
;	O
the	O
Advanced	B-KEY
Encryption	I-KEY
Standard	I-KEY
-LRB-	O
AES	B-KEY
-RRB-	O
.	O
The	O
article	O
briefly	O
examines	O
the	O
requirements	O
set	O
forth	O
for	O
its	O
development	O
,	O
defines	O
how	O
the	O
new	O
standard	O
is	O
implemented	O
,	O
and	O
describes	O
how	O
government	B-KEY
,	O
business	B-KEY
,	O
and	O
industry	B-KEY
can	O
transition	O
to	O
AES	B-KEY
with	O
minimum	O
impact	O
to	O
operations	O

Software	O
vendors	O
'	O
failure	O
fuels	O
consolidation	B-KEY
theories	O
-LSB-	O
telecom	B-KEY
interconnection	I-KEY
and	O
billing	O
-RSB-	O
As	O
independent	O
software	O
vendors	O
like	O
AP	B-KEY
Engines	I-KEY
fall	O
by	O
the	O
wayside	O
as	O
independent	O
entities	O
,	O
attrition	O
could	O
accelerate	O
consolidation	B-KEY
in	O
the	O
OSS	O
space	O

Optimization	O
of	O
element-by-element	B-KEY
FEM	O
in	O
HPF	B-KEY
1.1	O
In	O
this	O
study	O
,	O
Poisson	O
's	O
equation	O
is	O
numerically	O
evaluated	O
by	O
the	O
element-by-element	B-KEY
-LRB-	O
EBE	O
-RRB-	O
finite-element	O
method	O
in	O
a	O
parallel	O
environment	O
using	O
HPF	B-KEY
1.1	O
-LRB-	O
High-Performance	O
Fortran	O
-RRB-	O
.	O
In	O
order	O
to	O
achieve	O
high	O
parallel	O
efficiency	O
,	O
the	O
data	O
structures	O
have	O
been	O
altered	O
to	O
node-based	O
data	O
instead	O
of	O
mixtures	O
of	O
node	O
-	O
and	O
element-based	O
data	O
,	O
representing	O
a	O
node-based	O
EBE	O
finite-element	O
scheme	O
-LRB-	O
nEBE	O
-RRB-	O
.	O
The	O
parallel	O
machine	O
used	O
in	O
this	O
study	O
was	O
the	O
NEC	O
SX-4	O
,	O
and	O
experiments	O
were	O
performed	O
on	O
a	O
single	O
node	O
having	O
32	O
processors	O
sharing	O
common	O
memory	O
.	O
The	O
HPF	B-KEY
compiler	O
used	O
in	O
the	O
experiments	O
is	O
HPF/SX	O
Rev	O
2.0	O
released	O
in	O
1997	O
-LRB-	O
unofficial	O
-RRB-	O
,	O
which	O
supports	O
HPF	O
1.1	O
.	O
Models	O
containing	O
approximately	O
200	O
000	O
and	O
1,500,000	O
degrees	O
of	O
freedom	O
were	O
analyzed	O
in	O
order	O
to	O
evaluate	O
the	O
method	O
.	O
The	O
calculation	O
time	O
,	O
parallel	O
efficiency	O
,	O
and	O
memory	O
used	O
were	O
compared	O
.	O
The	O
performance	O
of	O
HPF	B-KEY
in	O
the	O
conjugate	B-KEY
gradient	I-KEY
solver	I-KEY
for	O
the	O
large	O
model	O
,	O
using	O
the	O
NEC	O
SX-4	O
compiler	O
option-noshrunk	O
,	O
was	O
about	O
85	O
%	O
that	O
of	O
the	O
message	B-KEY
passing	I-KEY
interface	O

Enlisting	O
on-line	B-KEY
residents	I-KEY
:	O
Expanding	O
the	O
boundaries	O
of	O
e-government	B-KEY
in	O
a	O
Japanese	B-KEY
rural	I-KEY
township	I-KEY
The	O
purpose	O
of	O
this	O
article	O
is	O
to	O
analyze	O
and	O
learn	O
from	O
an	O
unusual	O
way	O
in	O
which	O
local	B-KEY
bureaucrats	I-KEY
in	O
a	O
Japanese	B-KEY
rural	I-KEY
township	I-KEY
are	O
using	O
the	O
Internet	B-KEY
to	O
serve	O
their	O
constituents	O
by	O
enlisting	O
the	O
support	O
of	O
``	O
on-line	B-KEY
residents	I-KEY
.	O
''	O
Successful	O
e-government	B-KEY
requires	O
not	O
only	O
rethinking	O
the	O
potential	O
uses	O
of	O
computer	O
technology	O
,	O
but	O
in	O
adopting	O
new	O
patterns	O
of	O
decision-making	B-KEY
,	O
power	B-KEY
sharing	I-KEY
,	O
and	O
office	B-KEY
management	I-KEY
that	O
many	O
bureaucrats	O
may	O
not	O
be	O
predisposed	O
to	O
make	O
.	O
The	O
main	O
thesis	O
of	O
this	O
article	O
is	O
that	O
necessity	O
and	O
practicality	O
can	O
play	O
a	O
powerful	O
motivational	O
role	O
in	O
facilitating	O
the	O
incorporation	O
of	O
information	O
technology	O
-LRB-	O
IT	O
-RRB-	O
at	O
the	O
level	O
of	O
local	O
government	O
.	O
This	O
case	O
study	O
of	O
how	O
bureaucrats	O
in	O
Towa-cho	B-KEY
,	O
a	O
small	O
,	O
agricultural	O
town	O
in	O
Northeastern	O
Japan	O
,	O
have	O
harnessed	O
the	O
Internet	B-KEY
demonstrates	O
clearly	O
the	O
fundamentals	O
of	O
building	O
a	O
successful	O
e-government	B-KEY
framework	O
in	O
this	O
rural	B-KEY
municipality	I-KEY
,	O
similar	O
to	O
many	O
communities	O
in	O
Europe	O
and	O
North	O
America	O
today	O

New	O
water	B-KEY
management	I-KEY
system	I-KEY
begins	O
operation	O
at	O
US	B-KEY
projects	I-KEY
The	O
US	B-KEY
Army	I-KEY
Corps	I-KEY
of	I-KEY
Engineers	I-KEY
has	O
developed	O
a	O
new	O
automated	B-KEY
information	I-KEY
system	I-KEY
to	O
support	O
its	O
water	B-KEY
control	I-KEY
management	I-KEY
mission	O
.	O
The	O
new	O
system	O
provides	O
a	O
variety	O
of	O
decision	B-KEY
support	I-KEY
tools	I-KEY
,	O
enabling	O
water	B-KEY
control	I-KEY
managers	I-KEY
to	O
acquire	O
,	O
transform	O
,	O
verify	O
,	O
store	O
,	O
display	O
,	O
analyse	O
,	O
and	O
disseminate	O
data	O
and	O
information	O
efficiently	O
and	O
around	O
the	O
clock	O

Creating	O
the	O
right	O
mail	B-KEY
model	O
If	O
you	O
know	O
your	O
post	B-KEY
room	I-KEY
is	O
not	O
as	O
efficiently	O
organised	O
as	O
it	O
might	O
be	O
,	O
but	O
you	O
are	O
not	O
sure	O
how	O
best	O
to	O
go	O
about	O
making	O
improvements	O
,	O
then	O
consider	O
this	O
advice	O
from	O
John	O
Edgar	O
of	O
consultant	B-KEY
MCS	B-KEY

Choice	B-KEY
preferences	I-KEY
without	O
inferences	O
:	O
subconscious	B-KEY
priming	I-KEY
of	O
risk	B-KEY
attitudes	I-KEY
We	O
present	O
a	O
procedure	O
for	O
subconscious	B-KEY
priming	I-KEY
of	O
risk	B-KEY
attitudes	I-KEY
.	O
In	O
Experiment	O
1	O
,	O
we	O
were	O
reliably	O
able	O
to	O
induce	O
risk-seeking	O
or	O
risk-averse	B-KEY
preferences	I-KEY
across	O
a	O
range	O
of	O
decision	B-KEY
scenarios	I-KEY
using	O
this	O
priming	O
procedure	O
.	O
In	O
Experiment	O
2	O
,	O
we	O
showed	O
that	O
these	O
priming	O
effects	O
can	O
be	O
reversed	O
by	O
drawing	O
participants	O
'	O
attention	O
to	O
the	O
priming	O
event	O
.	O
Our	O
results	O
support	O
claims	O
that	O
the	O
formation	O
of	O
risk	O
preferences	O
can	O
be	O
based	O
on	O
preconscious	B-KEY
processing	I-KEY
,	O
as	O
for	O
example	O
postulated	O
by	O
the	O
affective	B-KEY
primacy	I-KEY
hypothesis	I-KEY
,	O
rather	O
than	O
rely	O
on	O
deliberative	B-KEY
mental	I-KEY
operations	I-KEY
,	O
as	O
posited	O
by	O
several	O
current	O
models	O
of	O
judgment	O
and	O
decision	O
making	O

Help-desk	O
support	O
is	O
key	O
to	O
wireless	B-KEY
success	O
-LSB-	O
finance	B-KEY
-RSB-	O
A	O
well	O
thought	O
out	O
help	B-KEY
desk	I-KEY
can	O
make	O
or	O
break	O
an	O
institution	O
's	O
mobile	O
play	O
.	O
Schwab	B-KEY
,	O
Ameritrade	B-KEY
and	O
RBC	B-KEY
are	O
taking	O
their	O
support	O
function	O
seriously	O

Optimal	B-KEY
control	I-KEY
using	O
the	O
transport	B-KEY
equation	I-KEY
:	O
the	O
Liouville	B-KEY
machine	I-KEY
Transport	O
theory	O
describes	O
the	O
scattering	B-KEY
behavior	I-KEY
of	O
physical	B-KEY
particles	I-KEY
such	O
as	O
photons	O
.	O
Here	O
we	O
show	O
how	O
to	O
connect	O
this	O
theory	O
to	O
optimal	B-KEY
control	I-KEY
theory	O
and	O
to	O
adaptive	B-KEY
behavior	I-KEY
of	O
agents	O
embedded	O
in	O
an	O
environment	O
.	O
Environments	O
and	O
tasks	O
are	O
defined	O
by	O
physical	O
boundary	O
conditions	O
.	O
Given	O
some	O
task	O
,	O
we	O
compute	O
a	O
set	O
of	O
probability	O
densities	O
on	O
continuous	O
state	O
and	O
action	O
and	O
time	O
.	O
From	O
these	O
densities	O
we	O
derive	O
an	O
optimal	O
policy	O
such	O
that	O
for	O
all	O
states	O
the	O
most	O
likely	O
action	O
maximizes	O
the	O
probability	O
of	O
reaching	O
a	O
predefined	O
goal	O
state	O
.	O
Liouville	O
's	O
conservation	O
theorem	O
tells	O
us	O
that	O
the	O
conditional	O
density	O
at	O
time	O
t	O
,	O
state	O
s	O
,	O
and	O
action	O
a	O
must	O
equal	O
the	O
density	O
at	O
t	O
+	O
dt	O
,	O
s	O
+	O
ds	O
,	O
a	O
+	O
da	O
.	O
Discretization	O
yields	O
a	O
linear	O
system	O
that	O
can	O
be	O
solved	O
directly	O
and	O
whose	O
solution	O
corresponds	O
to	O
an	O
optimal	O
policy	O
.	O
Discounted	O
reward	O
schemes	O
are	O
incorporated	O
naturally	O
by	O
taking	O
the	O
Laplace	O
transform	O
of	O
the	O
equations	O
.	O
The	O
Liouville	B-KEY
machine	I-KEY
quickly	O
solves	O
rather	O
complex	O
maze	O
problems	O

Williams	O
nears	O
end	O
of	O
Chapter	O
11	O
-LSB-	O
telecom	O
-RSB-	O
Leucadia	B-KEY
National	I-KEY
Corp.	I-KEY
comes	O
through	O
with	O
a	O
$	O
330	O
million	O
boost	O
for	O
Williams	O
Communications	B-KEY
,	I-KEY
which	B-KEY
should	O
keep	O
the	O
carrier	O
afloat	O
through	O
the	O
remainder	O
of	O
its	O
bankruptcy	O

A	O
novel	O
approach	O
for	O
the	O
detection	O
of	O
pathlines	O
in	O
X-ray	B-KEY
angiograms	I-KEY
:	O
the	O
wavefront	B-KEY
propagation	I-KEY
algorithm	I-KEY
Presents	O
a	O
new	O
pathline	O
approach	O
,	O
based	O
on	O
the	O
wavefront	B-KEY
propagation	I-KEY
principle	I-KEY
,	O
and	O
developed	O
in	O
order	O
to	O
reduce	O
the	O
variability	O
in	O
the	O
outcomes	O
of	O
the	O
quantitative	B-KEY
coronary	I-KEY
artery	I-KEY
analysis	I-KEY
.	O
This	O
novel	O
approach	O
,	O
called	O
wavepath	O
,	O
reduces	O
the	O
influence	O
of	O
the	O
user-defined	O
start	O
-	O
and	O
endpoints	O
of	O
the	O
vessel	B-KEY
segment	I-KEY
and	O
is	O
therefore	O
more	O
robust	O
and	O
improves	O
the	O
reproducibility	O
of	O
the	O
lesion	B-KEY
quantification	I-KEY
substantially	O
.	O
The	O
validation	O
study	O
shows	O
that	O
the	O
wavepath	B-KEY
method	I-KEY
is	O
totally	O
constant	O
in	O
the	O
middle	O
part	O
of	O
the	O
pathline	O
,	O
even	O
when	O
using	O
the	O
method	O
for	O
constructing	O
a	O
bifurcation	B-KEY
or	O
sidebranch	B-KEY
pathline	I-KEY
.	O
Furthermore	O
,	O
the	O
number	O
of	O
corrections	B-KEY
needed	O
to	O
guide	O
the	O
wavepath	O
through	O
the	O
correct	B-KEY
vessel	O
is	O
decreased	O
from	O
an	O
average	O
of	O
0.44	O
corrections	O
per	O
pathline	O
to	O
an	O
average	O
of	O
0.12	O
per	O
pathline	O
.	O
Therefore	O
,	O
it	O
can	O
be	O
concluded	O
that	O
the	O
wavepath	O
algorithm	O
improves	O
the	O
overall	O
analysis	O
substantially	O

High-speed	B-KEY
CMOS	I-KEY
circuits	I-KEY
with	O
parallel	B-KEY
dynamic	I-KEY
logic	I-KEY
and	O
speed-enhanced	B-KEY
skewed	I-KEY
static	I-KEY
logic	I-KEY
In	O
this	O
paper	O
,	O
we	O
describe	O
parallel	B-KEY
dynamic	I-KEY
logic	I-KEY
-LRB-	O
PDL	O
-RRB-	O
which	O
exhibits	O
high	O
speed	O
without	O
charge	O
sharing	O
problem	O
.	O
PDL	O
uses	O
only	O
parallel-connected	B-KEY
transistors	I-KEY
for	O
fast	O
logic	O
evaluation	O
and	O
is	O
a	O
good	O
candidate	O
for	O
high-speed	O
low-voltage	B-KEY
operation	I-KEY
.	O
It	O
has	O
less	O
back-bias	B-KEY
effect	I-KEY
compared	O
to	O
other	O
logic	O
styles	O
,	O
which	O
use	O
stacked	B-KEY
transistors	I-KEY
.	O
Furthermore	O
,	O
PDL	O
needs	O
no	O
signal	O
ordering	O
or	O
tapering	O
.	O
PDL	O
with	O
speed-enhanced	B-KEY
skewed	I-KEY
static	I-KEY
logic	I-KEY
renders	O
straightforward	O
logic	B-KEY
synthesis	I-KEY
without	O
the	O
usual	O
area	O
penalty	O
due	O
to	O
logic	O
duplication	O
.	O
Our	O
experimental	O
results	O
on	O
two	O
32-bit	O
carry	B-KEY
lookahead	I-KEY
adders	I-KEY
using	O
0.25	O
-	O
mu	O
m	O
CMOS	O
technology	O
show	O
that	O
PDL	O
with	O
speed-enhanced	O
skewed	O
static	O
-LRB-	O
SSS	O
-RRB-	O
look	O
reduces	O
the	O
delay	B-KEY
over	O
clock-delayed	O
-LRB-	O
CD	O
-RRB-	O
-	O
domino	O
by	O
15	O
%	O
-27	O
%	O
and	O
the	O
power-delay	O
product	O
by	O
20	O
%	O
-37	O
%	O

The	O
contiguity	B-KEY
in	O
R/M	O
An	O
r.e.	O
degree	O
c	O
is	O
contiguous	B-KEY
if	O
deg/sub	O
wtt	O
/	O
-LRB-	O
A	O
-RRB-	O
=	O
deg/sub	O
wtt	O
/	O
-LRB-	O
B	O
-RRB-	O
for	O
any	O
r.e.	O
sets	O
A	O
,	O
B	O
in	O
c	O
.	O
In	O
this	O
paper	O
,	O
we	O
generalize	O
the	O
notation	O
of	O
contiguity	B-KEY
to	O
the	O
structure	O
R/M	O
,	O
the	O
upper	B-KEY
semilattice	I-KEY
of	O
the	O
r.e.	O
degree	O
set	O
R	O
modulo	O
the	O
cappable	O
r.e.	O
degree	O
set	O
M	O
.	O
An	O
element	O
-LSB-	O
c	O
-RSB-	O
in	O
R/M	O
is	O
contiguous	B-KEY
if	O
-LSB-	O
deg/sub	O
wtt	O
/	O
-LRB-	O
A	O
-RRB-	O
-RSB-	O
=	O
-LSB-	O
deg/sub	O
wtt	O
/	O
-LRB-	O
B	O
-RRB-	O
-RSB-	O
for	O
any	O
r.e.	O
sets	O
A	O
,	O
B	O
such	O
that	O
deg/sub	O
T	O
/	O
-LRB-	O
A	O
-RRB-	O
,	O
deg/sub	O
T	O
/	O
-LRB-	O
B	O
-RRB-	O
in	O
-LSB-	O
c	O
-RSB-	O
.	O
It	O
is	O
proved	O
in	O
this	O
paper	O
that	O
every	O
nonzero	B-KEY
element	I-KEY
in	O
R/M	O
is	O
not	O
contiguous	B-KEY
,	O
i.e.	O
,	O
for	O
every	O
element	O
-LSB-	O
c	O
-RSB-	O
in	O
R/M	O
,	O
if	O
-LSB-	O
c	O
-RSB-	O
not	O
=	O
-LSB-	O
o	O
-RSB-	O
then	O
there	O
exist	O
at	O
least	O
two	O
r.e.	O
sets	O
A	O
,	O
B	O
such	O
that	O
deg/sub	O
T	O
/	O
-LRB-	O
A	O
-RRB-	O
,	O
deg/sub	O
T	O
/	O
-LRB-	O
B	O
-RRB-	O
in	O
-LSB-	O
c	O
-RSB-	O
and	O
-LSB-	O
deg/sub	O
wtt	O
/	O
-LRB-	O
A	O
-RRB-	O
-RSB-	O
not	O
=	O
-LSB-	O
deg/sub	O
wtt	O
/	O
-LRB-	O
B	O
-RRB-	O
-RSB-	O

Products	O
and	O
polymorphic	B-KEY
subtypes	I-KEY
This	O
paper	O
is	O
devoted	O
to	O
a	O
comprehensive	O
study	O
of	O
polymorphic	B-KEY
subtypes	I-KEY
with	O
products	O
.	O
We	O
first	O
present	O
a	O
sound	O
and	O
complete	O
Hilbert	B-KEY
style	I-KEY
axiomatization	I-KEY
of	O
the	O
relation	O
of	O
being	O
a	O
subtype	O
in	O
presence	O
of	O
to	O
,	O
*	O
type	O
constructors	O
and	O
the	O
For	O
all	O
quantifier	O
,	O
and	O
we	O
show	O
that	O
such	O
axiornatization	O
is	O
not	O
encodable	O
in	O
the	O
system	O
with	O
to	O
,	O
For	O
all	O
only	O
.	O
In	O
order	O
to	O
give	O
a	O
logical	B-KEY
semantics	I-KEY
to	O
such	O
a	O
subtyping	O
relation	O
,	O
we	O
propose	O
a	O
new	O
form	O
of	O
a	O
sequent	O
which	O
plays	O
a	O
key	O
role	O
in	O
a	O
natural	O
deduction	O
and	O
a	O
Gentzen	B-KEY
style	I-KEY
calculi	I-KEY
.	O
Interestingly	O
enough	O
,	O
the	O
sequent	O
must	O
have	O
the	O
form	O
E	O
implies	O
T	O
,	O
where	O
E	O
is	O
a	O
non-commutative	O
,	O
non-empty	O
sequence	O
of	O
typing	O
assumptions	O
and	O
T	O
is	O
a	O
finite	B-KEY
binary	I-KEY
tree	I-KEY
of	O
typing	O
judgements	O
,	O
each	O
of	O
them	O
behaving	O
like	O
a	O
pushdown	B-KEY
store	I-KEY
.	O
We	O
study	O
basic	O
metamathematical	O
properties	O
of	O
the	O
two	O
logical	O
systems	O
,	O
such	O
as	O
subject	O
reduction	O
and	O
cut	O
elimination	O
.	O
Some	O
decidability/undecidability	O
issues	O
related	O
to	O
the	O
presented	O
subtyping	O
relation	O
are	O
also	O
explored	O
:	O
as	O
expected	O
,	O
the	O
subtyping	O
over	O
to	O
,	O
*	O
,	O
For	O
all	O
is	O
undecidable	O
,	O
being	O
already	O
undecidable	O
for	O
the	O
to	O
,	O
For	O
all	O
fragment	O
-LRB-	O
as	O
proved	O
in	O
-LSB-	O
15	O
-RSB-	O
-RRB-	O
,	O
but	O
for	O
the	O
*	O
,	O
For	O
all	O
fragment	O
it	O
turns	O
out	O
to	O
be	O
decidable	B-KEY

Experimental	O
feedforward	O
and	O
feedback	O
control	O
of	O
a	O
one-dimensional	O
SMA	B-KEY
composite	O
The	O
control	O
of	O
embedded	B-KEY
shape	I-KEY
memory	I-KEY
alloy	I-KEY
-LRB-	O
SMA	B-KEY
-RRB-	O
actuators	O
has	O
recently	O
become	O
a	O
topic	O
of	O
interest	O
in	O
the	O
field	O
of	O
smart	B-KEY
structures	I-KEY
.	O
The	O
inherent	O
difficulties	O
associated	O
with	O
SMA	B-KEY
actuators	O
has	O
resulted	O
in	O
a	O
variety	O
of	O
approaches	O
.	O
Homogenization	B-KEY
provides	O
a	O
simplified	O
,	O
yet	O
mathematically	O
rigorous	O
,	O
method	O
of	O
determining	O
average	O
stress	O
and	O
strain	O
fields	O
in	O
a	O
composite	O
.	O
A	O
modified	O
constitutive	O
model	B-KEY
is	O
presented	O
based	O
on	O
experimental	O
results	O
demonstrating	O
the	O
inability	O
of	O
most	O
simple	O
phenomenological	O
models	B-KEY
to	O
capture	O
the	O
effective	O
behavior	O
of	O
SMAs	B-KEY
during	O
thermal	B-KEY
activation	I-KEY
.	O
A	O
feedforward	O
controller	O
is	O
presented	O
for	O
a	O
SMA	B-KEY
composite	O
based	O
on	O
the	O
homogenization	B-KEY
of	O
a	O
modified	O
phenomenological	O
model	B-KEY
for	O
SMAs	B-KEY
in	O
a	O
linear	B-KEY
matrix	I-KEY

Lob	O
's	O
theorem	O
as	O
a	O
limitation	B-KEY
on	I-KEY
mechanism	I-KEY
We	O
argue	O
that	O
Lob	O
's	O
Theorem	O
implies	O
a	O
limitation	B-KEY
on	I-KEY
mechanism	I-KEY
.	O
Specifically	O
,	O
we	O
argue	O
,	O
via	O
an	O
application	O
of	O
a	O
generalized	O
version	O
of	O
Lob	O
's	O
Theorem	O
,	O
that	O
any	O
particular	O
device	O
known	O
by	O
an	O
observer	O
to	O
be	O
mechanical	O
can	O
not	O
be	O
used	O
as	O
an	O
epistemic	B-KEY
authority	I-KEY
-LRB-	O
of	O
a	O
particular	O
type	O
-RRB-	O
by	O
that	O
observer	O
:	O
either	O
the	O
belief-set	B-KEY
of	O
such	O
an	O
authority	O
is	O
not	O
mechanizable	O
or	O
,	O
if	O
it	O
is	O
,	O
there	O
is	O
no	O
identifiable	O
formal	B-KEY
system	I-KEY
of	O
which	O
the	O
observer	O
can	O
know	O
-LRB-	O
or	O
truly	O
believe	O
-RRB-	O
it	O
to	O
be	O
the	O
theorem-set	B-KEY
.	O
This	O
gives	O
,	O
we	O
believe	O
,	O
an	O
important	O
and	O
hitherto	O
unnoticed	O
connection	O
between	O
mechanism	O
and	O
the	O
use	O
of	O
authorities	O
by	O
human-like	B-KEY
epistemic	I-KEY
agents	I-KEY

Mobile	O
banking	B-KEY
's	O
tough	O
sell	O
Banks	B-KEY
are	O
having	O
to	O
put	O
their	O
mobile-commerce	B-KEY
projects	O
on	O
hold	O
because	O
the	O
essential	O
technology	O
to	O
make	O
the	O
services	O
usable	O
,	O
in	O
particular	O
GPRS	B-KEY
-LRB-	O
general	O
packet	O
radio	O
service	O
-RRB-	O
has	O
n't	O
become	O
widely	O
available	O
.	O
It	O
is	O
estimated	O
that	O
by	O
the	O
end	O
of	O
2002	O
,	O
only	O
5	O
per	O
cent	O
of	O
adults	O
will	O
have	O
GPRS	B-KEY
phones	O
.	O
This	O
will	O
have	O
a	O
knock-on	O
effect	O
for	O
other	O
technologies	O
such	O
as	O
clickable	O
icons	O
and	O
multimedia	O
messaging	O
.	O
In	O
fact	O
banking	B-KEY
via	O
WAP	O
-LRB-	O
wireless	B-KEY
application	I-KEY
protocol	I-KEY
-RRB-	O
has	O
proved	O
to	O
be	O
a	O
frustrating	O
and	O
time-consuming	O
process	O
for	O
the	O
customer	O
.	O
Financial	O
firms	O
'	O
hopes	O
for	O
higher	O
mobile	O
usage	O
are	O
stymied	O
by	O
the	O
fact	O
that	O
improvements	O
to	O
the	O
systems	O
wo	O
n't	O
happen	O
as	O
fast	O
as	O
they	O
want	O
and	O
the	O
inadequacies	O
of	O
the	O
system	O
go	O
beyond	O
immature	O
technology	O
.	O
Financial	O
services	O
institutions	O
should	O
not	O
wait	O
for	O
customers	O
to	O
become	O
au	O
fait	O
with	O
their	O
WAP	O
.	O
Instead	O
they	O
should	O
be	O
the	O
ones	O
``	O
driving	O
the	O
traffic	O
''	O

Mammogram	B-KEY
synthesis	I-KEY
using	O
a	O
3D	B-KEY
simulation	I-KEY
.	O
I.	O
Breast	B-KEY
tissue	I-KEY
model	I-KEY
and	O
image	B-KEY
acquisition	I-KEY
simulation	I-KEY
A	O
method	O
is	O
proposed	O
for	O
generating	O
synthetic	O
mammograms	O
based	O
upon	O
simulations	O
of	O
breast	O
tissue	O
and	O
the	O
mammographic	O
imaging	O
process	O
.	O
A	O
computer	B-KEY
breast	I-KEY
model	I-KEY
has	O
been	O
designed	O
with	O
a	O
realistic	O
distribution	O
of	O
large	O
and	O
medium	O
scale	O
tissue	O
structures	O
.	O
Parameters	O
controlling	O
the	O
size	O
and	O
placement	O
of	O
simulated	O
structures	O
-LRB-	O
adipose	B-KEY
compartments	I-KEY
and	O
ducts	B-KEY
-RRB-	O
provide	O
a	O
method	O
for	O
consistently	O
modeling	O
images	O
of	O
the	O
same	O
simulated	O
breast	O
with	O
modified	O
position	O
or	O
acquisition	O
parameters	O
.	O
The	O
mammographic	O
imaging	O
process	O
is	O
simulated	O
using	O
a	O
compression	O
model	O
and	O
a	O
model	O
of	O
the	O
X-ray	B-KEY
image	I-KEY
acquisition	I-KEY
process	O
.	O
The	O
compression	O
model	O
estimates	O
breast	O
deformation	O
using	O
tissue	B-KEY
elasticity	I-KEY
parameters	I-KEY
found	O
in	O
the	O
literature	O
and	O
clinical	O
force	B-KEY
values	I-KEY
.	O
The	O
synthetic	O
mammograms	O
were	O
generated	O
by	O
a	O
mammogram	O
acquisition	O
model	O
using	O
a	O
monoenergetic	B-KEY
parallel	I-KEY
beam	I-KEY
approximation	I-KEY
applied	O
to	O
the	O
synthetically	O
compressed	O
breast	O
phantom	O

Speaker	B-KEY
identification	I-KEY
from	O
voice	O
using	O
neural	O
networks	O
The	O
paper	O
provides	O
three	O
different	O
schemes	O
for	O
speaker	B-KEY
identification	I-KEY
of	O
personnel	B-KEY
from	O
their	O
voice	O
using	O
artificial	B-KEY
neural	I-KEY
networks	I-KEY
.	O
The	O
first	O
scheme	O
recognizes	O
speakers	O
by	O
employing	O
the	O
classical	O
backpropagation	B-KEY
algorithm	I-KEY
pre-trained	B-KEY
with	O
known	B-KEY
voice	I-KEY
samples	I-KEY
of	O
the	O
persons	O
.	O
The	O
second	O
scheme	O
provides	O
a	O
framework	O
for	O
classifying	O
the	O
known	O
training	O
samples	O
of	O
the	O
voice	O
features	O
using	O
a	O
hierarchical	B-KEY
architecture	I-KEY
realized	O
with	O
a	O
self-organizing	B-KEY
feature	I-KEY
map	I-KEY
neural	O
net	O
.	O
The	O
first	O
scheme	O
is	O
highly	O
robust	O
as	O
it	O
is	O
capable	O
of	O
identifying	O
the	O
personnel	B-KEY
from	O
their	O
noisy	O
voice	O
samples	O
,	O
but	O
because	O
of	O
its	O
excessive	O
training	O
time	O
it	O
has	O
limited	O
applications	O
for	O
a	O
large	O
voice	O
database	O
.	O
The	O
second	O
scheme	O
though	O
not	O
so	O
robust	O
as	O
the	O
former	O
,	O
however	O
,	O
can	O
classify	O
an	O
unknown	O
voice	O
sample	O
to	O
its	O
nearest	O
class	O
.	O
The	O
time	O
needed	O
for	O
classification	B-KEY
by	O
the	O
first	O
scheme	O
is	O
always	O
unique	O
irrespective	O
of	O
the	O
voice	O
sample	O
.	O
It	O
is	O
proportional	O
to	O
the	O
number	O
of	O
feedforward	B-KEY
layers	I-KEY
in	O
the	O
network	O
.	O
The	O
time-requirement	O
of	O
the	O
second	O
classification	B-KEY
scheme	O
,	O
however	O
,	O
is	O
not	O
free	O
from	O
the	O
voice	O
features	O
and	O
is	O
proportional	O
to	O
the	O
number	O
of	O
2D	B-KEY
arrays	I-KEY
traversed	O
by	O
the	O
algorithm	O
on	O
the	O
hierarchical	O
structure	O
.	O
The	O
third	O
scheme	O
is	O
highly	O
robust	O
and	O
mis-classification	O
is	O
as	O
low	O
as	O
0.2	O
per	O
cent	O
.	O
The	O
third	O
scheme	O
combines	O
the	O
composite	O
benefits	O
of	O
a	O
radial	B-KEY
basis	I-KEY
function	I-KEY
neural	I-KEY
net	I-KEY
and	O
backpropagation	O
trained	O
neural	O
net	O

Web	O
services	O
boost	O
integration	O
Microsoft	B-KEY
and	O
IBM	B-KEY
have	O
announced	O
products	O
to	O
help	O
their	O
database	B-KEY
software	I-KEY
co-exist	O
with	O
competitors	O
'	O
offerings	O
.	O
The	O
products	O
use	O
web	B-KEY
services	I-KEY
technology	I-KEY
allowing	O
users	O
to	O
improve	O
integration	O
between	O
databases	O
and	O
application	O
software	O
from	O
rival	O
vendors	O

Analytic	O
PCA	O
construction	O
for	O
theoretical	O
analysis	O
of	O
lighting	B-KEY
variability	I-KEY
in	O
images	O
of	O
a	O
Lambertian	O
object	O
We	O
analyze	O
theoretically	O
the	O
subspace	O
best	O
approximating	O
images	O
of	O
a	O
convex	B-KEY
Lambertian	I-KEY
object	I-KEY
taken	O
from	O
the	O
same	O
viewpoint	O
,	O
but	O
under	O
different	O
distant	O
illumination	O
conditions	O
.	O
We	O
analytically	O
construct	O
the	O
principal	O
component	O
analysis	O
for	O
images	O
of	O
a	O
convex	B-KEY
Lambertian	I-KEY
object	I-KEY
,	O
explicitly	O
taking	O
attached	O
shadows	O
into	O
account	O
,	O
and	O
find	O
the	O
principal	B-KEY
eigenmodes	I-KEY
and	O
eigenvalues	O
with	O
respect	O
to	O
lighting	B-KEY
variability	I-KEY
.	O
Our	O
analysis	O
makes	O
use	O
of	O
an	O
analytic	O
formula	O
for	O
the	O
irradiance	B-KEY
in	O
terms	O
of	O
spherical-harmonic	O
coefficients	O
of	O
the	O
illumination	O
and	O
shows	O
,	O
under	O
appropriate	O
assumptions	O
,	O
that	O
the	O
principal	O
components	O
or	O
eigenvectors	O
are	O
identical	O
to	O
the	O
spherical	B-KEY
harmonic	I-KEY
basis	O
functions	O
evaluated	O
at	O
the	O
surface	B-KEY
normal	I-KEY
vectors	I-KEY
.	O
Our	O
main	O
contribution	O
is	O
in	O
extending	O
these	O
results	O
to	O
the	O
single-viewpoint	O
case	O
,	O
showing	O
how	O
the	O
principal	B-KEY
eigenmodes	I-KEY
and	O
eigenvalues	O
are	O
affected	O
when	O
only	O
a	O
limited	O
subset	O
-LRB-	O
the	O
upper	O
hemisphere	O
-RRB-	O
of	O
normals	O
is	O
available	O
and	O
the	O
spherical	B-KEY
harmonics	I-KEY
are	O
no	O
longer	O
orthonormal	O
over	O
the	O
restricted	O
domain	O
.	O
Our	O
results	O
are	O
very	O
close	O
,	O
both	O
qualitatively	O
and	O
quantitatively	O
,	O
to	O
previous	O
empirical	O
observations	O
and	O
represent	O
the	O
first	O
essentially	O
complete	O
theoretical	O
explanation	O
of	O
these	O
observations	O

Expert	O
advice	O
-	O
how	O
can	O
my	O
organisation	O
take	O
advantage	O
of	O
reverse	B-KEY
auctions	I-KEY
without	O
jeopardising	O
existing	O
supplier	B-KEY
relationships	I-KEY
?	O
In	O
a	O
recent	O
survey	O
,	O
AMR	O
Research	O
found	O
that	O
companies	O
that	O
use	O
reverse	B-KEY
auctions	I-KEY
to	O
negotiate	O
prices	O
with	O
suppliers	O
typically	O
achieve	O
savings	O
of	O
between	O
10	O
%	O
and	O
15	O
%	O
on	O
direct	O
goods	O
and	O
between	O
20	O
%	O
and	O
25	O
%	O
on	O
indirect	O
goods	O
,	O
and	O
can	O
slash	O
sourcing	O
cycle	O
times	O
from	O
months	O
to	O
weeks	O
.	O
Suppliers	O
,	O
however	O
,	O
are	O
less	O
enthusiastic	O
.	O
They	O
believe	O
that	O
these	O
savings	O
are	O
achieved	O
only	O
by	O
stripping	O
the	O
human	O
element	O
out	O
of	O
negotiations	O
and	O
evaluating	O
bids	O
on	O
price	O
alone	O
,	O
which	O
drives	O
down	O
their	O
profit	O
margins	O
.	O
As	O
a	O
result	O
,	O
reverse	B-KEY
auctions	I-KEY
carry	O
the	O
risk	O
of	O
jeopardising	O
long-term	O
and	O
trusted	O
relationships	O
.	O
Suppliers	O
that	O
have	O
not	O
been	O
involved	O
in	O
a	O
reverse	B-KEY
auction	I-KEY
before	O
typically	O
fear	O
the	O
bidding	O
event	O
itself	O
-	O
arguably	O
the	O
most	O
theatrical	O
and	O
,	O
therefore	O
,	O
most	O
hyped-up	O
part	O
of	O
the	O
process	O
.	O
Although	O
it	O
may	O
only	O
last	O
one	O
hour	O
,	O
weeks	O
of	O
preparation	B-KEY
go	O
into	O
setting	O
up	O
a	O
successful	O
bidding	O
event	O

Multi-agent	B-KEY
collaboration	I-KEY
for	O
B2B	B-KEY
workflow	I-KEY
monitoring	I-KEY
Business-to-business	O
-LRB-	O
B2B	O
-RRB-	O
application	O
environments	O
are	O
exceedingly	O
dynamic	O
and	O
competitive	O
.	O
This	O
dynamism	O
is	O
manifested	O
in	O
the	O
form	O
of	O
changing	B-KEY
process	I-KEY
requirements	I-KEY
and	O
time	B-KEY
constraints	I-KEY
.	O
However	O
,	O
current	O
workflow	B-KEY
management	I-KEY
technologies	O
have	O
difficulties	O
trying	O
to	O
solve	O
problems	O
,	O
such	O
as	O
:	O
how	O
to	O
deal	O
with	O
the	O
dynamic	O
nature	O
of	O
B2B	O
commerce	O
processes	O
,	O
how	O
to	O
manage	O
the	O
distributed	O
knowledge	O
and	O
recourses	O
,	O
and	O
how	O
to	O
reduce	O
the	O
transaction	B-KEY
risk	I-KEY
.	O
In	O
this	O
paper	O
,	O
a	O
collaborative	O
multi-agent	O
system	O
is	O
proposed	O
.	O
Multiple	O
intelligent	O
agents	O
in	O
our	O
system	O
can	O
work	O
together	O
not	O
only	O
to	O
identify	O
the	O
workflow	O
problems	O
,	O
but	O
also	O
to	O
solve	O
such	O
problems	O
,	O
by	O
applying	O
business	B-KEY
rules	I-KEY
,	O
such	O
as	O
re-organizing	O
the	O
procurement	O
and	O
the	O
transaction	O
processes	O
,	O
and	O
making	O
necessary	O
workflow	O
process	O
changes	O

Exploiting	O
randomness	B-KEY
in	O
quantum	B-KEY
information	I-KEY
processing	I-KEY
We	O
consider	O
how	O
randomness	B-KEY
can	O
be	O
made	O
to	O
play	O
a	O
useful	O
role	O
in	O
quantum	O
information	O
processing-in	O
particular	O
,	O
for	O
decoherence	B-KEY
control	I-KEY
and	O
the	O
implementation	O
of	O
quantum	B-KEY
algorithms	I-KEY
.	O
For	O
a	O
two-level	B-KEY
system	I-KEY
in	O
which	O
the	O
decoherence	O
channel	O
is	O
non-dissipative	O
,	O
we	O
show	O
that	O
decoherence	O
suppression	O
is	O
possible	O
if	O
memory	O
is	O
present	O
in	O
the	O
channel	O
.	O
Random	B-KEY
switching	O
between	O
two	O
potentially	O
harmful	O
noise	O
sources	O
can	O
then	O
provide	O
a	O
source	O
of	O
stochastic	O
control	O
.	O
Such	O
random	B-KEY
switching	O
can	O
also	O
be	O
used	O
in	O
an	O
advantageous	O
way	O
for	O
the	O
implementation	O
of	O
quantum	O
algorithms	O

OS	B-KEY
porting	I-KEY
and	O
application	B-KEY
development	I-KEY
for	O
SoC	O
To	O
deliver	O
improved	O
usability	O
in	O
high-end	O
portable	O
consumer	O
products	O
,	O
the	O
use	O
of	O
an	O
appropriate	O
consumer	B-KEY
operating	I-KEY
system	I-KEY
-LRB-	O
OS	O
-RRB-	O
is	O
becoming	O
far	O
more	O
widespread	O
.	O
Using	O
a	O
commercially	O
supported	O
OS	O
also	O
vastly	O
increases	O
the	O
availability	O
of	O
supported	O
applications	O
.	O
For	O
the	O
device	O
developer	O
,	O
this	O
trend	O
adds	O
major	O
complexity	O
to	O
the	O
problem	O
of	O
system	O
implementation	O
.	O
Porting	O
a	O
complete	O
operating	O
system	O
to	O
a	O
new	O
hardware	B-KEY
design	I-KEY
adds	O
significantly	O
to	O
the	O
development	O
burden	O
,	O
increasing	O
both	O
time-to-market	O
and	O
expense	O
.	O
Even	O
for	O
those	O
familiar	O
with	O
the	O
integration	O
of	O
a	O
real-time	O
OS	O
,	O
the	O
porting	O
,	O
validation	O
and	O
support	O
of	O
a	O
complex	O
platform	O
OS	O
is	O
a	O
formidable	O
task	O

A	O
study	O
on	O
an	O
automatic	B-KEY
seam	I-KEY
tracking	I-KEY
system	I-KEY
by	O
using	O
an	O
electromagnetic	B-KEY
sensor	I-KEY
for	O
sheet	B-KEY
metal	I-KEY
arc	I-KEY
welding	I-KEY
of	O
butt	B-KEY
joints	I-KEY
Many	O
sensors	O
,	O
such	O
as	O
the	O
vision	O
sensor	O
and	O
the	O
laser	O
displacement	O
sensor	O
,	O
have	O
been	O
developed	O
to	O
automate	O
the	O
arc	O
welding	O
process	O
.	O
However	O
,	O
these	O
sensors	O
have	O
some	O
problems	O
due	O
to	O
the	O
effects	O
of	O
arc	O
light	O
,	O
fumes	O
and	O
spatter	O
.	O
An	O
electromagnetic	B-KEY
sensor	I-KEY
,	O
which	O
utilizes	O
the	O
generation	O
of	O
an	O
eddy	O
current	O
,	O
was	O
developed	O
for	O
detecting	O
the	O
weld	O
line	O
of	O
a	O
butt	B-KEY
joint	I-KEY
in	O
which	O
the	O
root	B-KEY
gap	I-KEY
size	I-KEY
was	O
zero	O
.	O
An	O
automatic	B-KEY
seam	I-KEY
tracking	I-KEY
system	I-KEY
designed	O
for	O
sheet	B-KEY
metal	I-KEY
arc	I-KEY
welding	I-KEY
was	O
constructed	O
with	O
a	O
sensor	O
.	O
Through	O
experiments	O
,	O
it	O
was	O
revealed	O
that	O
the	O
system	O
had	O
an	O
excellent	O
seam	B-KEY
tracking	I-KEY
accuracy	I-KEY
of	O
the	O
order	O
of	O
+	O
or-0	O
.2	O
mm	O

Action	B-KEY
aggregation	I-KEY
and	O
defuzzification	B-KEY
in	O
Mamdani-type	B-KEY
fuzzy	I-KEY
systems	I-KEY
Discusses	O
the	O
issues	O
of	O
action	B-KEY
aggregation	I-KEY
and	O
defuzzification	B-KEY
in	O
Mamdani-type	B-KEY
fuzzy	I-KEY
systems	I-KEY
.	O
The	O
paper	O
highlights	O
the	O
shortcomings	O
of	O
defuzzification	B-KEY
techniques	O
associated	O
with	O
the	O
customary	O
interpretation	O
of	O
the	O
sentence	B-KEY
connective	I-KEY
`	O
and	O
'	O
by	O
means	O
of	O
the	O
set	B-KEY
union	I-KEY
operation	I-KEY
.	O
These	O
include	O
loss	O
of	O
smoothness	O
of	O
the	O
output	O
characteristic	O
and	O
inaccurate	O
mapping	O
of	O
the	O
fuzzy	B-KEY
response	I-KEY
.	O
The	O
most	O
appropriate	O
procedure	O
for	O
aggregating	O
the	O
outputs	O
of	O
different	O
fuzzy	B-KEY
rules	I-KEY
and	O
converting	O
them	O
into	O
crisp	B-KEY
signals	I-KEY
is	O
then	O
suggested	O
.	O
The	O
advantages	O
in	O
terms	O
of	O
increased	O
transparency	B-KEY
and	O
mapping	B-KEY
accuracy	I-KEY
of	O
the	O
fuzzy	B-KEY
response	I-KEY
are	O
demonstrated	O

Syndicators	O
turn	O
to	O
the	O
enterprise	O
Syndicators	O
have	O
started	O
reshaping	O
offerings	O
,	O
products	O
,	O
and	O
services	O
towards	O
the	O
marketplace	O
that	O
was	O
looking	O
for	O
enterprise-wide	B-KEY
content	I-KEY
syndication	I-KEY
technology	I-KEY
and	O
service	O
.	O
Syndication	O
companies	O
are	O
turning	O
themselves	O
into	O
infrastructure	B-KEY
companies	I-KEY
.	O
Many	O
syndication	O
companies	O
are	O
now	O
focusing	O
their	O
efforts	O
on	O
enterprise	B-KEY
clients	I-KEY
instead	O
of	O
the	O
risky	O
dot	O
coms	O

Scalable	B-KEY
hybrid	I-KEY
computation	I-KEY
with	O
spikes	B-KEY
We	O
outline	O
a	O
hybrid	B-KEY
analog-digital	I-KEY
scheme	I-KEY
for	O
computing	O
with	O
three	O
important	O
features	O
that	O
enable	O
it	O
to	O
scale	O
to	O
systems	O
of	O
large	O
complexity	O
:	O
First	O
,	O
like	O
digital	O
computation	O
,	O
which	O
uses	O
several	O
one-bit	O
precise	O
logical	O
units	O
to	O
collectively	O
compute	O
a	O
precise	O
answer	O
to	O
a	O
computation	O
,	O
the	O
hybrid	O
scheme	O
uses	O
several	O
moderate-precision	B-KEY
analog	I-KEY
units	I-KEY
to	O
collectively	O
compute	O
a	O
precise	O
answer	O
to	O
a	O
computation	O
.	O
Second	O
,	O
frequent	B-KEY
discrete	I-KEY
signal	I-KEY
restoration	I-KEY
of	O
the	O
analog	O
information	O
prevents	O
analog	B-KEY
noise	I-KEY
and	O
offset	O
from	O
degrading	O
the	O
computation	O
.	O
Third	O
,	O
a	O
state	O
machine	O
enables	O
complex	O
computations	O
to	O
be	O
created	O
using	O
a	O
sequence	O
of	O
elementary	O
computations	O
.	O
A	O
natural	O
choice	O
for	O
implementing	O
this	O
hybrid	O
scheme	O
is	O
one	O
based	O
on	O
spikes	B-KEY
because	O
spike-count	O
codes	O
are	O
digital	O
,	O
while	O
spike-time	O
codes	O
are	O
analog	O
.	O
We	O
illustrate	O
how	O
spikes	B-KEY
afford	O
easy	O
ways	O
to	O
implement	O
all	O
three	O
components	O
of	O
scalable	B-KEY
hybrid	I-KEY
computation	I-KEY
.	O
First	O
,	O
as	O
an	O
important	O
example	O
of	O
distributed	B-KEY
analog	I-KEY
computation	I-KEY
,	O
we	O
show	O
how	O
spikes	B-KEY
can	O
create	O
a	O
distributed	O
modular	O
representation	O
of	O
an	O
analog	O
number	O
by	O
implementing	O
digital	B-KEY
carry	I-KEY
interactions	I-KEY
between	O
spiking	B-KEY
analog	O
neurons	O
.	O
Second	O
,	O
we	O
show	O
how	O
signal	O
restoration	O
may	O
be	O
performed	O
by	O
recursive	O
spike-count	O
quantization	O
of	O
spike-time	O
codes	O
.	O
Third	O
,	O
we	O
use	O
spikes	B-KEY
from	O
an	O
analog	O
dynamical	O
system	O
to	O
trigger	O
state	O
transitions	O
in	O
a	O
digital	O
dynamical	O
system	O
,	O
which	O
reconfigures	O
the	O
analog	O
dynamical	O
system	O
using	O
a	O
binary	B-KEY
control	I-KEY
vector	I-KEY
;	O
such	O
feedback	B-KEY
interactions	I-KEY
between	O
analog	O
and	O
digital	O
dynamical	O
systems	O
create	O
a	O
hybrid	O
state	O
machine	O
-LRB-	O
HSM	O
-RRB-	O
.	O
The	O
HSM	O
extends	O
and	O
expands	O
the	O
concept	O
of	O
a	O
digital	O
finite-state-machine	B-KEY
to	O
the	O
hybrid	O
domain	O
.	O
We	O
present	O
experimental	O
data	O
from	O
a	O
two-neuron	O
HSM	O
on	O
a	O
chip	O
that	O
implements	O
error-correcting	B-KEY
analog-to-digital	I-KEY
conversion	I-KEY
with	O
the	O
concurrent	O
use	O
of	O
spike-time	O
and	O
spike-count	O
codes	O
.	O
We	O
also	O
present	O
experimental	O
data	O
from	O
silicon	B-KEY
circuits	I-KEY
that	O
implement	O
HSM-based	O
pattern	B-KEY
recognition	I-KEY
using	O
spike-time	O
synchrony	O
.	O
We	O
outline	O
how	O
HSMs	O
may	O
be	O
used	O
to	O
perform	O
learning	B-KEY
,	O
vector	B-KEY
quantization	I-KEY
,	O
spike	B-KEY
pattern	B-KEY
recognition	I-KEY
and	O
generation	O
,	O
and	O
how	O
they	O
may	O
be	O
reconfigured	O

A	O
quantum	B-KEY
full	I-KEY
adder	I-KEY
for	O
a	O
scalable	B-KEY
nuclear	I-KEY
spin	I-KEY
quantum	I-KEY
computer	I-KEY
We	O
demonstrate	O
a	O
strategy	O
for	O
implementation	O
a	O
quantum	B-KEY
full	I-KEY
adder	I-KEY
in	O
a	O
spin	O
chain	O
quantum	O
computer	O
.	O
As	O
an	O
example	O
,	O
we	O
simulate	O
a	O
quantum	B-KEY
full	I-KEY
adder	I-KEY
in	O
a	O
chain	O
containing	O
201	O
spins	O
.	O
Our	O
simulations	O
also	O
demonstrate	O
how	O
one	O
can	O
minimize	O
errors	O
generated	O
by	O
non-resonant	O
effects	O

HPF/JA	O
:	O
extensions	O
of	O
High	O
Performance	O
Fortran	O
for	O
accelerating	O
real-world	O
applications	O
This	O
paper	O
presents	O
a	O
set	O
of	O
extensions	O
on	O
High	B-KEY
Performance	I-KEY
Fortran	I-KEY
-LRB-	O
HPF	B-KEY
-RRB-	O
to	O
make	O
it	O
more	O
usable	O
for	O
parallelizing	O
real-world	O
production	O
codes	O
.	O
HPF	B-KEY
has	O
been	O
effective	O
for	O
programs	O
that	O
a	O
compiler	B-KEY
can	O
automatically	O
optimize	O
efficiently	O
.	O
However	O
,	O
once	O
the	O
compiler	B-KEY
can	O
not	O
,	O
there	O
have	O
been	O
no	O
ways	O
for	O
the	O
users	O
to	O
explicitly	O
parallelize	O
or	O
optimize	O
their	O
programs	O
.	O
In	O
order	O
to	O
resolve	O
the	O
situation	O
,	O
we	O
have	O
developed	O
a	O
set	O
of	O
HPF	B-KEY
extensions	O
-LRB-	O
HPF/JA	O
-RRB-	O
to	O
give	O
the	O
users	O
more	O
control	O
over	O
sophisticated	O
parallelization	O
and	O
communication	O
optimizations	O
.	O
They	O
include	O
parallelization	B-KEY
of	I-KEY
loops	I-KEY
with	O
complicated	O
reductions	O
,	O
asynchronous	O
communication	O
,	O
user-controllable	O
shadow	O
,	O
and	O
communication	O
pattern	O
reuse	O
for	O
irregular	O
remote	O
data	O
accesses	O
.	O
Preliminary	O
experiments	O
have	O
proved	O
that	O
the	O
extensions	O
are	O
effective	O
at	O
increasing	O
HPF	B-KEY
's	O
usability	O

Approximate	B-KEY
confidence	I-KEY
intervals	I-KEY
for	O
one	O
proportion	O
and	O
difference	B-KEY
of	I-KEY
two	I-KEY
proportions	I-KEY
Constructing	O
a	O
confidence	O
interval	O
for	O
a	O
binomial	B-KEY
proportion	I-KEY
or	O
the	O
difference	B-KEY
of	I-KEY
two	I-KEY
proportions	I-KEY
is	O
a	O
routine	O
exercise	O
in	O
daily	O
data	B-KEY
analysis	I-KEY
.	O
The	O
best-known	O
method	O
is	O
the	O
Wald	O
interval	O
based	O
on	O
the	O
asymptotic	O
normal	O
approximation	O
to	O
the	O
distribution	O
of	O
the	O
observed	O
sample	O
proportion	O
,	O
though	O
it	O
is	O
known	O
to	O
have	O
bad	O
performance	O
for	O
small	O
to	O
medium	O
sample	O
sizes	O
.	O
Agresti	O
et	O
al.	O
-LRB-	O
1998	O
,	O
2000	O
-RRB-	O
proposed	O
an	O
Adding-4	O
method	O
:	O
4	O
pseudo-observations	O
are	O
added	O
with	O
2	O
successes	O
and	O
2	O
failures	O
and	O
then	O
the	O
resulting	O
-LRB-	O
pseudo	O
-	O
-RRB-	O
sample	O
proportion	O
is	O
used	O
.	O
The	O
method	O
is	O
simple	O
and	O
performs	O
extremely	O
well	O
.	O
Here	O
we	O
propose	O
an	O
approximate	O
method	O
based	O
on	O
a	O
t-approximation	B-KEY
that	O
takes	O
account	O
of	O
the	O
uncertainty	B-KEY
in	O
estimating	O
the	O
variance	O
of	O
the	O
observed	O
-LRB-	O
pseudo	O
-	O
-RRB-	O
sample	O
proportion	O
.	O
It	O
follows	O
the	O
same	O
line	O
of	O
using	O
a	O
t-test	B-KEY
,	O
rather	O
than	O
z-test	O
,	O
in	O
testing	O
the	O
mean	O
of	O
a	O
normal	B-KEY
distribution	I-KEY
with	O
an	O
unknown	O
variance	O
.	O
For	O
some	O
circumstances	O
our	O
proposed	O
method	O
has	O
a	O
higher	O
coverage	B-KEY
probability	I-KEY
than	O
the	O
Adding-4	O
method	O

Trading	B-KEY
exchanges	I-KEY
:	O
online	B-KEY
marketplaces	I-KEY
evolve	O
Looks	O
at	O
how	O
trading	B-KEY
exchanges	I-KEY
are	O
evolving	O
rapidly	O
to	O
help	O
manufacturers	B-KEY
keep	O
up	O
with	O
customer	B-KEY
demand	I-KEY

Agents	O
in	O
e-commerce	O
:	O
state	B-KEY
of	I-KEY
the	I-KEY
art	I-KEY
This	O
paper	O
surveys	O
the	O
state	B-KEY
of	I-KEY
the	I-KEY
art	I-KEY
of	O
agent-mediated	B-KEY
electronic	I-KEY
commerce	I-KEY
-LRB-	O
e-commerce	O
-RRB-	O
,	O
especially	O
in	O
business-to-consumer	O
-LRB-	O
B2C	O
-RRB-	O
e-commerce	O
and	O
business-to-business	O
-LRB-	O
B2B	O
-RRB-	O
e-commerce	O
.	O
From	O
the	O
consumer	B-KEY
buying	I-KEY
behaviour	I-KEY
perspective	O
,	O
the	O
roles	O
of	O
agents	O
in	O
B2C	O
e-commerce	O
are	O
:	O
product	B-KEY
brokering	I-KEY
,	O
merchant	B-KEY
brokering	I-KEY
,	O
and	O
negotiation	B-KEY
.	O
The	O
applications	O
of	O
agents	O
in	O
B2B	O
e-commerce	O
are	O
mainly	O
in	O
supply	B-KEY
chain	I-KEY
management	I-KEY
.	O
Mobile	B-KEY
agents	I-KEY
,	O
evolutionary	B-KEY
agents	I-KEY
,	O
and	O
data-mining	B-KEY
agents	I-KEY
are	O
some	O
special	O
techniques	O
which	O
can	O
be	O
applied	O
in	O
agent-mediated	O
e-commerce	O
.	O
In	O
addition	O
,	O
some	O
technologies	O
for	O
implementation	O
are	O
briefly	O
reviewed	O
.	O
Finally	O
,	O
we	O
conclude	O
this	O
paper	O
by	O
discussions	O
on	O
the	O
future	O
directions	O
of	O
agent-mediated	O
e-commerce	O

Disability-related	B-KEY
special	I-KEY
libraries	I-KEY
One	O
of	O
the	O
ways	O
that	O
the	O
federal	B-KEY
government	I-KEY
works	O
to	O
improve	O
services	O
to	O
people	O
with	O
disabilities	O
is	O
to	O
fund	O
disability-related	B-KEY
information	I-KEY
centers	I-KEY
and	O
clearinghouses	O
that	O
provide	O
information	B-KEY
resources	I-KEY
and	O
referrals	O
to	O
disabled	O
individuals	O
,	O
their	O
family	O
members	O
,	O
service	O
providers	O
,	O
and	O
the	O
general	O
public	O
.	O
The	O
Teaching	O
Research	O
Division	O
of	O
Western	B-KEY
Oregon	I-KEY
University	I-KEY
operates	O
two	O
federally	O
funded	O
information	O
centers	O
for	O
people	O
with	O
disabilities	O
:	O
OBIRN	B-KEY
-LRB-	O
the	O
Oregon	B-KEY
Brain	I-KEY
Injury	I-KEY
Resource	I-KEY
Network	I-KEY
-RRB-	O
and	O
DB-LINK	B-KEY
-LRB-	O
the	O
National	B-KEY
Information	I-KEY
Clearinghouse	I-KEY
on	I-KEY
Children	I-KEY
who	I-KEY
are	I-KEY
Deaf-Blind	I-KEY
-RRB-	O
.	O
Both	O
have	O
developed	O
in-depth	O
library	B-KEY
collections	I-KEY
and	O
services	O
in	O
addition	O
to	O
typical	O
clearinghouse	O
services	O
.	O
The	O
authors	O
describe	O
how	O
OBIRN	B-KEY
and	O
DB-LINK	B-KEY
were	O
designed	O
and	O
developed	O
,	O
and	O
how	O
they	O
are	O
currently	O
structured	O
and	O
maintained	O
.	O
Both	O
information	O
centers	O
use	O
many	O
of	O
the	O
same	O
strategies	O
and	O
tools	O
in	O
day-to-day	O
operations	O
,	O
but	O
differ	O
in	O
a	O
number	O
of	O
ways	O
,	O
including	O
materials	O
and	O
clientele	O

Cat	O
and	O
class	O
:	O
what	O
use	O
are	O
these	O
skills	O
to	O
the	O
new	O
legal	B-KEY
information	I-KEY
professional	I-KEY
?	O
This	O
article	O
looks	O
at	O
the	O
cataloguing	B-KEY
and	O
classification	B-KEY
skills	O
taught	O
on	O
information	B-KEY
studies	I-KEY
courses	I-KEY
and	O
the	O
use	O
these	O
skills	O
are	O
to	O
new	O
legal	B-KEY
information	I-KEY
professionals	I-KEY
.	O
The	O
article	O
is	O
based	O
on	O
the	O
opinions	O
of	O
nine	O
new	O
legal	B-KEY
information	I-KEY
professionals	I-KEY
from	O
both	O
academic	O
and	O
law	B-KEY
firm	I-KEY
libraries	I-KEY

Score	B-KEY
tests	I-KEY
for	O
zero-inflated	O
Poisson	O
models	O
In	O
many	O
situations	O
count	B-KEY
data	I-KEY
have	O
a	O
large	O
proportion	O
of	O
zeros	O
and	O
the	O
zero-inflated	O
Poisson	O
regression	O
-LRB-	O
ZIP	O
-RRB-	O
model	O
may	O
be	O
appropriate	O
.	O
A	O
simple	O
score	B-KEY
test	I-KEY
for	O
zero-inflation	O
,	O
comparing	O
the	O
ZIP	O
model	O
with	O
a	O
constant	O
proportion	O
of	O
excess	B-KEY
zeros	I-KEY
to	O
a	O
standard	O
Poisson	O
regression	O
model	O
,	O
was	O
given	O
by	O
van	O
den	O
Broek	O
-LRB-	O
1995	O
-RRB-	O
.	O
We	O
extend	O
this	O
test	O
to	O
the	O
more	O
general	O
situation	O
where	O
the	O
zero	B-KEY
probability	I-KEY
is	O
allowed	O
to	O
depend	O
on	O
covariates	B-KEY
.	O
The	O
performance	O
of	O
this	O
test	O
is	O
evaluated	O
using	O
a	O
simulation	B-KEY
study	O
.	O
To	O
identify	O
potentially	O
important	O
covariates	B-KEY
in	O
the	O
zero-inflation	O
model	O
a	O
composite	B-KEY
test	I-KEY
is	O
proposed	O
.	O
The	O
use	O
of	O
the	O
general	O
score	B-KEY
test	I-KEY
and	O
the	O
composite	O
procedure	O
is	O
illustrated	O
on	O
two	O
examples	O
from	O
the	O
literature	O
.	O
The	O
composite	O
score	B-KEY
test	I-KEY
is	O
found	O
to	O
suggest	O
appropriate	O
models	O

Succession	O
in	O
standardization	B-KEY
:	O
grafting	O
XML	B-KEY
onto	O
SGML	B-KEY
Succession	O
in	O
standardization	B-KEY
is	O
often	O
a	O
problem	O
.	O
The	O
advantages	O
of	O
improvements	O
must	O
be	O
weighed	O
against	O
those	O
of	O
compatibility	O
.	O
If	O
compatibility	O
considerations	O
dominate	O
,	O
a	O
grafting	B-KEY
process	I-KEY
takes	O
place	O
.	O
According	O
to	O
our	O
taxonomy	O
of	O
succession	O
,	O
there	O
are	O
three	O
types	O
of	O
outcomes	O
.	O
A	O
Type	B-KEY
I	I-KEY
succession	I-KEY
,	O
where	O
grafting	O
is	O
successful	O
,	O
entails	O
compatibility	O
between	O
successors	O
,	O
technical	O
paradigm	O
compliance	O
and	O
continuity	O
in	O
the	O
standards	B-KEY
trajectory	O
.	O
In	O
this	O
paper	O
,	O
we	O
examine	O
issues	O
of	O
succession	O
and	O
focus	O
on	O
the	O
Extensible	B-KEY
Markup	I-KEY
Language	I-KEY
-LRB-	O
XML	B-KEY
-RRB-	O
.	O
It	O
was	O
to	O
be	O
grafted	O
on	O
the	O
Standard	B-KEY
Generalized	O
Markup	O
Language	O
-LRB-	O
SGML	O
-RRB-	O
,	O
a	O
stable	O
standard	O
since	O
1988	O
.	O
However	O
,	O
XML	B-KEY
was	O
a	O
profile	O
,	O
a	O
subset	O
and	O
an	O
extension	O
of	O
SGML	B-KEY
-LRB-	O
1988	O
-RRB-	O
.	O
Adaptation	O
of	O
SGML	B-KEY
was	O
needed	O
-LRB-	O
SGML	B-KEY
1999	O
-RRB-	O
to	O
forge	O
full	O
-LRB-	O
downward	O
-RRB-	O
compatibility	O
with	O
XML	B-KEY
-LRB-	O
1998	O
-RRB-	O
.	O
We	O
describe	O
the	O
grafting	O
efforts	O
and	O
analyze	O
their	O
outcomes	O
.	O
Our	O
conclusion	O
is	O
that	O
although	O
SGML	B-KEY
was	O
a	O
technical	O
exemplar	O
for	O
XML	B-KEY
developers	O
,	O
full	O
compatibility	O
was	O
not	O
achieved	O
.	O
The	O
widespread	O
use	O
of	O
HyperText	O
Mark-up	O
Language	O
-LRB-	O
HTML	O
-RRB-	O
exemplified	O
the	O
desirability	O
of	O
simplicity	O
in	O
XML	B-KEY
,	O
standardization	B-KEY
.	O
This	O
and	O
HTML	O
's	O
user	O
market	O
largely	O
explain	O
the	O
discontinuity	O
in	O
SGML-XML	O
succession	O

Does	O
social	O
capital	O
determine	O
innovation	B-KEY
?	O
To	O
what	O
extent	O
?	O
This	O
paper	O
deals	O
with	O
two	O
questions	O
:	O
Does	O
social	O
capital	O
determine	O
innovation	B-KEY
in	O
manufacturing	B-KEY
firms	I-KEY
?	O
If	O
it	O
is	O
the	O
case	O
,	O
to	O
what	O
extent	O
?	O
To	O
deal	O
with	O
these	O
questions	O
,	O
we	O
review	O
the	O
literature	O
on	O
innovation	B-KEY
in	O
order	O
to	O
see	O
how	O
social	O
capital	O
came	O
to	O
be	O
added	O
to	O
the	O
other	O
forms	O
of	O
capital	O
as	O
an	O
explanatory	O
variable	O
of	O
innovation	B-KEY
.	O
In	O
doing	O
so	O
,	O
we	O
have	O
been	O
led	O
to	O
follow	O
the	O
dominating	O
view	O
of	O
the	O
literature	O
on	O
social	O
capital	O
and	O
innovation	B-KEY
which	O
claims	O
that	O
social	O
capital	O
can	O
not	O
be	O
captured	O
through	O
a	O
single	O
indicator	O
,	O
but	O
that	O
it	O
actually	O
takes	O
many	O
different	O
forms	O
that	O
must	O
be	O
accounted	O
for	O
.	O
Therefore	O
,	O
to	O
the	O
traditional	O
explanatory	O
variables	O
of	O
innovation	B-KEY
,	O
we	O
have	O
added	O
five	O
forms	O
of	O
structural	B-KEY
social	I-KEY
capital	I-KEY
-LRB-	O
business	B-KEY
network	I-KEY
assets	I-KEY
,	O
information	B-KEY
network	I-KEY
assets	I-KEY
,	O
research	B-KEY
network	I-KEY
assets	I-KEY
,	O
participation	B-KEY
assets	I-KEY
,	O
and	O
relational	B-KEY
assets	I-KEY
-RRB-	O
and	O
one	O
form	O
of	O
cognitive	B-KEY
social	I-KEY
capital	I-KEY
-LRB-	O
reciprocal	B-KEY
trust	I-KEY
-RRB-	O
.	O
In	O
a	O
context	O
where	O
empirical	O
investigations	O
regarding	O
the	O
relations	O
between	O
social	O
capital	O
and	O
innovation	B-KEY
are	O
still	O
scanty	O
,	O
this	O
paper	O
makes	O
contributions	O
to	O
the	O
advancement	O
of	O
knowledge	O
in	O
providing	O
new	O
evidence	O
regarding	O
the	O
impact	O
and	O
the	O
extent	O
of	O
social	O
capital	O
on	O
innovation	B-KEY
at	O
the	O
two	O
decisionmaking	O
stages	O
considered	O
in	O
this	O
study	O

Design	O
of	O
an	O
adaptive	B-KEY
vibration	I-KEY
absorber	I-KEY
to	O
reduce	O
electrical	B-KEY
transformer	I-KEY
structural	B-KEY
vibration	I-KEY
This	O
paper	O
considers	O
the	O
design	O
of	O
a	O
vibration	O
absorber	O
to	O
reduce	O
structural	B-KEY
vibration	I-KEY
at	O
multiple	O
frequencies	O
,	O
with	O
an	O
enlarged	O
bandwidth	B-KEY
control	I-KEY
at	O
these	O
target	O
frequencies	O
.	O
While	O
the	O
basic	O
absorber	O
is	O
a	O
passive	O
device	O
a	O
control	O
system	O
has	O
been	O
added	O
to	O
facilitate	O
tuning	O
,	O
effectively	O
giving	O
the	O
combination	O
of	O
a	O
passive	O
and	O
active	O
device	O
,	O
which	O
leads	O
to	O
far	O
greater	O
stability	O
and	O
robustness	O
.	O
Experimental	O
results	O
demonstrating	O
the	O
effectiveness	O
of	O
the	O
absorber	O
are	O
also	O
described	O

Operator	B-KEY
splitting	I-KEY
and	O
approximate	B-KEY
factorization	I-KEY
for	O
taxis-diffusion-reaction	B-KEY
models	I-KEY
In	O
this	O
paper	O
we	O
consider	O
the	O
numerical	B-KEY
solution	I-KEY
of	O
2D	O
systems	O
of	O
certain	O
types	O
of	O
taxis-diffusion-reaction	O
equations	O
from	O
mathematical	B-KEY
biology	I-KEY
.	O
By	O
spatial	B-KEY
discretization	I-KEY
these	O
PDE	B-KEY
systems	I-KEY
are	O
approximated	O
by	O
systems	O
of	O
positive	O
,	O
nonlinear	B-KEY
ODEs	I-KEY
-LRB-	O
Method	O
of	O
Lines	O
-RRB-	O
.	O
The	O
aim	O
of	O
this	O
paper	O
is	O
to	O
examine	O
the	O
numerical	B-KEY
integration	I-KEY
of	O
these	O
ODE	O
systems	O
for	O
low	O
to	O
moderate	O
accuracy	O
by	O
means	O
of	O
splitting	O
techniques	O
.	O
An	O
important	O
consideration	O
is	O
maintenance	O
of	O
positivity	O
.	O
We	O
apply	O
operator	B-KEY
splitting	I-KEY
and	O
approximate	B-KEY
matrix	I-KEY
factorization	I-KEY
using	O
low	O
order	O
explicit	O
Runge-Kutta	B-KEY
methods	I-KEY
and	O
linearly	B-KEY
implicit	I-KEY
Runge-Kutta-Rosenbrock	I-KEY
methods	I-KEY
.	O
As	O
a	O
reference	O
method	O
the	O
general	O
purpose	O
solver	O
VODPK	O
is	O
applied	O

A	O
maximum-likelihood	B-KEY
surface	I-KEY
estimator	I-KEY
for	O
dense	B-KEY
range	I-KEY
data	I-KEY
Describes	O
how	O
to	O
estimate	O
3D	B-KEY
surface	I-KEY
models	I-KEY
from	O
dense	O
sets	O
of	O
noisy	B-KEY
range	I-KEY
data	I-KEY
taken	O
from	O
different	O
points	O
of	O
view	O
,	O
i.e.	O
,	O
multiple	O
range	O
maps	O
.	O
The	O
proposed	O
method	O
uses	O
a	O
sensor	B-KEY
model	I-KEY
to	O
develop	O
an	O
expression	O
for	O
the	O
likelihood	O
of	O
a	O
3D	O
surface	O
,	O
conditional	O
on	O
a	O
set	O
of	O
noisy	B-KEY
range	I-KEY
measurements	I-KEY
.	O
Optimizing	O
this	O
likelihood	O
with	O
respect	O
to	O
the	O
model	O
parameters	O
provides	O
an	O
unbiased	O
and	O
efficient	O
estimator	O
.	O
The	O
proposed	O
numerical	O
algorithms	O
make	O
this	O
estimation	O
computationally	O
practical	O
for	O
a	O
wide	O
variety	O
of	O
circumstances	O
.	O
The	O
results	O
from	O
this	O
method	O
compare	O
favorably	O
with	O
state-of-the-art	O
approaches	O
that	O
rely	O
on	O
the	O
closest-point	O
or	O
perpendicular	O
distance	O
metric	O
,	O
a	O
convenient	O
heuristic	B-KEY
that	O
produces	O
biased	B-KEY
solutions	I-KEY
and	O
fails	O
completely	O
when	O
surfaces	O
are	O
not	O
sufficiently	O
smooth	O
,	O
as	O
in	O
the	O
case	O
of	O
complex	B-KEY
scenes	I-KEY
or	O
noisy	B-KEY
range	I-KEY
measurements	I-KEY
.	O
Empirical	O
results	O
on	O
both	O
simulated	O
and	O
real	B-KEY
ladar	I-KEY
data	I-KEY
demonstrate	O
the	O
effectiveness	O
of	O
the	O
proposed	O
method	O
for	O
several	O
different	O
types	O
of	O
problems	O
.	O
Furthermore	O
,	O
the	O
proposed	O
method	O
offers	O
a	O
general	O
framework	O
that	O
can	O
accommodate	O
extensions	O
to	O
include	O
surface	O
priors	O
,	O
more	O
sophisticated	O
noise	O
models	O
,	O
and	O
other	O
sensing	O
modalities	O
,	O
such	O
as	O
sonar	B-KEY
or	O
synthetic	B-KEY
aperture	I-KEY
radar	I-KEY

Active	B-KEY
vibration	I-KEY
control	I-KEY
of	O
piezolaminated	B-KEY
smart	I-KEY
beams	I-KEY
This	O
paper	O
deals	O
with	O
the	O
active	B-KEY
vibration	I-KEY
control	I-KEY
of	O
beam	B-KEY
like	I-KEY
structures	I-KEY
with	O
distributed	O
piezoelectric	O
sensor	O
and	O
actuator	O
layers	O
bonded	O
on	O
top	O
and	O
bottom	B-KEY
surfaces	I-KEY
of	O
the	O
beam	O
.	O
A	O
finite	B-KEY
element	I-KEY
model	I-KEY
based	O
on	O
Euler-Bernoulli	B-KEY
beam	I-KEY
theory	I-KEY
has	O
been	O
developed	O
.	O
The	O
contribution	O
of	O
the	O
piezoelectric	O
sensor	O
and	O
actuator	O
layers	O
on	O
the	O
mass	B-KEY
and	O
stiffness	B-KEY
of	O
the	O
beam	O
is	O
considered	O
.	O
Three	O
types	O
of	O
classical	O
control	O
strategies	O
,	O
namely	O
direct	B-KEY
proportional	I-KEY
feedback	I-KEY
,	O
constant-gain	B-KEY
negative	I-KEY
velocity	I-KEY
feedback	I-KEY
and	O
Lyapunov	B-KEY
feedback	I-KEY
and	O
an	O
optimal	B-KEY
control	I-KEY
strategy	I-KEY
,	O
linear	B-KEY
quadratic	I-KEY
regulator	I-KEY
-LRB-	O
LQR	O
-RRB-	O
scheme	O
are	O
applied	O
to	O
study	O
their	O
control	B-KEY
effectiveness	I-KEY
.	O
Also	O
,	O
the	O
control	O
performance	O
with	O
different	O
types	O
of	O
loading	O
,	O
such	O
as	O
impulse	B-KEY
loading	I-KEY
,	O
step	B-KEY
loading	I-KEY
,	O
harmonic	O
and	O
random	B-KEY
loading	I-KEY
is	O
studied	O

Updating	O
systems	O
for	O
monitoring	O
and	O
controlling	O
power	O
equipment	O
on	O
the	O
basis	O
of	O
the	O
firmware	O
system	O
SARGON	O
The	O
economic	O
difficulties	O
experienced	O
by	O
the	O
power	B-KEY
industry	I-KEY
of	O
Russia	B-KEY
has	O
considerably	O
retarded	O
the	O
speed	O
of	O
commissioning	O
new	O
capacities	O
and	O
reconstructing	O
equipment	O
in	O
service	O
.	O
The	O
increasing	O
deterioration	O
of	O
the	O
equipment	O
at	O
power	O
stations	O
makes	O
the	O
problem	O
of	O
its	O
updating	O
very	O
acute	O
.	O
The	O
main	O
efforts	O
of	O
organizations	O
working	O
in	O
the	O
power	B-KEY
industry	I-KEY
are	O
now	O
focused	O
on	O
updating	O
all	O
kinds	O
of	O
equipment	O
installed	O
at	O
power	O
installations	O
.	O
The	O
necessary	O
condition	O
for	O
the	O
efficient	O
operation	O
of	O
power	O
equipment	O
is	O
to	O
carry	O
out	O
serious	O
modernization	O
of	O
systems	O
for	O
monitoring	O
and	O
control	O
-LRB-	O
SMC	O
-RRB-	O
of	O
technological	O
processes	O
.	O
The	O
specialists	O
at	O
ZAO	B-KEY
NVT-Avtomatika	I-KEY
have	O
developed	O
efficient	O
technology	O
for	O
updating	O
the	O
SMC	O
on	O
the	O
basis	O
of	O
the	O
firmware	O
system	O
SARGON	O
which	O
ensures	O
the	O
fast	O
introduction	O
of	O
high-quality	O
systems	O
of	O
automation	O
with	O
a	O
minimal	O
payback	O
time	O
of	O
the	O
capital	O
outlay	O
.	O
This	O
paper	O
discusses	O
the	O
updating	O
of	O
equipment	O
using	O
SARGON	O

More	O
than	O
the	O
money	O
-LSB-	O
software	B-KEY
project	I-KEY
-RSB-	O
Experiences	O
creating	O
budgets	B-KEY
for	O
large	O
software	B-KEY
projects	I-KEY
have	O
taught	O
manufacturers	O
that	O
it	O
is	O
not	O
about	O
the	O
money	O
-	O
it	O
is	O
about	O
what	O
one	O
really	O
needs	O
.	O
Before	O
a	O
company	O
can	O
begin	O
to	O
build	O
a	O
budget	B-KEY
for	O
a	O
software	O
.	O
project	O
,	O
it	O
has	O
to	O
have	O
a	O
good	O
understanding	O
of	O
what	O
business	O
issues	O
need	O
to	O
be	O
addressed	O
and	O
what	O
the	O
business	O
objectives	O
are	O
.	O
This	O
step	O
is	O
critical	O
because	O
it	O
defines	O
the	O
business	O
goals	O
,	O
outlines	O
the	O
metrics	O
for	O
success	O
,	O
sets	O
the	O
scope	O
for	O
the	O
project	O
,	O
and	O
defines	O
the	O
criteria	O
for	O
selecting	O
the	O
right	O
software	O

Stability	O
of	O
W-methods	O
with	O
applications	O
to	O
operator	B-KEY
splitting	I-KEY
and	O
to	O
geometric	B-KEY
theory	I-KEY
We	O
analyze	O
the	O
stability	O
properties	O
of	O
W-methods	O
applied	O
to	O
the	O
parabolic	B-KEY
initial	I-KEY
value	I-KEY
problem	I-KEY
u	O
'	O
+	O
Au	O
=	O
Bu	O
.	O
We	O
work	O
in	O
an	O
abstract	B-KEY
Banach	I-KEY
space	I-KEY
setting	O
,	O
assuming	O
that	O
A	O
is	O
the	O
generator	O
of	O
an	O
analytic	B-KEY
semigroup	I-KEY
and	O
that	O
B	O
is	O
relatively	O
bounded	O
with	O
respect	O
to	O
A	O
.	O
Since	O
W-methods	O
treat	O
the	O
term	O
with	O
A	O
implicitly	O
,	O
whereas	O
the	O
term	O
involving	O
B	O
is	O
discretized	O
in	O
an	O
explicit	O
way	O
,	O
they	O
can	O
be	O
regarded	O
as	O
splitting	O
methods	O
.	O
As	O
an	O
application	O
of	O
our	O
stability	O
results	O
,	O
convergence	O
for	O
nonsmooth	B-KEY
initial	I-KEY
data	I-KEY
is	O
shown	O
.	O
Moreover	O
,	O
the	O
layout	O
of	O
a	O
geometric	B-KEY
theory	I-KEY
for	O
discretizations	O
of	O
semilinear	O
parabolic	O
problems	O
u	O
'	O
+	O
Au	O
=	O
f	O
-LRB-	O
u	O
-RRB-	O
is	O
presented	O

Efficient	O
combinational	B-KEY
verification	I-KEY
using	O
overlapping	B-KEY
local	I-KEY
BDDs	I-KEY
and	O
a	O
hash	B-KEY
table	I-KEY
We	O
propose	O
a	O
novel	O
methodology	O
that	O
combines	O
local	O
BDDs	O
-LRB-	O
binary	B-KEY
decision	I-KEY
diagrams	I-KEY
-RRB-	O
with	O
a	O
hash	B-KEY
table	I-KEY
for	O
very	O
efficient	O
verification	O
of	O
combinational	O
circuits	O
.	O
The	O
main	O
purpose	O
of	O
this	O
technique	O
is	O
to	O
remove	O
the	O
considerable	O
overhead	O
associated	O
with	O
case-by-case	B-KEY
verification	I-KEY
of	O
internal	B-KEY
node	I-KEY
pairs	I-KEY
in	O
typical	O
internal	B-KEY
correspondence	I-KEY
based	I-KEY
verification	I-KEY
methods	O
.	O
Two	O
heuristics	B-KEY
based	O
on	O
the	O
number	O
of	O
structural	B-KEY
levels	I-KEY
of	O
circuitry	O
looked	O
at	O
and	O
the	O
total	O
number	O
of	O
nodes	O
in	O
the	O
BDD	B-KEY
manager	I-KEY
are	O
used	O
to	O
control	O
the	O
BDD	B-KEY
sizes	I-KEY
and	O
introduce	O
new	O
cutsets	B-KEY
based	O
on	O
already	O
found	O
equivalent	O
nodes	O
.	O
We	O
verify	O
the	O
ISCAS85	B-KEY
benchmark	I-KEY
circuits	I-KEY
and	O
demonstrate	O
significant	O
speedup	O
over	O
existing	O
methods	O
.	O
We	O
also	O
verify	O
several	O
hard	B-KEY
industrial	I-KEY
circuits	I-KEY
and	O
show	O
our	O
superiority	O
in	O
extracting	O
internal	B-KEY
equivalences	I-KEY

Development	O
of	O
visual	B-KEY
design	I-KEY
steering	I-KEY
as	O
an	O
aid	O
in	O
large-scale	B-KEY
multidisciplinary	I-KEY
design	I-KEY
optimization	I-KEY
.	O
II	O
.	O
Method	B-KEY
validation	I-KEY
For	O
pt	O
.	O
I	O
see	O
ibid.	O
,	O
pp.	O
412-24	O
.	O
Graph	B-KEY
morphing	I-KEY
,	O
the	O
first	O
concept	O
developed	O
under	O
the	O
newly	O
proposed	O
paradigm	O
of	O
visual	B-KEY
design	I-KEY
steering	I-KEY
-LRB-	O
VDS	O
-RRB-	O
,	O
is	O
applied	O
to	O
optimal	B-KEY
design	I-KEY
problems	I-KEY
.	O
Graph	B-KEY
morphing	I-KEY
,	O
described	O
in	O
Part	O
I	O
of	O
this	O
paper	O
,	O
can	O
be	O
used	O
to	O
provide	O
insights	O
to	O
a	O
designer	O
to	O
improve	O
efficiency	O
,	O
reliability	B-KEY
,	O
and	O
accuracy	B-KEY
of	O
an	O
optimal	O
design	O
in	O
less	O
cycle	B-KEY
time	I-KEY
.	O
It	O
is	O
demonstrated	O
in	O
this	O
part	O
of	O
the	O
paper	O
that	O
graph	B-KEY
morphing	I-KEY
can	O
be	O
used	O
to	O
provide	O
insights	O
into	O
design	B-KEY
variable	I-KEY
impact	I-KEY
,	O
constraint	B-KEY
redundancy	I-KEY
,	O
reasonable	O
values	O
for	O
constraint	B-KEY
allowable	I-KEY
limits	I-KEY
,	O
and	O
function	B-KEY
smoothness	I-KEY
,	O
that	O
otherwise	O
might	O
not	O
be	O
attainable	O

Direct	B-KEY
aperture	I-KEY
optimization	I-KEY
:	O
A	O
turnkey	B-KEY
solution	I-KEY
for	O
step-and-shoot	B-KEY
IMRT	I-KEY
IMRT	B-KEY
treatment	I-KEY
plans	I-KEY
for	O
step-and-shoot	O
delivery	O
have	O
traditionally	O
been	O
produced	O
through	O
the	O
optimization	O
of	O
intensity	B-KEY
distributions	I-KEY
-LRB-	O
or	O
maps	B-KEY
-RRB-	O
for	O
each	O
beam	B-KEY
angle	I-KEY
.	O
The	O
optimization	B-KEY
step	I-KEY
is	O
followed	O
by	O
the	O
application	O
of	O
a	O
leaf-sequencing	B-KEY
algorithm	I-KEY
that	O
translates	O
each	O
intensity	B-KEY
map	I-KEY
into	O
a	O
set	O
of	O
deliverable	O
aperture	O
shapes	O
.	O
In	O
this	O
article	O
,	O
we	O
introduce	O
an	O
automated	B-KEY
planning	I-KEY
system	I-KEY
in	O
which	O
we	O
bypass	O
the	O
traditional	O
intensity	O
optimization	O
,	O
and	O
instead	O
directly	O
optimize	O
the	O
shapes	O
and	O
the	O
weights	O
of	O
the	O
apertures	O
.	O
We	O
call	O
this	O
approach	O
``	O
direct	B-KEY
aperture	I-KEY
optimization	I-KEY
.	O
''	O
This	O
technique	O
allows	O
the	O
user	O
to	O
specify	O
the	O
maximum	O
number	O
of	O
apertures	O
per	O
beam	O
direction	O
,	O
and	O
hence	O
provides	O
significant	O
control	O
over	O
the	O
complexity	O
of	O
the	O
treatment	O
delivery	O
.	O
This	O
is	O
possible	O
because	O
the	O
machine	B-KEY
dependent	I-KEY
delivery	I-KEY
constraints	I-KEY
imposed	O
by	O
the	O
MLC	B-KEY
are	O
enforced	O
within	O
the	O
aperture	B-KEY
optimization	I-KEY
algorithm	I-KEY
rather	O
than	O
in	O
a	O
separate	O
leaf-sequencing	O
step	O
.	O
The	O
leaf	B-KEY
settings	I-KEY
and	O
the	O
aperture	B-KEY
intensities	I-KEY
are	O
optimized	O
simultaneously	O
using	O
a	O
simulated	B-KEY
annealing	I-KEY
algorithm	I-KEY
.	O
We	O
have	O
tested	O
direct	B-KEY
aperture	I-KEY
optimization	I-KEY
on	O
a	O
variety	O
of	O
patient	B-KEY
cases	I-KEY
using	O
the	O
EGS4/BEAM	B-KEY
Monte	I-KEY
Carlo	I-KEY
package	I-KEY
for	O
our	O
dose	B-KEY
calculation	I-KEY
engine	I-KEY
.	O
The	O
results	O
demonstrate	O
that	O
direct	B-KEY
aperture	I-KEY
optimization	I-KEY
can	O
produce	O
highly	B-KEY
conformal	I-KEY
step-and-shoot	I-KEY
treatment	I-KEY
plans	I-KEY
using	O
only	O
three	O
to	O
five	O
apertures	O
per	O
beam	O
direction	O
.	O
As	O
compared	O
with	O
traditional	O
optimization	O
strategies	O
,	O
our	O
studies	O
demonstrate	O
that	O
direct	B-KEY
aperture	I-KEY
optimization	I-KEY
can	O
result	O
in	O
a	O
significant	O
reduction	O
in	O
both	O
the	O
number	O
of	O
beam	B-KEY
segments	I-KEY
and	O
the	O
number	O
of	O
monitor	B-KEY
units	I-KEY
.	O
Direct	B-KEY
aperture	I-KEY
optimization	I-KEY
therefore	O
produces	O
highly	B-KEY
efficient	I-KEY
treatment	I-KEY
deliveries	I-KEY
that	O
maintain	O
the	O
full	B-KEY
dosimetric	I-KEY
benefits	I-KEY
of	O
IMRT	O

Setup	O
cost	O
and	O
lead	B-KEY
time	I-KEY
reductions	I-KEY
on	O
stochastic	B-KEY
inventory	I-KEY
models	I-KEY
with	O
a	O
service	B-KEY
level	I-KEY
constraint	I-KEY
The	O
stochastic	B-KEY
inventory	I-KEY
models	I-KEY
analyzed	O
in	O
this	O
paper	O
explore	O
the	O
problem	O
of	O
lead	O
time	O
associated	O
with	O
setup	B-KEY
cost	I-KEY
reductions	I-KEY
for	O
the	O
continuous	O
review	O
and	O
periodic	B-KEY
review	I-KEY
inventory	I-KEY
models	I-KEY
.	O
For	O
these	O
two	O
models	O
with	O
a	O
mixture	O
of	O
backorders	B-KEY
and	O
lost	B-KEY
sales	I-KEY
,	O
we	O
respectively	O
assume	O
that	O
their	O
mean	O
and	O
variance	O
of	O
the	O
lead	B-KEY
time	I-KEY
demand	I-KEY
and	O
protection	B-KEY
interval	I-KEY
-LRB-	O
i.e.	O
,	O
lead	O
time	O
plus	O
review	O
period	O
-RRB-	O
demand	O
are	O
known	O
,	O
but	O
their	O
probability	B-KEY
distributions	I-KEY
are	O
unknown	O
.	O
We	O
develop	O
a	O
minimax	B-KEY
distribution	I-KEY
free	I-KEY
procedure	I-KEY
to	O
find	O
the	O
optimal	O
solution-for	O
each	O
case	O

Three-dimensional	O
periodic	B-KEY
Voronoi	I-KEY
grain	I-KEY
models	I-KEY
and	O
micromechanical	O
FE-simulations	O
of	O
a	O
two-phase	B-KEY
steel	I-KEY
A	O
three-dimensional	B-KEY
model	I-KEY
is	O
proposed	O
for	O
modeling	O
of	O
microstructures	O
.	O
The	O
model	O
is	O
based	O
on	O
the	O
finite	O
element	O
method	O
with	O
periodic	B-KEY
boundary	I-KEY
conditions	I-KEY
.	O
The	O
Voronoi	B-KEY
algorithm	I-KEY
is	O
used	O
to	O
generate	O
the	O
geometrical	B-KEY
model	I-KEY
,	O
which	O
has	O
a	O
periodic	O
grain	O
structure	O
that	O
follows	O
the	O
original	O
boundaries	O
of	O
the	O
Voronoi	O
cells	O
.	O
As	O
an	O
application	O
,	O
the	O
model	O
is	O
used	O
to	O
model	O
a	O
two-phase	O
ferrite/pearlite	O
steel	O
.	O
It	O
is	O
shown	O
that	O
periodic	O
cells	O
with	O
only	O
five	O
grains	O
generate	O
representative	O
stress-strain	B-KEY
curves	I-KEY

The	O
Tattletale	B-KEY
technique	I-KEY
Practical	O
experience	O
has	O
taught	O
many	O
Java	B-KEY
developers	O
one	O
thing	O
:	O
critical	B-KEY
resources	I-KEY
-LRB-	O
mutexes	B-KEY
,	O
database	B-KEY
connections	I-KEY
,	O
transactions	B-KEY
,	O
file	B-KEY
handles	I-KEY
,	O
etc.	O
-RRB-	O
require	O
timely	O
and	O
systematic	O
release	O
.	O
Unfortunately	O
,	O
Java	B-KEY
's	O
garbage	B-KEY
collector	I-KEY
is	O
not	O
up	O
to	O
that	O
job	O
.	O
According	O
to	O
the	O
Java	B-KEY
Language	O
Specification	O
,	O
there	O
are	O
no	O
guarantees	O
when	O
a	O
garbage	B-KEY
collector	I-KEY
will	O
run	O
,	O
when	O
it	O
will	O
collect	O
an	O
object	O
,	O
or	O
when	O
it	O
will	O
finalize	O
an	O
object	O
-	O
if	O
ever	O
.	O
Even	O
more	O
unfortunately	O
,	O
Java	B-KEY
's	O
counterpart	O
to	O
the	O
C++	O
destructor	O
-LRB-	O
the	O
finally	O
block	O
-RRB-	O
is	O
both	O
tedious	O
and	O
error-prone	O
,	O
requiring	O
developers	O
to	O
constantly	O
remember	O
and	O
duplicate	O
resource-releasing	B-KEY
code	I-KEY
.	O
Consequently	O
,	O
even	O
good	O
Java	B-KEY
developers	O
can	O
forget	O
to	O
release	O
critical	B-KEY
resources	I-KEY
.	O
There	O
is	O
a	O
light	O
at	O
the	O
end	O
of	O
the	O
tunnel	O
.	O
Java	B-KEY
may	O
make	O
it	O
easier	O
to	O
leak	O
critical	B-KEY
resources	I-KEY
,	O
but	O
it	O
also	O
provides	O
the	O
necessary	O
mechanisms	O
to	O
easily	O
track	O
them	O
down	O
.	O
The	O
Tattletale	B-KEY
technique	I-KEY
is	O
a	O
simple	O
method	O
for	O
designing	O
new	O
classes	O
and	O
retrofitting	O
existing	O
classes	O
to	O
quickly	O
and	O
easily	O
detect	O
the	O
offending	O
code	O
responsible	O
for	O
leaking	O
resources	O

A	O
scalable	B-KEY
model	I-KEY
of	O
cerebellar	B-KEY
adaptive	I-KEY
timing	I-KEY
and	O
sequencing	O
:	O
the	O
recurrent	O
slide	O
and	O
latch	O
-LRB-	O
RSL	O
-RRB-	O
model	O
From	O
the	O
dawn	O
of	O
modern	O
neural	B-KEY
network	I-KEY
theory	I-KEY
,	O
the	O
mammalian	B-KEY
cerebellum	I-KEY
has	O
been	O
a	O
favored	O
object	O
of	O
mathematical	O
modeling	O
studies	O
.	O
Early	O
studies	O
focused	O
on	O
the	O
fanout	O
,	O
convergence	O
,	O
thresholding	O
,	O
and	O
learned	O
weighting	O
of	O
perceptual-motor	O
signals	O
within	O
the	O
cerebellar	O
cortex	O
.	O
This	O
led	O
to	O
the	O
still	O
viable	O
idea	O
that	O
the	O
granule	B-KEY
cell	I-KEY
stage	I-KEY
in	O
the	O
cerebellar	O
cortex	O
performs	O
a	O
sparse	B-KEY
expansive	I-KEY
recoding	I-KEY
of	O
the	O
time-varying	B-KEY
input	I-KEY
vector	I-KEY
.	O
This	O
recoding	O
reveals	O
and	O
emphasizes	O
combinations	O
in	O
a	O
distributed	B-KEY
representation	I-KEY
that	O
serves	O
as	O
a	O
basis	O
for	O
the	O
learned	O
,	O
state-dependent	O
control	O
actions	O
engendered	O
by	O
cerebellar	O
outputs	O
to	O
movement	O
related	O
centers	O
.	O
To	O
make	O
optimal	O
use	O
of	O
available	O
signals	O
,	O
the	O
cerebellum	O
must	O
be	O
able	O
to	O
sift	O
the	O
evolving	O
state	O
representation	O
for	O
the	O
most	O
reliable	O
predictors	O
of	O
the	O
need	O
for	O
control	O
actions	O
,	O
and	O
to	O
use	O
those	O
predictors	O
even	O
if	O
they	O
appear	O
only	O
transiently	O
and	O
well	O
in	O
advance	O
of	O
the	O
optimal	O
time	O
for	O
initiating	O
the	O
control	O
action	O
.	O
The	O
paper	O
proposes	O
a	O
modification	O
to	O
prior	O
,	O
population	O
,	O
models	O
for	O
cerebellar	B-KEY
adaptive	I-KEY
timing	I-KEY
and	O
sequencing	O
.	O
Since	O
it	O
replaces	O
a	O
population	O
with	O
a	O
single	O
element	O
,	O
the	O
proposed	O
RSL	O
model	O
is	O
in	O
one	O
sense	O
maximally	O
efficient	O
,	O
and	O
therefore	O
optimal	O
from	O
the	O
perspective	O
of	O
scalability	O

Exploratory	O
study	O
of	O
the	O
adoption	O
of	O
manufacturing	B-KEY
technology	I-KEY
innovations	I-KEY
in	O
the	O
USA	B-KEY
and	O
the	O
UK	B-KEY
Manufacturing	O
technologies	O
,	O
appropriately	O
implemented	O
,	O
provide	O
competitive	B-KEY
advantage	I-KEY
to	O
manufacturers	O
.	O
The	O
use	O
of	O
manufacturing	O
technologies	O
across	O
countries	O
is	O
difficult	O
to	O
compare	O
.	O
One	O
such	O
comparison	O
has	O
been	O
provided	O
in	O
the	O
literature	O
with	O
a	O
study	O
of	O
US	O
and	O
Japanese	O
practices	O
in	O
advanced	O
manufacturing	O
technology	O
use	O
using	O
a	O
common	O
questionnaire	O
.	O
The	O
present	O
study	O
compares	O
the	O
use	O
of	O
17	O
different	O
technologies	O
in	O
similar	O
industries	O
in	O
the	O
USA	B-KEY
-LRB-	O
n	O
=	O
1025	O
-RRB-	O
and	O
UK	B-KEY
-LRB-	O
n	O
=	O
166	O
-RRB-	O
using	O
a	O
common	O
questionnaire	O
.	O
Largely	O
,	O
there	O
are	O
remarkable	O
similarities	O
between	O
the	O
two	O
countries	O
.	O
This	O
may	O
partly	O
correlate	O
with	O
the	O
heavy	O
traffic	O
in	O
foreign	B-KEY
direct	I-KEY
investment	I-KEY
between	O
the	O
two	O
nations	O
.	O
Notable	O
differences	O
are	O
-LRB-	O
1	O
-RRB-	O
across-the-board	O
,	O
US	O
manufacturers	O
are	O
ahead	O
of	O
the	O
UK	B-KEY
firms	O
in	O
computerized	O
integration	O
with	O
units	O
inside	O
and	O
outside	O
manufacturing	O
organizations	O
;	O
-LRB-	O
2	O
-RRB-	O
US	O
manufacturers	O
show	O
higher	O
labour	B-KEY
productivity	I-KEY
,	O
which	O
is	O
consistent	O
with	O
macro-economic	B-KEY
data	I-KEY
,	O
and	O
-LRB-	O
3	O
-RRB-	O
more	O
UK	B-KEY
manufacturers	O
report	O
the	O
use	O
of	O
soft	B-KEY
technologies	I-KEY
such	O
as	O
just-in-time	B-KEY
,	O
total	B-KEY
quality	I-KEY
manufacturing	I-KEY
and	O
manufacturing	B-KEY
cells	I-KEY
.	O
Hypotheses	O
for	O
future	O
investigation	O
are	O
proposed	O

Study	O
of	O
ambiguities	O
inherent	O
to	O
the	O
spectral	B-KEY
analysis	I-KEY
of	O
Voigt	O
profiles-a	O
modified	O
Simplex	O
approach	O
In	O
pulsed	B-KEY
spectrometries	I-KEY
,	O
temporal	B-KEY
transients	I-KEY
are	O
often	O
analyzed	O
directly	O
in	O
the	O
temporal	O
domain	O
,	O
assuming	O
they	O
consist	O
only	O
of	O
purely	O
exponentially	O
decaying	O
sinusoids	O
.	O
When	O
experimental	O
spectra	O
actually	O
consist	O
of	O
Gaussian	O
or	O
Voigt	B-KEY
profiles	I-KEY
-LRB-	O
Gauss-Lorentz	B-KEY
profiles	I-KEY
-RRB-	O
,	O
we	O
show	O
that	O
the	O
direct	O
methods	O
may	O
erroneously	O
interpret	O
such	O
lines	O
as	O
the	O
sum	O
of	O
two	O
or	O
more	O
Lorentzian	O
profiles	O
.	O
Using	O
a	O
Nelder	B-KEY
and	I-KEY
Mead	I-KEY
Simplex	I-KEY
method	I-KEY
,	O
modified	O
by	O
introducing	O
new	O
means	O
to	O
avoid	O
degeneracies	O
and	O
quenchings	O
in	O
secondary	O
minima	O
,	O
we	O
demonstrate	O
that	O
a	O
large	O
number	O
of	O
different	O
solutions	O
can	O
be	O
obtained	O
with	O
equivalent	O
accuracy	B-KEY
over	O
the	O
limited	B-KEY
acquisition	I-KEY
time	I-KEY
interval	I-KEY
,	O
with	O
final	B-KEY
peak	I-KEY
parameters	I-KEY
devoid	O
of	O
physical	O
or	O
chemical	O
meaning	O

Senate	O
to	O
Powell	O
:	O
regulate	O
more	O
-LSB-	O
FCC	B-KEY
-RSB-	O
FCC	B-KEY
Chairman	O
Michael	O
Powell	O
pitched	O
a	O
six-step	O
market-based	O
recovery	B-KEY
plan	I-KEY
to	O
the	O
Senate	O
last	O
week	O
,	O
but	O
two	O
members	O
of	O
the	O
Commerce	O
Committee	O
told	O
him	O
telecom	O
's	O
revival	O
requires	O
more	O
reliance	O
on	O
regulation	O

Sensitivity	B-KEY
calibration	I-KEY
of	O
ultrasonic	O
detectors	O
based	O
using	O
ADD	O
diagrams	O
The	O
paper	O
considers	O
basic	O
problems	O
related	O
to	O
utilization	O
of	O
ADD	B-KEY
diagrams	I-KEY
in	O
calibrating	B-KEY
sensitivity	O
of	O
ultrasonic	B-KEY
detectors	I-KEY
.	O
We	O
suggest	O
that	O
a	O
convenient	O
tool	O
for	O
solving	O
such	O
problems	O
can	O
be	O
the	O
software	B-KEY
package	I-KEY
ADD	O
Universal	O
.	O
Version	O
2.1	O
designed	O
for	O
plotting	O
individual	O
ADD	B-KEY
diagrams	I-KEY
for	O
normal	O
and	O
slanted	B-KEY
transducers	I-KEY
.	O
The	O
software	O
is	O
compatible	O
with	O
the	O
contemporary	O
operational	O
system	O
Windows-95	O
-LRB-	O
98	O
-RRB-	O
.	O
Reference	B-KEY
signals	I-KEY
for	O
calibration	B-KEY
are	O
generated	O
in	O
a	O
sample	O
with	O
cylindrical	B-KEY
holes	I-KEY

Discreteness	B-KEY
and	O
relevance	B-KEY
:	O
a	O
reply	O
to	O
Roman	O
Poznanski	O
In	O
reply	O
to	O
Poznanski	O
-LRB-	O
see	O
ibid.	O
,	O
p.	O
435	O
,	O
2002	O
-RRB-	O
on	O
discreteness	B-KEY
and	O
relevance	B-KEY
,	O
Eliasmith	O
claims	O
that	O
all	O
of	O
the	O
concerns	O
voiced	O
by	O
Poznanski	O
in	O
his	O
reply	O
fail	O
to	O
offer	O
a	O
serious	O
challenge	O
to	O
the	O
idea	O
that	O
continuity	B-KEY
is	O
irrelevant	O
to	O
a	O
good	O
understanding	O
of	O
cognitive	B-KEY
systems	I-KEY
.	O
Eliasmith	O
hopes	O
that	O
it	O
is	O
evident	O
that	O
he	O
does	O
not	O
claim	O
that	O
the	O
process	O
in	O
neural	B-KEY
systems	I-KEY
is	O
discrete	B-KEY
,	O
but	O
rather	O
that	O
a	O
complete	O
characterization	O
of	O
the	O
process	O
can	O
be	O
discrete	B-KEY
;	O
these	O
of	O
course	O
are	O
significantly	O
different	O
claims	O

A	O
3-stage	O
pipelined	O
architecture	O
for	O
multi-view	B-KEY
images	I-KEY
decoder	I-KEY
In	O
this	O
paper	O
,	O
we	O
proposed	O
the	O
architecture	O
of	O
the	O
decoder	O
which	O
implements	O
the	O
multi-view	B-KEY
images	I-KEY
decoding	I-KEY
algorithm	O
.	O
The	O
study	O
of	O
the	O
hardware	B-KEY
structure	I-KEY
of	O
the	O
multi-view	O
image	O
processing	O
has	O
not	O
been	O
accomplished	O
.	O
The	O
proposed	O
multi-view	B-KEY
images	I-KEY
decoder	I-KEY
operates	O
in	O
a	O
three	O
stage	O
pipelined	O
manner	O
and	O
extracts	O
the	O
depth	O
of	O
the	O
pixels	O
of	O
the	O
decoded	O
image	O
every	O
clock	O
.	O
The	O
multi-view	B-KEY
images	I-KEY
decoder	I-KEY
consists	O
of	O
three	O
modules	O
,	O
Node	B-KEY
selector	I-KEY
which	O
transfers	O
the	O
value	O
of	O
the	O
nodes	O
repeatedly	O
and	O
Depth	B-KEY
Extractor	I-KEY
which	O
extracts	O
the	O
depth	O
of	O
each	O
pixel	O
from	O
the	O
four	O
values	O
of	O
the	O
nodes	O
and	O
Affine	B-KEY
Transformer	I-KEY
which	O
generates	O
the	O
projecting	O
position	O
on	O
the	O
image	O
plane	O
from	O
the	O
values	O
of	O
the	O
pixels	O
and	O
the	O
specified	O
viewpoint	B-KEY
.	O
The	O
proposed	O
architecture	O
is	O
designed	O
and	O
simulated	O
by	O
the	O
Max	O
+	O
PlusII	O
design	O
tool	O
and	O
the	O
operating	B-KEY
frequency	I-KEY
is	O
30	B-KEY
MHz	I-KEY
.	O
The	O
image	O
can	O
be	O
constructed	O
in	O
a	O
real	O
time	O
by	O
the	O
decoder	O
with	O
the	O
proposed	O
architecture	O

On	O
biorthogonal	B-KEY
nonuniform	I-KEY
filter	I-KEY
banks	I-KEY
and	O
tree	O
structures	O
This	O
paper	O
concerns	O
biorthogonal	B-KEY
nonuniform	I-KEY
filter	I-KEY
banks	I-KEY
.	O
It	O
is	O
shown	O
that	O
a	O
tree	B-KEY
structured	I-KEY
filter	I-KEY
bank	I-KEY
is	O
biorthogonal	O
if	O
it	O
is	O
equivalent	O
to	O
a	O
tree	B-KEY
structured	I-KEY
filter	I-KEY
bank	I-KEY
whose	O
matching	O
constituent	O
levels	O
on	O
the	O
analysis	O
and	O
synthesis	O
sides	O
are	O
themselves	O
biorthogonal	B-KEY
pairs	I-KEY
.	O
We	O
then	O
show	O
that	O
a	O
stronger	O
statement	O
can	O
be	O
made	O
about	O
dyadic	B-KEY
filter	I-KEY
banks	I-KEY
in	O
general	O
:	O
That	O
a	O
dyadic	B-KEY
filter	I-KEY
bank	I-KEY
is	O
biorthogonal	O
if	O
both	O
the	O
analysis	O
and	O
synthesis	O
banks	O
can	O
be	O
decomposed	O
into	O
dyadic	B-KEY
trees	I-KEY
.	O
We	O
further	O
show	O
that	O
these	O
decompositions	O
are	O
stability	O
and	O
FIR	B-KEY
preserving	I-KEY
.	O
These	O
results	O
,	O
derived	O
for	O
filter	O
banks	O
having	O
filters	O
with	O
rational	B-KEY
transfer	I-KEY
functions	I-KEY
,	O
thus	O
extend	O
some	O
of	O
the	O
earlier	O
comparable	O
results	O
for	O
orthonormal	O
filter	O
banks	O

SubSeven	B-KEY
's	O
Honey	B-KEY
Pot	I-KEY
program	I-KEY
A	O
serious	O
security	B-KEY
threat	I-KEY
today	O
are	O
malicious	B-KEY
executables	I-KEY
,	O
especially	O
new	O
,	O
unseen	O
malicious	B-KEY
executables	I-KEY
often	O
arriving	O
as	O
email	B-KEY
attachments	I-KEY
.	O
These	O
new	O
malicious	B-KEY
executables	I-KEY
are	O
created	O
at	O
the	O
rate	O
of	O
thousands	O
every	O
year	O
and	O
pose	O
a	O
serious	O
threat	O
.	O
Current	O
anti-virus	B-KEY
systems	I-KEY
attempt	O
to	O
detect	O
these	O
new	O
malicious	O
programs	O
with	O
heuristics	O
generated	O
by	O
hand	O
.	O
This	O
approach	O
is	O
costly	O
and	O
often	O
ineffective	O
.	O
We	O
introduce	O
the	O
Trojan	B-KEY
Horse	I-KEY
SubSeven	B-KEY
,	O
its	O
capabilities	O
and	O
influence	O
over	O
intrusion	B-KEY
detection	I-KEY
systems	I-KEY
.	O
A	O
Honey	B-KEY
Pot	I-KEY
program	I-KEY
is	O
implemented	O
,	O
simulating	O
the	O
SubSeven	B-KEY
Server	O
.	O
The	O
Honey	B-KEY
Pot	I-KEY
Program	I-KEY
provides	O
feedback	O
and	O
stores	O
data	O
to	O
and	O
from	O
the	O
SubSeven	B-KEY
's	O
client	O

Fractional	B-KEY
differentiation	I-KEY
in	O
passive	B-KEY
vibration	I-KEY
control	I-KEY
From	O
a	O
single-degree-of-freedom	O
model	O
used	O
to	O
illustrate	O
the	O
concept	O
of	O
vibration	B-KEY
isolation	I-KEY
,	O
a	O
method	O
to	O
transform	O
the	O
design	O
for	O
a	O
suspension	B-KEY
into	O
a	O
design	O
for	O
a	O
robust	B-KEY
controller	I-KEY
is	O
presented	O
.	O
Fractional	B-KEY
differentiation	I-KEY
is	O
used	O
to	O
model	O
the	O
viscoelastic	B-KEY
behaviour	I-KEY
of	O
the	O
suspension	B-KEY
.	O
The	O
use	O
of	O
fractional	B-KEY
differentiation	I-KEY
not	O
only	O
permits	O
optimisation	O
of	O
just	O
four	O
suspension	B-KEY
parameters	O
,	O
showing	O
the	O
`	O
compactness	O
'	O
of	O
the	O
fractional	O
derivative	O
operator	O
,	O
but	O
also	O
leads	O
to	O
robustness	O
of	O
the	O
suspension	B-KEY
's	O
performance	O
to	O
uncertainty	O
of	O
the	O
sprung	B-KEY
mass	I-KEY
.	O
As	O
an	O
example	O
,	O
an	O
engine	B-KEY
suspension	I-KEY
is	O
studied	O

Speedera	O
:	O
Web	O
without	O
the	O
wait	O
There	O
's	O
no	O
greater	O
testament	O
to	O
the	O
utility	O
of	O
the	O
Internet	O
than	O
the	O
fact	O
that	O
hundreds	O
of	O
millions	O
of	O
people	O
worldwide	O
are	O
willing	O
to	O
wait	O
for	O
Web	O
pages	O
as	O
they	O
build	O
incrementally	O
on	O
screen	O
.	O
But	O
while	O
users	O
may	O
put	O
up	O
with	O
the	O
``	O
World	O
Wide	O
Wait	O
,	O
''	O
they	O
definitely	O
do	O
n't	O
like	O
it	O
.	O
That	O
's	O
where	O
Content	B-KEY
Delivery	I-KEY
Networks	I-KEY
come	O
in	O
.	O
CDNs	O
ca	O
n't	O
turn	O
a	O
footpath	O
into	O
a	O
freeway	O
,	O
but	O
they	O
can	O
help	O
data	O
in	O
transit	O
take	O
advantage	O
of	O
shortcuts	O
and	O
steer	O
clear	O
of	O
traffic	O
jams	O
.	O
And	O
while	O
enhancing	O
the	O
responsiveness	O
of	O
Web	B-KEY
interaction	I-KEY
,	O
CDNs	O
also	O
enhance	O
the	O
prospects	O
of	O
their	O
clients	O
,	O
who	O
need	O
engaged	O
visitors	O
to	O
keep	O
their	O
Web-based	B-KEY
business	I-KEY
models	I-KEY
afloat	O
.	O
``	O
Our	O
mission	O
is	O
to	O
improve	O
the	O
quality	O
of	O
the	O
Internet	B-KEY
experience	I-KEY
for	O
end-users	O
,	O
''	O
says	O
Gordon	O
Smith	O
,	O
vice	O
president	O
of	O
marketing	O
at	O
Speedera	O
Networks	O
in	O
Santa	O
Clara	O
,	O
California	O
,	O
``	O
and	O
to	O
enable	O
Web-site	B-KEY
operators	I-KEY
to	O
provide	O
better	O
delivery	B-KEY
quality	I-KEY
,	O
performance	O
,	O
scalability	B-KEY
,	O
and	O
security	B-KEY
through	O
an	O
outsourced	B-KEY
service	I-KEY
model	I-KEY
that	O
slashes	O
IT	O
costs	O
.	O
''	O

Behavior	O
of	O
Runge-Kutta	B-KEY
discretizations	I-KEY
near	O
equilibria	B-KEY
of	O
index	B-KEY
2	I-KEY
differential	I-KEY
algebraic	I-KEY
systems	I-KEY
We	O
analyze	O
Runge-Kutta	B-KEY
discretizations	I-KEY
applied	O
to	O
index	O
2	O
differential	O
algebraic	O
equations	O
-LRB-	O
DAE	O
's	O
-RRB-	O
near	O
equilibria	B-KEY
.	O
We	O
compare	O
the	O
geometric	B-KEY
properties	I-KEY
of	O
the	O
numerical	O
and	O
the	O
exact	O
solutions	O
.	O
It	O
is	O
shown	O
that	O
projected	O
and	O
half-explicit	B-KEY
Runge-Kutta	I-KEY
methods	I-KEY
reproduce	O
the	O
qualitative	O
features	O
of	O
the	O
continuous	B-KEY
system	I-KEY
in	O
the	O
vicinity	O
of	O
an	O
equilibrium	O
correctly	O
.	O
The	O
proof	O
combines	O
cut-off	O
and	O
scaling	B-KEY
techniques	I-KEY
for	O
index	O
2	O
differential	O
algebraic	O
equations	O
with	O
some	O
invariant	B-KEY
manifold	I-KEY
results	O
of	O
Schropp	O
-LRB-	O
Geometric	B-KEY
properties	I-KEY
of	O
Runge-Kutta	B-KEY
discretizations	I-KEY
for	O
index	O
2	O
differential	O
algebraic	O
equations	O
,	O
Konstanzer	O
Schriften	O
in	O
Mathematik	O
und	O
Informatik	O
128	O
-RRB-	O
and	O
classical	O
results	O
for	O
discretized	B-KEY
ordinary	I-KEY
differential	I-KEY
equations	I-KEY

Laser-based	O
internal	B-KEY
profile	I-KEY
measurement	O
system	O
An	O
automatic	O
laser-based	O
system	O
to	O
measure	O
the	O
internal	B-KEY
profiles	I-KEY
of	O
various	O
structures	O
has	O
been	O
developed	O
.	O
The	O
system	O
uses	O
a	O
point	B-KEY
laser	I-KEY
source	I-KEY
through	O
a	O
rotating	B-KEY
optical	I-KEY
device	I-KEY
fixed	O
on	O
to	O
a	O
laser	O
measurement	O
meter	O
.	O
A	O
notebook	O
computer	O
with	O
custom	O
software	O
is	O
used	O
to	O
control	O
the	O
laser	B-KEY
meter	I-KEY
and	O
rotating	B-KEY
device	I-KEY
to	O
estimate	O
the	O
scanned	O
profile	O
shape	O
and	O
to	O
determine	O
the	O
resulting	O
cross-section	O
area	O
.	O
The	O
information	O
provided	O
by	O
this	O
system	O
is	O
essential	O
to	O
construction	O
industry	O
,	O
including	O
window	O
and	O
door	O
builders	O
;	O
the	O
glass	O
,	O
panel	O
,	O
board	O
,	O
and	O
floor	B-KEY
tile	I-KEY
manufacturers	I-KEY
;	O
carpet	B-KEY
venders	I-KEY
;	O
and	O
building	B-KEY
contractors	I-KEY
for	O
cost	B-KEY
estimation	I-KEY
and	O
production	B-KEY
control	I-KEY
.	O
As	O
a	O
result	O
,	O
the	O
lead	O
time	O
for	O
delivering	O
the	O
customized	B-KEY
windowpanes	I-KEY
,	O
woodwork	O
,	O
floor	O
tiles	O
,	O
and	O
ceilings	O
can	O
be	O
reduced	O
.	O
Applications	O
of	O
this	O
system	O
for	O
measuring	O
the	O
shapes	O
of	O
window	B-KEY
frames	I-KEY
and	O
floor	O
plans	O
are	O
described	O
and	O
demonstrated	O
.	O
The	O
measurement	O
accuracy	O
is	O
evaluated	O
and	O
analyzed	O
.	O
Results	O
have	O
indicated	O
that	O
the	O
measurement	O
accuracy	O
can	O
be	O
achieved	O
within	O
4	O
%	O
of	O
the	O
measurement	O
distance	O
,	O
for	O
typical	O
window	O
designs	O
and	O
floor	O
patterns	O
required	O
by	O
major	O
window	O
manufacturers	O
.	O
Recommendations	O
to	O
improve	O
the	O
system	O
are	O
also	O
included	O

Government	B-KEY
budget	I-KEY
and	O
accounting	B-KEY
information	I-KEY
policy	I-KEY
and	O
practice	O
in	O
Taiwan	B-KEY
The	O
principal	O
government	B-KEY
budget	I-KEY
and	O
accounting	B-KEY
information	I-KEY
policies	I-KEY
in	O
Taiwan	B-KEY
are	O
founded	O
on	O
the	O
ability	O
to	O
provide	O
integrated	O
,	O
consistent	O
,	O
and	O
timely	O
information	O
for	O
government	B-KEY
managers	I-KEY
to	O
make	O
more	O
rational	B-KEY
decisions	I-KEY
concerning	O
national	B-KEY
resource	I-KEY
allocation	I-KEY
and	O
evaluation	O
.	O
A	O
specific	O
accounting	O
organization	O
system	O
has	O
been	O
designed	O
for	O
this	O
purpose	O
.	O
This	O
paper	O
analyzes	O
information	O
policies	O
and	O
practices	O
according	O
to	O
the	O
relevant	O
laws	O
and	O
regulations	O
,	O
identifies	O
issues	O
regarding	O
the	O
policies	O
,	O
and	O
presents	O
strategies	O
to	O
resolve	O
the	O
issues	O

A	O
robust	B-KEY
H/sub	O
infinity	O
/	O
control	O
approach	O
for	O
induction	O
motors	O
This	O
paper	O
deals	O
with	O
the	O
robustness	B-KEY
and	O
stability	B-KEY
of	O
an	O
induction	B-KEY
motor	I-KEY
control	I-KEY
structure	O
against	O
internal	O
and	O
external	B-KEY
disturbances	I-KEY
.	O
In	O
the	O
proposed	O
control	O
scheme	O
,	O
we	O
have	O
used	O
an	O
H/sub	O
infinity	O
/	O
controller	O
with	O
field	B-KEY
orientation	I-KEY
and	O
input-output	B-KEY
linearization	I-KEY
to	O
achieve	O
the	O
above-specified	O
features	O
.	O
Simulation	O
results	O
are	O
included	O
to	O
illustrate	O
the	O
control	O
approach	O
performances	O

Asymptotical	B-KEY
stability	I-KEY
in	O
discrete-time	O
neural	O
networks	O
In	O
this	O
work	O
,	O
we	O
present	O
a	O
proof	O
of	O
the	O
existence	O
of	O
a	O
fixed	B-KEY
point	I-KEY
and	O
a	O
generalized	B-KEY
sufficient	I-KEY
condition	I-KEY
that	O
guarantees	O
the	O
stability	B-KEY
of	O
it	O
in	O
discrete-time	B-KEY
neural	I-KEY
networks	I-KEY
by	O
using	O
the	O
Lyapunov	B-KEY
function	I-KEY
method	I-KEY
.	O
We	O
also	O
show	O
that	O
for	O
both	O
symmetric	O
and	O
asymmetric	B-KEY
connections	I-KEY
,	O
the	O
unique	B-KEY
attractor	I-KEY
is	O
a	O
fixed	B-KEY
point	I-KEY
when	O
several	O
conditions	O
are	O
satisfied	O
.	O
This	O
is	O
an	O
extended	O
result	O
of	O
Chen	O
and	O
Aihara	O
-LRB-	O
see	O
Physica	O
D	O
,	O
vol	O
.	O
104	O
,	O
no.	O
3/4	O
,	O
p.	O
286-325	O
,	O
1997	O
-RRB-	O
.	O
In	O
particular	O
,	O
we	O
further	O
study	O
the	O
stability	B-KEY
of	O
equilibrium	O
in	O
discrete-time	B-KEY
neural	I-KEY
networks	I-KEY
with	O
the	O
connection	B-KEY
weight	I-KEY
matrix	I-KEY
in	O
form	O
of	O
an	O
interval	B-KEY
matrix	I-KEY
.	O
Finally	O
,	O
several	O
examples	O
are	O
shown	O
to	O
illustrate	O
and	O
reinforce	O
our	O
theory	O

Blending	O
parametric	O
patches	O
with	O
subdivision	B-KEY
surfaces	I-KEY
In	O
this	O
paper	O
the	O
problem	O
of	O
blending	O
parametric	O
surfaces	O
using	O
subdivision	B-KEY
patches	I-KEY
is	O
discussed	O
.	O
A	O
new	O
approach	O
,	O
named	O
removing-boundary	O
,	O
is	O
presented	O
to	O
generate	O
piecewise-smooth	O
subdivision	B-KEY
surfaces	I-KEY
through	O
discarding	O
the	O
outmost	O
quadrilaterals	O
of	O
the	O
open	O
meshes	O
derived	O
by	O
each	O
subdivision	O
step	O
.	O
Then	O
the	O
approach	O
is	O
employed	O
both	O
to	O
blend	O
parametric	B-KEY
bicubic	I-KEY
B-spline	I-KEY
surfaces	I-KEY
and	O
to	O
fill	O
n-sided	O
holes	O
.	O
It	O
is	O
easy	O
to	O
produce	O
piecewise-smooth	O
subdivision	B-KEY
surfaces	I-KEY
with	O
both	O
convex	O
and	O
concave	O
corners	O
on	O
the	O
boundary	O
,	O
and	O
limit	O
surfaces	O
are	O
guaranteed	O
to	O
be	O
C/sup	O
2	O
/	O
continuous	O
on	O
the	O
boundaries	O
except	O
for	O
a	O
few	O
singular	O
points	O
by	O
the	O
removing-boundary	O
approach	O
.	O
Thus	O
the	O
blending	O
method	O
is	O
very	O
efficient	O
and	O
the	O
blending	O
surface	O
generated	O
is	O
of	O
good	O
effect	O

The	O
Internet	B-KEY
,	O
knowledge	B-KEY
and	O
the	O
academy	B-KEY
As	O
knowledge	B-KEY
is	O
released	O
from	O
the	O
bounds	O
of	O
libraries	O
,	O
as	O
research	O
becomes	O
no	O
longer	O
confined	O
to	O
the	O
academy	B-KEY
,	O
and	O
education/certification	O
is	O
available	O
,	O
any	O
time/any	O
place	O
,	O
the	O
university	O
and	O
the	O
faculty	O
must	O
redefine	O
themselves	O
.	O
Liberal	B-KEY
studies	I-KEY
,	O
once	O
the	O
core	O
,	O
and	O
currently	O
eschewed	O
in	O
favor	O
of	O
science	O
and	O
technology	O
,	O
will	O
be	O
reborn	O
in	O
those	O
institutions	O
that	O
can	O
rise	O
above	O
the	O
mundane	O
and	O
embrace	O
an	O
emerging	O
``	O
third	O
culture	O
''	O

LAN-based	B-KEY
building	I-KEY
maintenance	I-KEY
and	I-KEY
surveillance	I-KEY
robot	I-KEY
The	O
building	O
and	O
construction	O
industry	O
is	O
the	O
major	O
industry	O
of	O
Hong	O
Kong	O
as	O
in	O
many	O
developed	O
countries	O
around	O
the	O
world	O
.	O
After	O
the	O
commissioning	O
of	O
a	O
high-rise	B-KEY
building	I-KEY
or	O
a	O
large	O
estate	O
,	O
substantial	O
manpower	O
,	O
both	O
inside	O
the	O
management	O
centre	O
under	O
a	O
standby	O
manner	O
,	O
as	O
well	O
as	O
surveillance	O
for	O
security	B-KEY
purposes	I-KEY
around	O
the	O
whole	O
building	O
,	O
is	O
required	O
for	O
daily	O
operation	O
to	O
ensure	O
a	O
quality	O
environment	O
for	O
the	O
occupants	O
.	O
If	O
the	O
surveillance	O
job	O
can	O
be	O
done	O
by	O
robots	O
,	O
the	O
efficiency	O
can	O
be	O
highly	O
enhanced	O
,	O
resulting	O
in	O
a	O
great	O
saving	O
of	O
manpower	O
and	O
the	O
improved	O
safety	O
of	O
the	O
management	O
staff	O
as	O
a	O
by-product	O
.	O
Furthermore	O
,	O
if	O
the	O
robot	O
can	O
retrieve	O
commands	O
from	O
the	O
building	B-KEY
management	I-KEY
system	I-KEY
via	O
a	O
local	B-KEY
area	I-KEY
network	I-KEY
-LRB-	O
LAN	O
-RRB-	O
,	O
further	O
savings	O
in	O
manpower	O
can	O
be	O
achieved	O
in	O
terms	O
of	O
first-line	B-KEY
fault	I-KEY
attendance	I-KEY
by	O
human	O
management	O
staff	O
.	O
This	O
paper	O
describes	O
the	O
development	O
of	O
a	O
robot	O
prototype	O
here	O
in	O
Hong	O
Kong	O
,	O
which	O
can	O
handle	O
some	O
daily	O
routine	O
maintenance	O
works	O
and	O
surveillance	O
responsibilities	O
.	O
The	O
hardware	B-KEY
structure	I-KEY
of	O
the	O
robot	O
and	O
its	O
on-board	O
devices	O
are	O
described	O
.	O
Real-time	O
images	O
captured	O
by	O
a	O
camera	O
on	O
the	O
robot	O
with	O
pan/tilt/zoom	B-KEY
functions	I-KEY
can	O
be	O
transmitted	O
back	O
to	O
the	O
central	O
management	O
office	O
via	O
a	O
local	B-KEY
area	I-KEY
network	I-KEY
.	O
The	O
interface	O
between	O
the	O
robot	O
and	O
the	O
building	O
automation	O
system	O
-LRB-	O
BAS	O
-RRB-	O
of	O
the	O
building	O
is	O
discussed	O
.	O
This	O
is	O
the	O
first	O
key	O
achievement	O
of	O
this	O
project	O
with	O
a	O
strong	O
implication	O
on	O
reducing	O
the	O
number	O
of	O
human	O
staff	O
to	O
manage	O
a	O
modem	O
building	O
.	O
Teleoperation	B-KEY
of	O
the	O
robot	O
via	O
the	O
Internet	O
or	O
intranet	O
is	O
also	O
possible	O
,	O
which	O
is	O
the	O
second	O
achievement	O
of	O
this	O
project	O
.	O
Finally	O
,	O
the	O
robot	O
can	O
identify	O
its	O
physical	O
position	O
inside	O
the	O
building	O
by	O
a	O
landmark	B-KEY
recognition	I-KEY
method	I-KEY
based	O
on	O
standard	O
CAD	O
drawings	O
,	O
which	O
is	O
the	O
third	O
achievement	O
of	O
this	O
project	O
.	O
The	O
main	O
goal	O
of	O
this	O
paper	O
is	O
not	O
the	O
description	O
of	O
some	O
groundbreaking	O
technology	O
in	O
robotic	O
development	O
.	O
It	O
is	O
mainly	O
intended	O
to	O
convince	O
building	O
designers	O
and	O
managers	O
to	O
incorporate	O
robotic	O
systems	O
when	O
they	O
are	O
managing	O
modem	O
buildings	O
to	O
save	O
manpower	O
and	O
improve	O
efficiency	O

Modular	O
and	O
visual	B-KEY
specification	I-KEY
of	O
hybrid	B-KEY
systems	I-KEY
:	O
an	O
introduction	O
to	O
HyCharts	B-KEY
Visual	B-KEY
description	I-KEY
techniques	I-KEY
are	O
particularly	O
important	O
for	O
the	O
design	O
of	O
hybrid	B-KEY
systems	I-KEY
,	O
because	O
specifications	O
of	O
such	O
systems	O
usually	O
have	O
to	O
be	O
discussed	O
between	O
engineers	O
from	O
a	O
number	O
of	O
different	O
disciplines	O
.	O
Modularity	O
is	O
vital	O
for	O
hybrid	B-KEY
systems	I-KEY
not	O
only	O
because	O
it	O
allows	O
to	O
handle	O
large	O
systems	O
,	O
but	O
also	O
because	O
it	O
permits	O
to	O
think	O
in	O
terms	O
of	O
components	B-KEY
,	O
which	O
is	O
familiar	O
to	O
engineers	O
.	O
Based	O
on	O
two	O
different	O
interpretations	O
for	O
hierarchic	B-KEY
graphs	I-KEY
and	O
on	O
a	O
clear	O
hybrid	B-KEY
computation	I-KEY
model	I-KEY
,	O
we	O
develop	O
HyCharts	B-KEY
.	O
HyCharts	B-KEY
consist	O
of	O
two	O
modular	O
visual	O
formalisms	O
,	O
one	O
for	O
the	O
specification	O
of	O
the	O
architecture	O
and	O
one	O
for	O
the	O
specification	O
of	O
the	O
behavior	O
of	O
hybrid	B-KEY
systems	I-KEY
.	O
The	O
operators	O
on	O
hierarchic	B-KEY
graphs	I-KEY
enable	O
us	O
to	O
give	O
a	O
surprisingly	O
simple	O
denotational	B-KEY
semantics	I-KEY
for	O
many	O
concepts	O
known	O
from	O
statechart-like	O
formalisms	O
.	O
Due	O
to	O
a	O
very	O
general	O
composition	O
operator	O
,	O
HyCharts	B-KEY
can	O
easily	O
be	O
composed	O
with	O
description	O
techniques	O
from	O
other	O
engineering	O
disciplines	O
.	O
Such	O
heterogeneous	B-KEY
system	I-KEY
specifications	I-KEY
seem	O
to	O
be	O
particularly	O
appropriate	O
for	O
hybrid	B-KEY
systems	I-KEY
because	O
of	O
their	O
interdisciplinary	O
character	O

Soft	O
options	O
for	O
software	B-KEY
upgrades	I-KEY
?	O
Several	O
new	O
products	O
claim	O
to	O
take	O
the	O
work	O
out	O
of	O
installing	O
software	O
and	O
patches	O
,	O
and	O
even	O
migrating	O
operating	O
systems	O
.	O
Software	O
migration	O
products	O
fall	O
into	O
two	O
broad	O
categories	O
.	O
The	O
drive	O
imaging	O
type	O
is	O
designed	O
to	O
make	O
exact	O
copies	O
of	O
a	O
hard	O
disk	O
,	O
either	O
an	O
entire	O
drive	O
or	O
certain	O
directories	O
,	O
so	O
you	O
can	O
use	O
it	O
to	O
back	O
up	O
data	O
.	O
The	O
application	O
management	O
type	O
is	O
designed	O
for	O
more	O
incremental	O
upgrades	O
and	O
often	O
provides	O
additional	O
features	O
such	O
as	O
the	O
ability	O
to	O
monitor	O
or	O
control	O
users	O
'	O
access	O
to	O
applications	O

From	O
FREE	O
to	O
FEE	O
-LSB-	O
online	B-KEY
advertising	I-KEY
market	I-KEY
-RSB-	O
As	O
the	O
online	B-KEY
advertising	I-KEY
market	I-KEY
continues	O
to	O
struggle	O
,	O
many	O
online	O
content	O
marketers	O
are	O
wrestling	O
with	O
the	O
issue	O
of	O
how	O
to	O
add	O
at	O
least	O
some	O
level	O
of	O
paid	B-KEY
subscription	I-KEY
income	I-KEY
to	O
their	O
revenue	O
mix	O
in	O
order	O
to	O
reach	O
or	O
improve	O
profitability	O
.	O
Since	O
the	O
business	O
of	O
selling	B-KEY
content	I-KEY
online	I-KEY
is	O
still	O
in	O
its	O
infancy	O
,	O
and	O
many	O
consumers	O
clearly	O
still	O
think	O
of	O
Web	O
content	O
as	O
simply	O
and	O
rightfully	O
free	O
,	O
few	O
roadmaps	O
are	O
available	O
to	O
show	O
the	O
way	O
to	O
effective	O
marketing	B-KEY
strategies	I-KEY
,	O
but	O
some	O
guiding	O
principles	O
have	O
emerged	O

Acceptance	O
of	O
a	O
price	O
discount	O
:	O
the	O
role	O
of	O
the	O
semantic	O
relatedness	O
between	O
purchases	O
and	O
the	O
comparative	B-KEY
price	I-KEY
format	I-KEY
Two	O
studies	O
are	O
reported	O
where	O
people	O
are	O
asked	O
to	O
accept	O
or	O
not	O
a	O
price	O
reduction	O
on	O
a	O
target	O
product	O
.	O
In	O
the	O
high	O
-LRB-	O
low	O
-RRB-	O
relative	O
saving	O
version	O
,	O
the	O
regular	O
price	O
of	O
the	O
target	O
product	O
is	O
low	O
-LRB-	O
high	O
-RRB-	O
.	O
In	O
both	O
versions	O
,	O
the	O
absolute	O
value	O
of	O
the	O
price	O
reduction	O
is	O
the	O
same	O
as	O
well	O
as	O
the	O
total	O
of	O
regular	O
prices	O
of	O
planned	B-KEY
purchases	I-KEY
.	O
As	O
first	O
reported	O
by	O
Tversky	O
and	O
Kahneman	O
-LRB-	O
1981	O
-RRB-	O
,	O
findings	O
show	O
that	O
the	O
majority	O
of	O
people	O
accept	O
the	O
price	O
discount	O
in	O
the	O
high-relative	O
saving	O
version	O
whereas	O
the	O
minority	O
do	O
it	O
in	O
the	O
low	O
one	O
.	O
In	O
Study	O
1	O
,	O
findings	O
show	O
that	O
the	O
previous	O
preference	B-KEY
reversal	I-KEY
disappears	O
when	O
planned	B-KEY
purchases	I-KEY
are	O
strongly	O
related	O
.	O
Also	O
,	O
a	O
previously	O
unreported	O
preference	B-KEY
reversal	I-KEY
is	O
found	O
.	O
The	O
majority	O
of	O
people	O
accept	O
the	O
price	O
discount	O
when	O
the	O
products	O
are	O
weakly	O
related	O
whereas	O
the	O
minority	O
accept	O
when	O
the	O
products	O
are	O
strongly	O
related	O
.	O
In	O
Study	O
2	O
,	O
findings	O
show	O
that	O
the	O
classic	O
preference	B-KEY
reversal	I-KEY
disappears	O
as	O
a	O
function	O
of	O
the	O
comparative	B-KEY
price	I-KEY
format	I-KEY
.	O
Also	O
,	O
another	O
previously	O
unreported	O
preference	B-KEY
reversal	I-KEY
is	O
found	O
.	O
When	O
the	O
offered	O
price	O
reduction	O
relates	O
to	O
a	O
low-priced	B-KEY
product	I-KEY
,	O
people	O
are	O
more	O
inclined	O
to	O
accept	O
it	O
with	O
a	O
control	O
than	O
a	O
minimal	O
comparative	B-KEY
price	I-KEY
format	I-KEY
.	O
Findings	O
reported	O
in	O
Studies	O
1	O
and	O
2	O
are	O
interpreted	O
in	O
terms	O
of	O
mental	B-KEY
accounting	I-KEY
shifts	I-KEY

UPSILON	B-KEY
:	O
universal	B-KEY
programming	I-KEY
system	I-KEY
with	O
incomplete	O
lazy	O
object	O
notation	O
This	O
paper	O
presents	O
a	O
new	O
model	O
of	O
computation	O
that	O
differs	O
from	O
prior	O
models	O
in	O
that	O
it	O
emphasizes	O
data	O
over	O
flow	O
control	O
,	O
has	O
no	O
named	O
variables	O
and	O
has	O
an	O
object-oriented	B-KEY
flavor	I-KEY
.	O
We	O
prove	O
that	O
this	O
model	O
is	O
a	O
complete	O
and	O
confluent	O
acceptable	O
programming	B-KEY
system	I-KEY
and	O
has	O
a	O
usable	B-KEY
type	I-KEY
theory	I-KEY
.	O
A	O
new	O
data	B-KEY
synchronization	I-KEY
primitive	I-KEY
is	O
introduced	O
in	O
order	O
to	O
achieve	O
the	O
above	O
properties	O
.	O
Subtle	O
variations	O
of	O
the	O
model	O
are	O
shown	O
to	O
fall	O
short	O
of	O
having	O
all	O
these	O
necessary	O
properties	O

PageFlex	B-KEY
+	O
MediaRich	B-KEY
=	O
PageRich	B-KEY
Layout	B-KEY
and	O
graphics	B-KEY
innovators	O
collaborate	O
on	O
fully	O
variable	O
combination	O
.	O
Pageflex	B-KEY
and	O
Equilibrium	O
have	O
melded	O
their	O
respective	O
EDIT	O
and	O
MediaRich	B-KEY
technologies	O
to	O
make	O
a	O
variable-data	O
composition	B-KEY
engine	O
with	O
a	O
Web	O
interface	O
.	O
Though	O
a	O
first-generation	O
effort	O
,	O
it	O
shows	O
substantial	O
promise	O

Chemical	B-KEY
production	I-KEY
in	O
the	O
superlative	B-KEY
-LSB-	O
formaldehyde	O
plant	O
process	B-KEY
control	I-KEY
system	I-KEY
and	O
remote	O
I/O	O
system	O
-RSB-	O
BASF	B-KEY
commissioned	O
the	O
largest	O
formaldehyde	O
production	O
plant	O
in	O
the	O
world	O
,	O
in	O
December	O
2000	O
,	O
with	O
an	O
annual	O
capacity	O
of	O
180000	O
t	O
.	O
The	O
new	O
plant	O
,	O
built	O
to	O
meet	O
the	O
growing	O
demand	O
for	O
formaldehyde	O
,	O
sets	O
new	O
standards	O
.	O
Its	O
size	O
,	O
technology	O
and	O
above	O
all	O
its	O
cost-effectiveness	O
give	O
it	O
a	O
leading	O
position	O
internationally	O
.	O
To	O
maintain	O
such	O
high	O
standards	O
by	O
the	O
automation	B-KEY
technology	I-KEY
,	O
in	O
addition	O
to	O
the	O
trail-blazing	B-KEY
Simatic	I-KEY
PCS	I-KEY
7	I-KEY
process	B-KEY
control	I-KEY
system	I-KEY
from	O
Siemens	B-KEY
,	O
BASF	B-KEY
selected	O
the	O
innovative	O
remote	O
I/O	O
system	O
I.S.	O
1	O
from	O
R.	O
STAHL	B-KEY
Schaltgerate	I-KEY
GmbH	I-KEY
to	I-KEY
record	O
and	O
to	O
output	O
field	O
signals	O
in	O
hazardous	O
areas	O
Zone	O
1	O
and	O
2	O
.	O
This	O
combination	O
completely	O
satisfied	O
all	O
technical	O
requirements	O
and	O
also	O
had	O
the	O
best	O
price-performance	B-KEY
ratio	I-KEY
of	O
all	O
the	O
solutions	O
.	O
25	O
remote	O
I/O	O
field	O
stations	O
were	O
designed	O
and	O
matched	O
to	O
the	O
needs	O
of	O
the	O
formaldehyde	O
plant	O

Programmatic	O
efforts	O
encouraging	O
women	B-KEY
to	O
enter	O
the	O
information	O
technology	O
workforce	O
For	O
over	O
a	O
decade	O
the	O
National	B-KEY
Science	I-KEY
Foundation	I-KEY
-LRB-	O
NSF	O
-RRB-	O
has	O
been	O
supporting	O
projects	O
designed	O
to	O
improve	O
opportunities	O
for	O
women	B-KEY
in	O
computing	B-KEY
.	O
From	O
an	O
initial	O
emphasis	O
on	O
increasing	O
the	O
number	O
of	O
women	B-KEY
in	O
graduate	B-KEY
school	I-KEY
studying	O
computer	B-KEY
science	O
and	O
engineering	B-KEY
,	O
NSF	O
's	O
current	O
emphasis	O
has	O
broadened	O
to	O
include	O
research	O
studies	O
examining	O
the	O
underlying	O
reasons	O
why	O
women	B-KEY
are	O
underrepresented	O
in	O
the	O
information	O
technology	O
-LRB-	O
IT	O
-RRB-	O
workforce	O
.	O
This	O
paper	O
describes	O
the	O
recent	O
history	B-KEY
of	O
NSF	O
's	O
activities	O
in	O
this	O
area	O
and	O
the	O
subsequent	O
emergence	O
of	O
a	O
research	O
portfolio	O
addressing	O
the	O
underrepresentation	O
issue	O

Mammogram	B-KEY
synthesis	I-KEY
using	O
a	O
3D	B-KEY
simulation	I-KEY
.	O
II	O
.	O
Evaluation	O
of	O
synthetic	B-KEY
mammogram	I-KEY
texture	I-KEY
We	O
have	O
evaluated	O
a	O
method	O
for	O
synthesizing	O
mammograms	O
by	O
comparing	O
the	O
texture	O
of	O
clinical	O
and	O
synthetic	O
mammograms	O
.	O
The	O
synthesis	O
algorithm	O
is	O
based	O
upon	O
simulations	O
of	O
breast	O
tissue	O
and	O
the	O
mammographic	O
imaging	O
process	O
.	O
Mammogram	O
texture	O
was	O
synthesized	O
by	O
projections	O
of	O
simulated	O
adipose	B-KEY
tissue	I-KEY
compartments	I-KEY
.	O
It	O
was	O
hypothesized	O
that	O
the	O
synthetic	O
and	O
clinical	O
texture	O
have	O
similar	O
properties	O
,	O
assuming	O
that	O
the	O
mammogram	O
texture	O
reflects	O
the	O
3D	B-KEY
tissue	I-KEY
distribution	I-KEY
.	O
The	O
size	O
of	O
the	O
projected	O
compartments	O
was	O
computed	O
by	O
mathematical	B-KEY
morphology	I-KEY
.	O
The	O
texture	O
energy	O
and	O
fractal	B-KEY
dimension	I-KEY
were	O
also	O
computed	O
and	O
analyzed	O
in	O
terms	O
of	O
the	O
distribution	O
of	O
texture	O
features	O
within	O
four	O
different	O
tissue	O
regions	O
in	O
clinical	O
and	O
synthetic	O
mammograms	O
.	O
Comparison	O
of	O
the	O
cumulative	B-KEY
distributions	I-KEY
of	O
the	O
mean	O
features	O
computed	O
from	O
95	O
mammograms	O
showed	O
that	O
the	O
synthetic	B-KEY
images	I-KEY
simulate	O
the	O
mean	O
features	O
of	O
the	O
texture	O
of	O
clinical	O
mammograms	O
.	O
Correlation	O
of	O
clinical	O
and	O
synthetic	O
texture	O
feature	O
histograms	O
,	O
averaged	O
over	O
all	O
images	O
,	O
showed	O
that	O
the	O
synthetic	B-KEY
images	I-KEY
can	O
simulate	O
the	O
range	O
of	O
features	O
seen	O
over	O
a	O
large	O
group	O
of	O
mammograms	O
.	O
The	O
best	O
agreement	O
with	O
clinical	O
texture	O
was	O
achieved	O
for	O
simulated	O
compartments	O
with	O
radii	O
of	O
4-13	O
.3	O
mm	O
in	O
predominantly	O
adipose	O
tissue	O
regions	O
,	O
and	O
radii	O
of	O
2.7-5	O
.33	O
and	O
1.3-2	O
.7	O
mm	O
in	O
retroareolar	O
and	O
dense	B-KEY
fibroglandular	I-KEY
tissue	I-KEY
regions	I-KEY
,	O
respectively	O

Does	O
classicism	B-KEY
explain	O
universality	B-KEY
?	O
Arguments	O
against	O
a	O
pure	O
classical	B-KEY
component	O
of	O
mind	O
One	O
of	O
the	O
hallmarks	O
of	O
human	B-KEY
cognition	I-KEY
is	O
the	O
capacity	O
to	O
generalize	O
over	O
arbitrary	O
constituents	O
.	O
Marcus	O
-LRB-	O
Cognition	O
66	O
,	O
p.	O
153	O
;	O
Cognitive	O
Psychology	O
37	O
,	O
p.	O
243	O
,	O
1998	O
-RRB-	O
argued	O
that	O
this	O
capacity	O
,	O
called	O
``	O
universal	B-KEY
generalization	O
''	O
-LRB-	O
universality	O
-RRB-	O
,	O
is	O
not	O
supported	O
by	O
connectionist	O
models	O
.	O
Instead	O
,	O
universality	B-KEY
is	O
best	O
explained	O
by	O
classical	B-KEY
symbol	O
systems	O
,	O
with	O
connectionism	O
as	O
its	O
implementation	O
.	O
Here	O
it	O
is	O
argued	O
that	O
universality	B-KEY
is	O
also	O
a	O
problem	O
for	O
classicism	B-KEY
in	O
that	O
the	O
syntax-sensitive	B-KEY
rules	I-KEY
that	O
are	O
supposed	O
to	O
provide	O
causal	B-KEY
explanations	I-KEY
of	O
mental	B-KEY
processes	I-KEY
are	O
either	O
too	O
strict	O
,	O
precluding	O
possible	O
generalizations	O
;	O
or	O
too	O
lax	O
,	O
providing	O
no	O
information	O
as	O
to	O
the	O
appropriate	O
alternative	O
.	O
Consequently	O
,	O
universality	B-KEY
is	O
not	O
explained	O
by	O
a	O
classical	B-KEY
theory	O

Networking	O
in	O
the	O
palm	O
of	O
your	O
hand	O
-LSB-	O
PDA	B-KEY
buyer	O
's	O
guide	O
-RSB-	O
As	O
PDAs	B-KEY
move	O
beyond	O
the	O
personal	O
space	O
and	O
into	O
the	O
enterprise	O
,	O
you	O
need	O
to	O
get	O
a	O
firm	O
grip	O
on	O
the	O
options	O
available	O
for	O
your	O
users	O
.	O
What	O
operating	B-KEY
system	I-KEY
do	O
you	O
choose	O
?	O
What	O
features	O
do	O
you	O
and	O
your	O
company	O
need	O
?	O
How	O
will	O
these	O
devices	O
fit	O
into	O
the	O
existing	O
corporate	B-KEY
infrastructure	I-KEY
?	O
What	O
about	O
developer	B-KEY
support	I-KEY
?	O

Resonant	O
controllers	O
for	O
smart	B-KEY
structures	I-KEY
In	O
this	O
paper	O
we	O
propose	O
a	O
special	O
type	O
of	O
colocated	O
feedback	B-KEY
controller	I-KEY
for	O
smart	B-KEY
structures	I-KEY
.	O
The	O
controller	O
is	O
a	O
parallel	O
combination	O
of	O
high-Q	B-KEY
resonant	I-KEY
circuits	I-KEY
.	O
Each	O
of	O
the	O
resonant	O
circuits	O
is	O
tuned	O
to	O
a	O
pole	O
-LRB-	O
or	O
the	O
resonant	B-KEY
frequency	I-KEY
-RRB-	O
of	O
the	O
smart	B-KEY
structure	I-KEY
.	O
It	O
is	O
proven	O
that	O
the	O
parallel	O
combination	O
of	O
resonant	O
controllers	O
is	O
stable	O
with	O
an	O
infinite	O
gain	O
margin	O
.	O
Only	O
one	O
set	O
of	O
actuator-sensor	B-KEY
can	O
damp	B-KEY
multiple	B-KEY
resonant	I-KEY
modes	I-KEY
with	O
the	O
resonant	O
controllers	O
.	O
Experimental	O
results	O
are	O
presented	O
to	O
show	O
the	O
robustness	O
of	O
the	O
proposed	O
controller	O
in	O
damping	B-KEY
multimode	B-KEY
resonances	I-KEY

A	O
knowledge-based	B-KEY
approach	I-KEY
for	O
managing	O
urban	O
infrastructures	O
This	O
paper	O
presents	O
a	O
knowledge	O
e-based	O
approach	O
dedicated	O
to	O
the	O
efficient	O
management	O
,	O
regulation	B-KEY
,	O
interactive	O
and	O
dynamic	B-KEY
monitoring	I-KEY
of	O
urban	O
infrastructures	O
.	O
This	O
approach	O
identifies	O
the	O
data	O
and	O
related	O
treatments	O
common	O
to	O
several	O
municipal	B-KEY
activities	I-KEY
and	O
defines	O
the	O
requirements	O
and	O
functionalities	O
of	O
the	O
computer	O
tools	O
developed	O
to	O
improve	O
the	O
delivery	O
and	O
coordination	O
of	O
municipal	O
services	O
to	O
the	O
population	O
.	O
The	O
resulting	O
cooperative	B-KEY
system	I-KEY
called	O
SIGIU	B-KEY
is	O
composed	O
of	O
a	O
set	O
of	O
integrated	B-KEY
operating	I-KEY
systems	I-KEY
-LRB-	O
SYDEX	B-KEY
-RRB-	O
and	O
the	O
global	O
planning	O
and	O
coordination	B-KEY
system	I-KEY
-LRB-	O
SYGEC	B-KEY
-RRB-	O
.	O
The	O
objective	O
is	O
to	O
integrate	O
the	O
set	O
of	O
SYDEX	B-KEY
and	O
the	O
SYGEC	B-KEY
into	O
a	O
single	O
coherent	O
system	O
for	O
all	O
the	O
SIGIU	B-KEY
's	O
users	O
according	O
to	O
their	O
tasks	O
,	O
their	O
roles	O
,	O
and	O
their	O
responsibilities	O
within	O
the	O
municipal	O
administration	O
.	O
SIGIU	B-KEY
is	O
provided	O
by	O
different	O
measurement	O
and	O
monitoring	O
instruments	O
installed	O
on	O
some	O
system	O
's	O
elements	O
to	O
be	O
supervised	O
.	O
In	O
this	O
context	O
,	O
the	O
information	O
can	O
be	O
presented	O
in	O
different	O
forms	O
:	O
video	B-KEY
,	O
pictures	O
,	O
data	O
and	O
alarms	O
.	O
One	O
of	O
SIGIU	B-KEY
's	O
objectives	O
is	O
the	O
real-time	B-KEY
management	I-KEY
of	O
urban	O
infrastructures	O
'	O
control	O
mechanisms	O
.	O
To	O
carry	O
out	O
this	O
process	O
,	O
the	O
alarm	B-KEY
control	I-KEY
agent	I-KEY
creates	O
a	O
mobile	B-KEY
agent	I-KEY
associated	O
with	O
the	O
alarm	O
,	O
which	O
is	O
sent	O
to	O
a	O
mobile	O
station	O
and	O
warns	O
an	O
operator	O
.	O
Preliminary	O
implementation	O
results	O
show	O
that	O
SIGIU	B-KEY
supports	O
effectively	O
and	O
efficiently	O
the	O
decision	O
making	O
process	O
related	O
to	O
managing	O
urban	O
infrastructures	O

What	O
's	O
in	O
a	O
name	O
?	O
-LSB-	O
mobile	B-KEY
telephony	I-KEY
branding	B-KEY
-RSB-	O
Mobile	O
operators	O
are	O
frantically	O
consolidating	B-KEY
businesses	I-KEY
into	O
single	O
international	O
brands	B-KEY

Elastically	B-KEY
adaptive	I-KEY
deformable	I-KEY
models	I-KEY
We	O
present	O
a	O
technique	O
for	O
the	O
automatic	B-KEY
adaptation	I-KEY
of	O
a	O
deformable	O
model	O
's	O
elastic	B-KEY
parameters	I-KEY
within	O
a	O
Kalman	B-KEY
filter	I-KEY
framework	I-KEY
for	O
shape	B-KEY
estimation	I-KEY
applications	O
.	O
The	O
novelty	O
of	O
the	O
technique	O
is	O
that	O
the	O
model	O
's	O
elastic	B-KEY
parameters	I-KEY
are	O
not	O
constant	O
,	O
but	O
spatio-temporally	O
varying	O
.	O
The	O
variation	O
of	O
the	O
elastic	B-KEY
parameters	I-KEY
depends	O
on	O
the	O
distance	O
of	O
the	O
model	O
from	O
the	O
data	O
and	O
the	O
rate	O
of	O
change	O
of	O
this	O
distance	O
.	O
Each	O
pass	O
of	O
the	O
algorithm	O
uses	O
physics-based	B-KEY
modeling	I-KEY
techniques	I-KEY
to	O
iteratively	O
adjust	O
both	O
the	O
geometric	O
and	O
the	O
elastic	B-KEY
degrees	I-KEY
of	I-KEY
freedom	I-KEY
of	O
the	O
model	O
in	O
response	O
to	O
forces	O
that	O
are	O
computed	O
from	O
the	O
discrepancy	O
between	O
the	O
model	O
and	O
the	O
data	O
.	O
By	O
augmenting	O
the	O
state	B-KEY
equations	I-KEY
of	O
an	O
extended	B-KEY
Kalman	I-KEY
filter	I-KEY
to	O
incorporate	O
these	O
additional	O
variables	O
,	O
we	O
are	O
able	O
to	O
significantly	O
improve	O
the	O
quality	O
of	O
the	O
shape	B-KEY
estimation	I-KEY
.	O
Therefore	O
,	O
the	O
model	O
's	O
elastic	B-KEY
parameters	I-KEY
are	O
always	O
initialized	O
to	O
the	O
same	O
value	O
and	O
they	O
are	O
subsequently	O
modified	O
depending	O
on	O
the	O
data	O
and	O
the	O
noise	O
distribution	O
.	O
We	O
present	O
results	O
demonstrating	O
the	O
effectiveness	O
of	O
our	O
method	O
for	O
both	O
two-dimensional	O
and	O
three-dimensional	O
data	O

A	O
novel	O
control	B-KEY
logic	I-KEY
for	O
fast	O
valving	O
operations	O
This	O
letter	O
proposes	O
new	O
control	B-KEY
logic	I-KEY
for	O
operating	O
parallel	O
valves	O
in	O
fast	O
valving	O
schemes	O
in	O
order	O
to	O
improve	O
the	O
transient	B-KEY
stability	I-KEY
performance	O
of	O
power	O
systems	O
.	O
A	O
fast	O
valving	O
scheme	O
using	O
parallel	O
valves	O
overcomes	O
many	O
of	O
the	O
limitations	O
of	O
the	O
conventional	O
scheme	O
.	O
The	O
proposed	O
control	B-KEY
logic	I-KEY
for	O
operating	O
these	O
valves	O
has	O
been	O
applied	O
to	O
a	O
typical	O
single	B-KEY
machine	I-KEY
infinite	I-KEY
bus	I-KEY
system	I-KEY
.	O
Single	O
as	O
well	O
as	O
multiple	B-KEY
stroke	I-KEY
operations	I-KEY
for	O
controlling	O
the	O
turbine	O
power	O
output	O
have	O
been	O
studied	O
with	O
the	O
new	O
control	O
sequences	O
.	O
Encouraging	O
results	O
have	O
been	O
shown	O
over	O
the	O
conventional	O
schemes	O
of	O
fast	O
valving	O

Efficient	O
allocation	O
of	O
knowledge	O
in	O
distributed	B-KEY
business	I-KEY
structures	I-KEY
Accelerated	B-KEY
business	I-KEY
processes	I-KEY
demand	O
new	O
concepts	O
and	O
realizations	O
of	O
information	B-KEY
systems	I-KEY
and	O
knowledge	B-KEY
databases	I-KEY
.	O
This	O
paper	O
presents	O
the	O
concept	O
of	O
the	O
collaborative	B-KEY
information	I-KEY
space	I-KEY
-LRB-	O
CIS	O
-RRB-	O
,	O
which	O
supplies	O
the	O
necessary	O
tools	O
to	O
transform	O
individual	O
knowledge	O
into	O
collective	O
useful	O
information	O
.	O
The	O
creation	O
of	O
`	O
information	B-KEY
objects	I-KEY
'	O
in	O
the	O
CIS	O
allows	O
an	O
efficient	O
allocation	O
of	O
information	O
in	O
all	O
business	B-KEY
process	I-KEY
steps	I-KEY
at	O
any	O
time	O
.	O
Furthermore	O
,	O
the	O
specific	O
availability	O
of	O
heterogeneous	O
,	O
distributed	O
data	O
is	O
realized	O
by	O
a	O
Web-based	B-KEY
user	I-KEY
interface	I-KEY
,	O
which	O
enables	O
effective	O
search	O
by	O
a	O
multidimensionally	B-KEY
hierarchical	I-KEY
composition	I-KEY

Implementation	O
and	O
evaluation	O
of	O
HPF/SX	B-KEY
V2	I-KEY
We	O
are	O
developing	O
HPF/SX	B-KEY
V2	I-KEY
,	O
a	O
High	O
Performance	O
Fortran	O
-LRB-	O
HPF	O
-RRB-	O
compiler	B-KEY
for	O
vector	O
parallel	B-KEY
machines	O
.	O
It	O
provides	O
some	O
unique	O
extensions	O
as	O
well	O
as	O
the	O
features	O
of	O
HPF	O
2.0	O
and	O
HPF/JA	O
.	O
In	O
particular	O
,	O
this	O
paper	O
describes	O
four	O
of	O
them	O
:	O
-LRB-	O
1	O
-RRB-	O
the	O
ON	O
directive	O
of	O
HPF	O
2.0	O
;	O
-LRB-	O
2	O
-RRB-	O
the	O
REFLECT	O
and	O
LOCAL	O
directives	O
of	O
HPF/JA	O
;	O
-LRB-	O
3	O
-RRB-	O
vectorization	O
directives	O
;	O
and	O
-LRB-	O
4	O
-RRB-	O
automatic	O
parallelization	B-KEY
.	O
We	O
evaluate	O
these	O
features	O
through	O
some	O
benchmark	B-KEY
programs	O
on	O
NEC	O
SX-5	O
.	O
The	O
results	O
show	O
that	O
each	O
of	O
them	O
achieved	O
a	O
5-8	O
times	O
speedup	O
in	O
8-CPU	O
parallel	B-KEY
execution	O
and	O
the	O
four	O
features	O
are	O
useful	O
for	O
vector	O
parallel	B-KEY
execution	O
.	O
We	O
also	O
evaluate	O
the	O
overall	O
performance	O
of	O
HPF/SX	B-KEY
V2	I-KEY
by	O
using	O
over	O
30	O
well-known	O
benchmark	B-KEY
programs	O
from	O
HPFBench	O
,	O
APR	O
Benchmarks	B-KEY
,	O
GENESIS	O
Benchmarks	B-KEY
,	O
and	O
NAS	O
Parallel	B-KEY
Benchmarks	B-KEY
.	O
About	O
half	O
of	O
the	O
programs	O
showed	O
good	O
performance	O
,	O
while	O
the	O
other	O
half	O
suggest	O
weakness	O
of	O
the	O
compiler	B-KEY
,	O
especially	O
on	O
its	O
runtimes	O
.	O
It	O
is	O
necessary	O
to	O
improve	O
them	O
to	O
put	O
the	O
compiler	B-KEY
to	O
practical	O
use	O

Identifying	O
multivariate	B-KEY
discordant	I-KEY
observations	I-KEY
:	O
a	O
computer-intensive	B-KEY
approach	I-KEY
The	O
problem	O
of	O
identifying	O
multiple	B-KEY
outliers	I-KEY
in	O
a	O
multivariate	B-KEY
normal	I-KEY
sample	I-KEY
is	O
approached	O
via	O
successive	O
testing	O
using	O
P-values	B-KEY
rather	O
than	O
tabled	B-KEY
critical	I-KEY
values	I-KEY
.	O
Caroni	O
and	O
Prescott	O
-LRB-	O
Appl	O
.	O
Statist	O
.	O
41	O
,	O
p.	O
355	O
,	O
1992	O
-RRB-	O
proposed	O
a	O
generalization	O
of	O
the	O
EDR-ESD	O
procedure	O
of	O
Rosner	O
-LRB-	O
Technometrics	O
,	O
25	O
,	O
1983	O
-RRB-	O
-RRB-	O
.	O
Venter	O
and	O
Viljoen	O
-LRB-	O
Comput	O
.	O
Statist	O
.	O
Data	O
Anal	O
.	O
29	O
,	O
p.	O
261	O
,	O
1999	O
-RRB-	O
introduced	O
a	O
computer	O
intensive	O
method	O
to	O
identify	O
outliers	O
in	O
a	O
univariate	B-KEY
outlier	I-KEY
situation	O
.	O
We	O
now	O
generalize	O
this	O
method	O
to	O
the	O
multivariate	B-KEY
outlier	I-KEY
situation	O
and	O
compare	O
this	O
new	O
procedure	O
with	O
that	O
of	O
Caroni	O
and	O
Prescott	O
-LRB-	O
Appl	O
.	O
Statist	O
.	O
4	O
,	O
p.	O
355	O
,	O
1992	O
-RRB-	O

Learning	O
nonregular	O
languages	O
:	O
a	O
comparison	O
of	O
simple	O
recurrent	O
networks	O
and	O
LSTM	B-KEY
Rodriguez	O
-LRB-	O
2001	O
-RRB-	O
examined	O
the	O
learning	O
ability	O
of	O
simple	O
recurrent	O
nets	O
-LRB-	O
SRNs	O
-RRB-	O
-LRB-	O
Elman	O
,	O
1990	O
-RRB-	O
on	O
simple	O
context-sensitive	O
and	O
context-free	B-KEY
languages	I-KEY
.	O
In	O
response	O
to	O
Rodriguez	O
's	O
-LRB-	O
2001	O
-RRB-	O
article	O
,	O
we	O
compare	O
the	O
performance	B-KEY
of	O
simple	O
recurrent	O
nets	O
and	O
long	O
short-term	B-KEY
memory	I-KEY
recurrent	I-KEY
nets	I-KEY
on	O
context-free	O
and	O
context-sensitive	B-KEY
languages	I-KEY

An	O
improved	O
fuzzy	B-KEY
MCDM	I-KEY
model	I-KEY
based	O
on	O
ideal	O
and	O
anti-ideal	B-KEY
concepts	I-KEY
Liang	O
presented	O
-LRB-	O
1999	O
-RRB-	O
a	O
fuzzy	O
multiple	O
criteria	O
decision	O
making	O
-LRB-	O
MCDM	O
-RRB-	O
method	O
based	O
on	O
the	O
concepts	O
of	O
ideal	O
and	O
anti-ideal	O
points	O
.	O
Despite	O
its	O
merits	O
,	O
Liang	O
method	O
has	O
the	O
following	O
limitations	O
:	O
-LRB-	O
i	O
-RRB-	O
the	O
objective	O
criteria	O
are	O
converted	O
into	O
dimensionless	B-KEY
indices	I-KEY
and	O
the	O
subjective	O
criteria	O
are	O
not	O
converted	O
,	O
which	O
may	O
prevent	O
compatibility	O
for	O
these	O
criteria	O
,	O
-LRB-	O
ii	O
-RRB-	O
the	O
formulas	O
for	O
converting	O
objective	O
criteria	O
are	O
not	O
reliable	O
,	O
and	O
-LRB-	O
iii	O
-RRB-	O
an	O
unreliable	O
ranking	O
method	O
,	O
i.e.	O
maximizing	O
set	O
and	O
minimizing	O
set	O
,	O
is	O
applied	O
to	O
rank	O
the	O
fuzzy	O
numbers	O
.	O
This	O
paper	O
applies	O
the	O
Hsu	O
and	O
Chen	O
method	O
and	O
suggests	O
a	O
fuzzy	B-KEY
number	I-KEY
ranking	I-KEY
method	O
to	O
propose	O
an	O
improved	O
fuzzy	B-KEY
MCDM	I-KEY
model	I-KEY
based	O
on	O
ideal	O
and	O
anti-ideal	B-KEY
concepts	I-KEY
to	O
overcome	O
the	O
shortcomings	O
of	O
the	O
Liang	O
method	O
.	O
Numerical	O
examples	O
demonstrate	O
the	O
effectiveness	O
and	O
feasibility	O
of	O
the	O
proposed	O
ranking	O
method	O
and	O
the	O
improved	O
model	O
,	O
respectively	O

Implementing	B-KEY
:	O
it	O
's	O
all	O
about	O
processes	O
Looks	O
at	O
how	O
the	O
key	O
to	O
successful	O
technology	B-KEY
deployment	I-KEY
can	O
be	O
found	O
in	O
a	O
set	O
of	O
four	O
basic	O
disciplines	O

An	O
efficient	O
algorithm	O
for	O
sequential	O
generation	O
of	O
failure	O
states	O
in	O
a	O
network	O
with	O
multi-mode	O
components	O
In	O
this	O
work	O
,	O
a	O
new	O
algorithm	O
for	O
the	O
sequential	O
generation	O
of	O
failure	O
states	O
in	O
a	O
network	O
with	O
multi-mode	O
components	O
is	O
proposed	O
.	O
The	O
algorithm	O
presented	O
in	O
the	O
paper	O
transforms	O
the	O
state	B-KEY
enumeration	I-KEY
problem	I-KEY
into	O
a	O
K-shortest	B-KEY
paths	I-KEY
problem	I-KEY
.	O
Taking	O
advantage	O
of	O
the	O
inherent	O
efficiency	O
of	O
an	O
algorithm	O
for	O
shortest	O
paths	O
enumeration	O
and	O
also	O
of	O
the	O
characteristics	O
of	O
the	O
reliability	O
problem	O
in	O
which	O
it	O
will	O
be	O
used	O
,	O
an	O
algorithm	O
with	O
lower	O
complexity	O
than	O
the	O
best	O
algorithm	O
in	O
the	O
literature	O
for	O
solving	O
this	O
problem	O
,	O
was	O
obtained	O
.	O
Computational	O
results	O
will	O
be	O
presented	O
for	O
comparing	O
the	O
efficiency	O
of	O
both	O
algorithms	O
in	O
terms	O
of	O
CPU	B-KEY
time	I-KEY
and	O
for	O
problems	O
of	O
different	O
size	O

Using	O
k-nearest-neighbor	B-KEY
classification	I-KEY
in	O
the	O
leaves	O
of	O
a	O
tree	O
We	O
construct	O
a	O
hybrid	O
-LRB-	O
composite	O
-RRB-	O
classifier	O
by	O
combining	O
two	O
classifiers	O
in	O
common	O
use	O
-	O
classification	B-KEY
trees	I-KEY
and	O
k-nearest-neighbor	O
-LRB-	O
k-NN	O
-RRB-	O
.	O
In	O
our	O
scheme	O
we	O
divide	O
the	O
feature	O
space	O
up	O
by	O
a	O
classification	B-KEY
tree	I-KEY
,	O
and	O
then	O
classify	O
test	O
set	O
items	O
using	O
the	O
k-NN	B-KEY
rule	I-KEY
just	O
among	O
those	O
training	O
items	O
in	O
the	O
same	O
leaf	O
as	O
the	O
test	O
item	O
.	O
This	O
reduces	O
somewhat	O
the	O
computational	B-KEY
load	I-KEY
associated	O
with	O
k-NN	O
,	O
and	O
it	O
produces	O
a	O
classification	O
rule	O
that	O
performs	O
better	O
than	O
either	O
trees	O
or	O
the	O
usual	O
k-NN	O
in	O
a	O
number	O
of	O
well-known	O
data	B-KEY
sets	I-KEY

High-speed	B-KEY
consistency	I-KEY
checking	I-KEY
for	O
hypothetical	B-KEY
reasoning	I-KEY
systems	O
using	O
inference	B-KEY
path	I-KEY
network	I-KEY
Hypothetical	B-KEY
reasoning	I-KEY
is	O
popular	O
in	O
fault	B-KEY
diagnostics	I-KEY
and	O
design	O
systems	O
,	O
but	O
slow	O
reasoning	B-KEY
speed	I-KEY
is	O
its	O
drawback	O
.	O
The	O
goal	O
of	O
the	O
current	O
study	O
is	O
developing	O
hypothetical	B-KEY
reasoning	I-KEY
based	O
on	O
an	O
inference	B-KEY
path	I-KEY
network	I-KEY
,	O
which	O
would	O
overcome	O
this	O
drawback	O
.	O
In	O
hypothetical	B-KEY
reasoning	I-KEY
systems	O
based	O
on	O
an	O
inference	B-KEY
path	I-KEY
network	I-KEY
,	O
there	O
is	O
much	O
room	O
for	O
improvement	O
regarding	O
the	O
computing	O
costs	O
of	O
connotation	O
processing	O
and	O
consistency	O
checking	O
.	O
The	O
authors	O
of	O
this	O
study	O
demonstrate	O
improvement	O
ideas	O
regarding	O
one	O
of	O
these	O
problems	O
,	O
namely	O
,	O
consistency	O
checking	O
.	O
First	O
,	O
the	O
authors	O
obtained	O
necessary	O
and	O
sufficient	O
conditions	O
under	O
which	O
inconsistencies	B-KEY
occur	O
during	O
hypothesis	B-KEY
composition	I-KEY
.	O
Based	O
on	O
the	O
obtained	O
results	O
,	O
the	O
authors	O
proposed	O
an	O
algorithm	O
for	O
speeding	B-KEY
up	I-KEY
the	O
process	O
of	O
consistency	O
checking	O
.	O
Processing	O
with	O
this	O
algorithm	O
in	O
its	O
core	O
consists	O
of	O
transforming	O
the	O
inference	B-KEY
path	I-KEY
network	I-KEY
in	O
such	O
a	O
way	O
that	O
inconsistencies	B-KEY
do	O
not	O
occur	O
during	O
the	O
hypothesis	B-KEY
composition	I-KEY
,	O
under	O
the	O
condition	O
of	O
unchanged	O
solution	O
hypotheses	O
.	O
The	O
efficiency	O
of	O
this	O
algorithm	O
was	O
confirmed	O
by	O
tests	O

Electronic	B-KEY
reserves	I-KEY
at	O
University	B-KEY
College	I-KEY
London	I-KEY
:	O
understanding	O
the	O
needs	O
of	O
academic	O
departments	O
This	O
article	O
describes	O
a	O
recent	O
project	O
at	O
University	B-KEY
College	I-KEY
London	I-KEY
to	O
explore	O
the	O
feasibility	O
of	O
providing	O
a	O
service	O
to	O
improve	O
access	O
to	O
electronic	B-KEY
course	I-KEY
materials	I-KEY
.	O
Funded	O
by	O
the	O
Higher	B-KEY
Education	I-KEY
Funding	I-KEY
Council	I-KEY
for	I-KEY
England	I-KEY
-LRB-	O
HEFCE	O
-RRB-	O
,	O
the	O
project	O
was	O
not	O
simply	O
to	O
set	O
up	O
an	O
electronic	B-KEY
reserve	I-KEY
.	O
By	O
undertaking	O
a	O
needs	O
analysis	O
of	O
academic	O
departments	O
,	O
the	O
project	O
was	O
able	O
to	O
tailor	O
the	O
design	O
of	O
the	O
new	O
service	O
appropriately	O
.	O
While	O
new	O
initiatives	O
in	O
libraries	O
are	O
often	O
established	O
using	O
project	O
funding	O
,	O
this	O
work	O
was	O
unique	O
in	O
being	O
research-led	O
.	O
It	O
also	O
involved	O
collaboration	O
between	O
library	O
and	O
computing	B-KEY
staff	I-KEY
and	O
learning	B-KEY
technologists	I-KEY

Information	B-KEY
access	I-KEY
for	O
all	O
:	O
meeting	O
the	O
needs	O
of	O
deaf	B-KEY
and	O
hard	O
of	O
hearing	O
people	O
Discusses	O
the	O
nature	O
of	O
deafness	B-KEY
and	O
hearing	B-KEY
impairments	I-KEY
,	O
with	O
particular	O
reference	O
to	O
the	O
impact	O
which	O
the	O
onset	O
of	O
hearing	O
loss	O
presents	O
at	O
various	O
ages	O
.	O
The	O
author	O
goes	O
on	O
to	O
present	O
practical	O
tips	O
for	O
interacting	O
with	O
deaf	B-KEY
and	O
hard	B-KEY
of	I-KEY
hearing	I-KEY
clients	I-KEY
in	O
various	O
communication	B-KEY
contexts	I-KEY
,	O
including	O
sightreading	B-KEY
,	O
TTY	B-KEY
communications	I-KEY
,	O
and	O
ASL	B-KEY
interpreters	I-KEY
.	O
An	O
annotated	O
list	O
of	O
suggested	O
readings	O
is	O
appended	O

A	O
collocation	B-KEY
formulation	I-KEY
of	O
multistep	B-KEY
methods	I-KEY
for	O
variable	B-KEY
step-size	I-KEY
extensions	I-KEY
Multistep	B-KEY
methods	I-KEY
are	O
classically	O
constructed	O
by	O
specially	O
designed	O
difference	B-KEY
operators	I-KEY
on	O
an	O
equidistant	B-KEY
time	I-KEY
grid	I-KEY
.	O
To	O
make	O
them	O
practically	O
useful	O
,	O
they	O
have	O
to	O
be	O
implemented	O
by	O
varying	O
the	O
step-size	O
according	O
to	O
some	O
error-control	B-KEY
algorithm	I-KEY
.	O
It	O
is	O
well	O
known	O
how	O
to	O
extend	O
Adams	O
and	O
BDF	O
formulas	O
to	O
a	O
variable	O
step-size	O
formulation	O
.	O
In	O
this	O
paper	O
we	O
present	O
a	O
collocation	O
approach	O
to	O
construct	O
variable	O
step-size	O
formulas	O
.	O
We	O
make	O
use	O
of	O
piecewise	B-KEY
polynomials	I-KEY
to	O
show	O
that	O
every	O
k-step	B-KEY
method	I-KEY
of	O
order	O
k	O
+1	O
has	O
a	O
variable	O
step-size	O
polynomial	O
collocation	B-KEY
formulation	I-KEY

Buying	O
into	O
the	O
relationship	O
-LSB-	O
business	B-KEY
software	I-KEY
-RSB-	O
Choosing	O
the	O
right	O
software	O
to	O
improve	O
business	O
processes	O
can	O
have	O
a	O
huge	O
impact	O
on	O
a	O
company	O
's	O
efficiency	O
and	O
profitability	O
.	O
While	O
it	O
is	O
sometimes	O
hard	O
to	O
get	O
beyond	O
vendor	O
hype	O
about	O
software	O
features	O
and	O
functionality	B-KEY
and	O
know	O
what	O
to	O
realistically	O
expect	O
,	O
it	O
is	O
even	O
more	O
difficult	O
to	O
determine	O
if	O
the	O
vendor	O
is	O
the	O
right	O
vendor	O
to	O
partner	O
with	O
.	O
Thus	O
picking	O
the	O
right	O
software	O
is	O
important	O
,	O
but	O
companies	O
have	O
to	O
realize	O
that	O
what	O
they	O
are	O
really	O
buying	O
into	O
is	O
a	O
relationship	O
with	O
the	O
vendor	O

Ten	O
years	O
of	O
strategies	O
to	O
increase	O
participation	O
of	O
women	B-KEY
in	O
computing	B-KEY
programs	I-KEY
.	O
The	O
Central	B-KEY
Queensland	I-KEY
University	I-KEY
experience	O
:	O
1999-2001	O
In	O
the	O
late	O
eighties	O
,	O
the	O
participation	O
rate	O
of	O
women	B-KEY
in	O
information	O
technology	O
courses	O
in	O
most	O
Australian	O
Universities	O
was	O
around	O
25	O
%	O
.	O
This	O
low	O
level	O
of	O
women	B-KEY
's	O
participation	O
in	O
computing	O
courses	O
occurs	O
not	O
only	O
in	O
Australia	O
but	O
also	O
overseas	O
.	O
More	O
studies	O
indicate	O
that	O
the	O
participation	O
rates	O
have	O
not	O
improved	O
and	O
in	O
fact	O
may	O
be	O
even	O
further	O
in	O
decline	O
.	O
Participation	O
rates	O
in	O
the	O
workforce	O
also	O
appear	O
to	O
be	O
in	O
decline	O
.	O
Concerned	O
at	O
the	O
imbalance	O
within	O
Australia	O
,	O
the	O
Federal	O
government	O
directed	O
all	O
Australian	O
Universities	O
to	O
increase	O
the	O
number	O
of	O
women	B-KEY
in	O
courses	O
leading	O
to	O
a	O
professional	O
computing	O
qualification	O
-LRB-	O
i.e.	O
,	O
information	O
technology	O
courses	O
-RRB-	O
to	O
40	O
%	O
of	O
students	O
by	O
1995	O
.	O
This	O
paper	O
details	O
one	O
Australian	O
university	O
's	O
approach	O
,	O
over	O
a	O
10	O
year	O
period	O
-LRB-	O
1991-2001	O
-RRB-	O
,	O
to	O
redress	O
this	O
imbalance	O
.	O
We	O
provide	O
examples	O
of	O
intervention	O
strategies	O
developed	O
and	O
the	O
outcomes	O
for	O
these	O
strategies	O
.	O
We	O
present	O
the	O
outcomes	O
against	O
a	O
background	O
frame	O
of	O
the	O
Australian	B-KEY
Higher	I-KEY
Education	I-KEY
scene	O
of	O
that	O
decade	O
which	O
was	O
influenced	O
by	O
funding	O
levels	O
to	O
universities	O
in	O
general	O
and	O
to	O
equity	O
programs	O
in	O
particular	O
.	O
We	O
present	O
data	O
related	O
to	O
the	O
participation	O
of	O
women	B-KEY
in	O
computing	B-KEY
programs	I-KEY
along	O
with	O
snapshots	O
of	O
the	O
overall	O
changing	O
student	O
demographics	B-KEY
over	O
this	O
period	O

National	B-KEY
learning	I-KEY
systems	I-KEY
:	O
a	O
new	O
approach	O
on	O
technological	B-KEY
change	I-KEY
in	O
late	B-KEY
industrializing	I-KEY
economies	I-KEY
and	O
evidences	O
from	O
the	O
cases	O
of	O
Brazil	B-KEY
and	O
South	B-KEY
Korea	I-KEY
The	O
paper	O
has	O
two	O
intertwined	O
parts	O
.	O
The	O
first	O
one	O
is	O
a	O
proposal	O
for	O
a	O
conceptual	O
and	O
theoretical	O
framework	O
to	O
understand	O
technical	O
change	O
in	O
late	B-KEY
industrializing	I-KEY
economies	I-KEY
.	O
The	O
second	O
part	O
develops	O
a	O
kind	O
of	O
empirical	O
test	O
of	O
the	O
usefulness	O
of	O
that	O
new	O
framework	O
by	O
means	O
of	O
a	O
comparative	O
study	O
of	O
the	O
Brazilian	O
and	O
South	O
Korean	O
cases	O
.	O
All	O
the	O
four	O
types	O
of	O
macroevidences	O
of	O
the	O
technical	O
change	O
processes	O
of	O
Brazil	B-KEY
and	O
Korea	O
corroborate	O
,	O
directly	O
or	O
indirectly	O
,	O
the	O
hypothesis	O
of	O
the	O
existence	O
of	O
actual	O
cases	O
of	O
national	B-KEY
learning	I-KEY
systems	I-KEY
-LRB-	O
NLSs	O
-RRB-	O
of	O
passive	O
and	O
active	O
nature	O
,	O
as	O
it	O
is	O
shown	O
to	O
be	O
the	O
cases	O
of	O
Brazil	B-KEY
and	O
South	B-KEY
Korea	I-KEY
,	O
respectively	O
.	O
The	O
contrast	O
between	O
the	O
two	O
processes	O
of	O
technical	O
change	O
prove	O
remarkable	O
,	O
despite	O
both	O
processes	O
being	O
essentially	O
confined	O
to	O
learning	O
.	O
The	O
concepts	O
of	O
passive	O
and	O
active	O
NLSs	O
show	O
how	O
useful	O
they	O
are	O
to	O
apprehend	O
the	O
diversity	O
of	O
those	O
realities	O
,	O
and	O
,	O
consequently	O
,	O
to	O
avoid	O
,	O
for	O
instance	O
,	O
interpretations	O
that	O
misleadingly	O
suppose	O
-LRB-	O
based	O
on	O
conventional	O
economic	O
theory	O
-RRB-	O
that	O
those	O
countries	O
have	O
a	O
similar	O
lack	O
of	O
technological	O
dynamism	O

Moving	O
into	O
the	O
mainstream	O
-LSB-	O
product	B-KEY
lifecycle	I-KEY
management	I-KEY
-RSB-	O
Product	B-KEY
lifecycle	I-KEY
management	I-KEY
-LRB-	O
PLM	O
-RRB-	O
is	O
widely	O
recognised	O
by	O
most	O
manufacturing	B-KEY
companies	I-KEY
,	O
as	O
manufacturers	O
begin	O
to	O
identify	O
and	O
implement	O
targeted	O
projects	O
intended	O
to	O
deliver	O
return-on	O
investment	O
in	O
a	O
timely	O
fashion	O
.	O
Vendors	O
are	O
also	O
releasing	O
second-generation	O
PLM	O
products	O
that	O
are	O
packaged	O
,	O
out-of-the-box	O
solutions	O

The	O
agile	O
revolution	O
-LSB-	O
business	B-KEY
agility	I-KEY
-RSB-	O
There	O
is	O
a	O
new	O
business	O
revolution	O
in	O
the	O
air	O
.	O
The	O
theory	O
is	O
there	O
,	O
the	O
technology	O
is	O
evolving	O
,	O
fast	O
.	O
It	O
is	O
all	O
about	O
agility	O

Restoration	O
of	O
archival	O
documents	O
using	O
a	O
wavelet	B-KEY
technique	I-KEY
This	O
paper	O
addresses	O
a	O
problem	O
of	O
restoring	O
handwritten	B-KEY
archival	I-KEY
documents	I-KEY
by	O
recovering	O
their	O
contents	O
from	O
the	O
interfering	O
handwriting	O
on	O
the	O
reverse	O
side	O
caused	O
by	O
the	O
seeping	O
of	O
ink	O
.	O
We	O
present	O
a	O
novel	O
method	O
that	O
works	O
by	O
first	O
matching	O
both	O
sides	O
of	O
a	O
document	O
such	O
that	O
the	O
interfering	O
strokes	O
are	O
mapped	O
with	O
the	O
corresponding	O
strokes	O
originating	O
from	O
the	O
reverse	O
side	O
.	O
This	O
facilitates	O
the	O
identification	O
of	O
the	O
foreground	O
and	O
interfering	O
strokes	O
.	O
A	O
wavelet	B-KEY
reconstruction	I-KEY
process	I-KEY
then	O
iteratively	O
enhances	O
the	O
foreground	O
strokes	O
and	O
smears	O
the	O
interfering	O
strokes	O
so	O
as	O
to	O
strengthen	O
the	O
discriminating	O
capability	O
of	O
an	O
improved	O
Canny	B-KEY
edge	I-KEY
detector	I-KEY
against	O
the	O
interfering	O
strokes	O
.	O
The	O
method	O
has	O
been	O
shown	O
to	O
restore	O
the	O
documents	O
effectively	O
with	O
average	O
precision	O
and	O
recall	O
rates	O
for	O
foreground	O
text	O
extraction	O
at	O
84	O
percent	O
and	O
96	O
percent	O
,	O
respectively	O

The	O
results	O
of	O
experimental	O
studies	O
of	O
the	O
reflooding	O
of	O
fuel-rod	O
assemblies	O
from	O
above	O
and	O
problems	O
for	O
future	O
investigations	O
Problems	O
in	O
studying	O
the	O
reflooding	O
of	O
assemblies	O
from	O
above	O
conducted	O
at	O
foreign	O
and	O
Russian	B-KEY
experimental	I-KEY
installations	I-KEY
are	O
considered	O
.	O
The	O
efficiency	O
of	O
cooling	O
and	O
flow	B-KEY
reversal	I-KEY
under	O
countercurrent	B-KEY
flow	I-KEY
of	O
steam	B-KEY
and	O
water	B-KEY
,	O
as	O
well	O
as	O
the	O
scale	O
effect	O
are	O
analyzed	O
.	O
The	O
tasks	O
for	O
future	O
experiments	O
that	O
are	O
necessary	O
for	O
the	O
development	O
of	O
modern	O
correlations	O
for	O
the	O
loss-of-coolant	O
accident	O
-LRB-	O
LOCA	O
-RRB-	O
computer	O
codes	O
are	O
stated	O

Numerical	B-KEY
approximation	I-KEY
of	O
nonlinear	O
BVPs	O
by	O
means	O
of	O
BVMs	B-KEY
Boundary	B-KEY
Value	I-KEY
Methods	I-KEY
-LRB-	O
BVMs	B-KEY
-RRB-	O
would	O
seem	O
to	O
be	O
suitable	O
candidates	O
for	O
the	O
solution	O
of	O
nonlinear	B-KEY
Boundary	I-KEY
Value	I-KEY
Problems	I-KEY
-LRB-	O
BVPs	O
-RRB-	O
.	O
They	O
have	O
been	O
successfully	O
used	O
for	O
solving	O
linear	O
BVPs	O
together	O
with	O
a	O
mesh	B-KEY
selection	I-KEY
strategy	I-KEY
based	O
on	O
the	O
conditioning	O
of	O
the	O
linear	O
systems	O
.	O
Our	O
aim	O
is	O
to	O
extend	O
this	O
approach	O
so	O
as	O
to	O
use	O
them	O
for	O
the	O
numerical	B-KEY
approximation	I-KEY
of	O
nonlinear	O
problems	O
.	O
For	O
this	O
reason	O
,	O
we	O
consider	O
the	O
quasi-linearization	B-KEY
technique	I-KEY
that	O
is	O
an	O
application	O
of	O
the	O
Newton	B-KEY
method	I-KEY
to	O
the	O
nonlinear	B-KEY
differential	I-KEY
equation	I-KEY
.	O
Consequently	O
,	O
each	O
iteration	O
requires	O
the	O
solution	O
of	O
a	O
linear	O
BVP	O
.	O
In	O
order	O
to	O
guarantee	O
the	O
convergence	O
to	O
the	O
solution	O
of	O
the	O
continuous	O
nonlinear	O
problem	O
,	O
it	O
is	O
necessary	O
to	O
determine	O
how	O
accurately	O
the	O
linear	O
BVPs	O
must	O
be	O
solved	O
.	O
For	O
this	O
goal	O
,	O
suitable	O
stopping	B-KEY
criteria	I-KEY
on	O
the	O
residual	O
and	O
on	O
the	O
error	O
for	O
each	O
linear	O
BVP	O
are	O
given	O
.	O
Numerical	O
experiments	O
on	O
stiff	B-KEY
problems	I-KEY
give	O
rather	O
satisfactory	O
results	O
,	O
showing	O
that	O
the	O
experimental	O
code	O
,	O
called	O
TOM	O
,	O
that	O
uses	O
a	O
class	O
of	O
BVMs	B-KEY
and	O
the	O
quasi-linearization	B-KEY
technique	I-KEY
,	O
may	O
be	O
competitive	O
with	O
well	O
known	O
solvers	O
for	O
BVPs	O

The	O
semi-algebraic	O
theory	O
of	O
stochastic	O
games	O
The	O
asymptotic	B-KEY
behavior	I-KEY
of	O
the	O
min-max	B-KEY
value	I-KEY
of	O
a	O
finite-state	B-KEY
zero-sum	I-KEY
discounted	I-KEY
stochastic	I-KEY
game	I-KEY
,	O
as	O
the	O
discount	B-KEY
rate	I-KEY
approaches	O
0	O
,	O
has	O
been	O
studied	O
in	O
the	O
past	O
using	O
the	O
theory	O
of	O
real-closed	O
fields	O
.	O
We	O
use	O
the	O
theory	O
of	O
semi-algebraic	O
sets	O
and	O
mappings	O
to	O
prove	O
some	O
asymptotic	O
properties	O
of	O
the	O
min-max	B-KEY
value	I-KEY
,	O
which	O
hold	O
uniformly	O
for	O
all	O
stochastic	O
games	O
in	O
which	O
the	O
number	O
of	O
states	O
and	O
players	O
'	O
actions	O
are	O
predetermined	O
to	O
some	O
fixed	O
values	O
.	O
As	O
a	O
corollary	O
,	O
we	O
prove	O
a	O
uniform	B-KEY
polynomial	I-KEY
convergence	I-KEY
rate	I-KEY
of	O
the	O
value	O
of	O
the	O
N-stage	B-KEY
game	I-KEY
to	O
the	O
value	O
of	O
the	O
nondiscount	O
game	O
,	O
over	O
a	O
bounded	O
set	O
of	O
payoffs	O

An	O
exactly	B-KEY
solvable	I-KEY
random	I-KEY
satisfiability	I-KEY
problem	I-KEY
We	O
introduce	O
a	O
new	O
model	O
for	O
the	O
generation	O
of	O
random	O
satisfiability	O
problems	O
.	O
It	O
is	O
an	O
extension	O
of	O
the	O
hyper-SAT	B-KEY
model	I-KEY
of	O
Ricci-Tersenghi	O
,	O
Weigt	O
and	O
Zecchina	O
-LRB-	O
2001	O
-RRB-	O
,	O
which	O
is	O
a	O
variant	O
of	O
the	O
famous	O
K-SAT	O
model	O
:	O
it	O
is	O
extended	O
to	O
q-state	B-KEY
variables	I-KEY
and	O
relates	O
to	O
a	O
different	O
choice	O
of	O
the	O
statistical	B-KEY
ensemble	I-KEY
.	O
The	O
model	O
has	O
an	O
exactly	O
solvable	O
statistic	O
:	O
the	O
critical	O
exponents	O
and	O
scaling	O
functions	O
of	O
the	O
SAT/UNSAT	O
transition	O
are	O
calculable	O
at	O
zero	O
temperature	O
,	O
with	O
no	O
need	O
of	O
replicas	O
,	O
also	O
with	O
exact	B-KEY
finite-size	I-KEY
corrections	I-KEY
.	O
We	O
also	O
introduce	O
an	O
exact	B-KEY
duality	I-KEY
of	O
the	O
model	O
,	O
and	O
show	O
an	O
analogy	O
of	O
thermodynamic	B-KEY
properties	I-KEY
with	O
the	O
random	B-KEY
energy	I-KEY
model	I-KEY
of	O
disordered	B-KEY
spin	I-KEY
system	I-KEY
theory	I-KEY
.	O
Relations	O
with	O
error	B-KEY
correcting	I-KEY
codes	I-KEY
are	O
also	O
discussed	O

A	O
fractional-flow	O
model	O
of	O
serial	B-KEY
manufacturing	I-KEY
systems	I-KEY
with	O
rework	B-KEY
and	O
its	O
reachability	B-KEY
and	O
controllability	B-KEY
properties	O
A	O
dynamic	B-KEY
fractional-flow	I-KEY
model	I-KEY
of	O
a	O
serial	B-KEY
manufacturing	I-KEY
system	I-KEY
incorporating	O
rework	B-KEY
is	O
considered	O
.	O
Using	O
some	O
results	O
on	O
reachability	B-KEY
and	O
controllability	B-KEY
of	O
positive	B-KEY
linear	I-KEY
systems	I-KEY
the	O
ability	O
of	O
serial	B-KEY
manufacturing	I-KEY
systems	I-KEY
with	O
rework	B-KEY
to	O
``	O
move	O
in	O
space	O
''	O
,	O
that	O
is	O
their	O
reachability	B-KEY
and	O
controllability	B-KEY
properties	O
,	O
are	O
studied	O
.	O
These	O
properties	O
are	O
important	O
not	O
only	O
for	O
optimising	O
the	O
performance	O
of	O
the	O
manufacturing	O
system	O
,	O
possibly	O
off-line	O
,	O
but	O
also	O
to	O
improve	O
its	O
functioning	O
by	O
using	O
feedback	B-KEY
control	I-KEY
online	O

A	O
new	O
high	O
resolution	O
color	O
flow	O
system	O
using	O
an	O
eigendecomposition-based	B-KEY
adaptive	I-KEY
filter	I-KEY
for	O
clutter	O
rejection	O
We	O
present	O
a	O
new	O
signal	B-KEY
processing	I-KEY
strategy	I-KEY
for	O
high	B-KEY
frequency	I-KEY
color	I-KEY
flow	I-KEY
mapping	I-KEY
in	O
moving	B-KEY
tissue	I-KEY
environments	I-KEY
.	O
A	O
new	O
application	O
of	O
an	O
eigendecomposition-based	O
clutter	B-KEY
rejection	I-KEY
filter	I-KEY
is	O
presented	O
with	O
modifications	O
to	O
deal	O
with	O
high	B-KEY
blood-to-clutter	I-KEY
ratios	I-KEY
-LRB-	O
BCR	O
-RRB-	O
.	O
Additionally	O
,	O
a	O
new	O
method	O
for	O
correcting	O
blood	O
velocity	O
estimates	O
with	O
an	O
estimated	B-KEY
tissue	I-KEY
motion	I-KEY
profile	I-KEY
is	O
detailed	O
.	O
The	O
performance	O
of	O
the	O
clutter	O
filter	O
and	O
velocity	O
estimation	O
strategies	O
is	O
quantified	O
using	O
a	O
new	O
swept-scan	B-KEY
signal	I-KEY
model	I-KEY
.	O
In	B-KEY
vivo	I-KEY
color	I-KEY
flow	I-KEY
images	I-KEY
are	O
presented	O
to	O
illustrate	O
the	O
potential	O
of	O
the	O
system	O
for	O
mapping	O
blood	O
flow	O
in	O
the	O
microcirculation	B-KEY
with	O
external	O
tissue	O
motion	O

How	O
much	O
should	O
publishers	B-KEY
spend	O
on	O
technology	O
?	O
A	O
study	O
confirms	O
that	O
spending	O
on	O
publishing-specific	O
information	O
technology	O
-LRB-	O
IT	O
-RRB-	O
resources	O
is	O
growing	O
much	O
faster	O
than	O
IT	B-KEY
spending	I-KEY
for	O
general	O
business	O
activities	O
,	O
at	O
least	O
among	O
leading	O
publishers	B-KEY
in	O
the	O
scientific	O
,	O
technical	O
and	O
medical	O
-LRB-	O
STM	O
-RRB-	O
market	O
.	O
The	O
survey	O
asked	O
about	O
information	O
technology	O
funding	O
and	O
staffing	O
levels-past	O
,	O
present	O
and	O
future-and	O
also	O
inquired	O
about	O
activities	O
in	O
content	B-KEY
management	I-KEY
,	O
Web	B-KEY
delivery	I-KEY
,	O
computer	B-KEY
support	I-KEY
and	O
customer	B-KEY
relationship	I-KEY
management	I-KEY
.	O
The	O
results	O
provide	O
a	O
starting	O
point	O
for	O
measuring	O
information	O
technology	O
growth	O
and	O
budget	B-KEY
allocations	O
in	O
this	O
publishing	B-KEY
segment	O

The	O
crossing	B-KEY
number	I-KEY
of	O
P	O
-LRB-	O
N	O
,	O
3	O
-RRB-	O
It	O
is	O
proved	O
that	O
the	O
crossing	B-KEY
number	I-KEY
of	O
the	O
generalized	B-KEY
Petersen	I-KEY
graph	I-KEY
P	O
-LRB-	O
3k	O
+	O
h	O
,	O
3	O
-RRB-	O
is	O
k	O
+	O
h	O
if	O
h	O
in	O
-LCB-	O
0	O
,	O
2	O
-RCB-	O
and	O
k	O
+	O
3	O
if	O
h	O
=	O
1	O
,	O
for	O
each	O
k	O
>	O
or	O
=	O
3	O
,	O
with	O
the	O
single	O
exception	O
of	O
P	O
-LRB-	O
9,3	O
-RRB-	O
,	O
whose	O
crossing	B-KEY
number	I-KEY
is	O
2	O

Comparison	O
between	O
discrete	B-KEY
STFT	I-KEY
and	O
wavelets	B-KEY
for	O
the	O
analysis	O
of	O
power	B-KEY
quality	I-KEY
events	I-KEY
This	O
paper	O
deals	O
with	O
the	O
comparison	O
of	O
signal	B-KEY
processing	I-KEY
tools	I-KEY
for	O
power	O
quality	O
analysis	O
.	O
Two	O
signal	B-KEY
processing	I-KEY
techniques	I-KEY
are	O
considered	O
:	O
the	O
wavelet	B-KEY
filters	O
and	O
the	O
discrete	O
short-time	O
Fourier	O
transforms	O
-LRB-	O
STFT	O
-RRB-	O
.	O
Then	O
,	O
examples	O
of	O
the	O
two	O
most	O
frequent	O
disturbances	O
met	O
in	O
the	O
power	O
system	O
are	O
chosen	O
.	O
An	O
adjustable	B-KEY
speed	I-KEY
drive	I-KEY
with	O
a	O
six-pulse	B-KEY
converter	I-KEY
using	O
EMTP/ATP	B-KEY
is	O
designed	O
and	O
normal	O
energizing	O
of	O
utility	B-KEY
capacitors	I-KEY
is	O
presented	O
.	O
The	O
analysis	O
is	O
tested	O
on	O
a	O
system	O
consisting	O
of	O
13	O
buses	O
and	O
is	O
representative	O
of	O
a	O
medium-sized	B-KEY
industrial	I-KEY
plant	I-KEY
.	O
Finally	O
,	O
each	O
kind	O
of	O
electrical	B-KEY
disturbance	I-KEY
is	O
analyzed	O
with	O
examples	O
representing	O
each	O
tool	O
.	O
A	O
qualitative	O
comparison	O
of	O
results	O
shows	O
the	O
advantages	O
and	O
drawbacks	O
of	O
each	O
signal	B-KEY
processing	I-KEY
technique	I-KEY
applied	O
to	O
power	O
quality	O
analysis	O

A	O
comparative	O
study	O
of	O
some	O
generalized	B-KEY
rough	I-KEY
approximations	I-KEY
In	O
this	O
paper	O
we	O
focus	O
upon	O
a	O
comparison	O
of	O
some	O
generalized	B-KEY
rough	I-KEY
approximations	I-KEY
of	O
sets	O
,	O
where	O
the	O
classical	B-KEY
indiscernibility	I-KEY
relation	I-KEY
is	O
generalized	O
to	O
any	O
binary	B-KEY
reflexive	I-KEY
relation	I-KEY
.	O
We	O
aim	O
at	O
finding	O
the	O
best	O
of	O
several	O
candidates	O
for	O
generalized	B-KEY
rough	I-KEY
approximation	I-KEY
mappings	O
,	O
where	O
both	O
definability	O
of	O
sets	O
by	O
elementary	O
granules	O
of	O
information	O
as	O
well	O
as	O
the	O
issue	O
of	O
distinction	O
among	O
positive	O
,	O
negative	O
,	O
and	O
border	O
regions	O
of	O
a	O
set	O
are	O
taken	O
into	O
account	O

The	O
visible	O
cement	O
data	O
set	O
With	O
advances	O
in	O
x-ray	B-KEY
microtomography	I-KEY
,	O
it	O
is	O
now	O
possible	O
to	O
obtain	O
three-dimensional	O
representations	O
of	O
a	O
material	O
's	O
microstructure	B-KEY
with	O
a	O
voxel	B-KEY
size	I-KEY
of	O
less	O
than	O
one	O
micrometer	O
.	O
The	O
Visible	O
Cement	O
Data	O
Set	O
represents	O
a	O
collection	O
of	O
3-D	O
data	O
sets	O
obtained	O
using	O
the	O
European	B-KEY
Synchrotron	I-KEY
Radiation	I-KEY
Facility	I-KEY
in	O
Grenoble	O
,	O
France	O
in	O
September	O
2000	O
.	O
Most	O
of	O
the	O
images	O
obtained	O
are	O
for	O
hydrating	B-KEY
portland	I-KEY
cement	I-KEY
pastes	I-KEY
,	O
with	O
a	O
few	O
data	O
sets	O
representing	O
hydrating	O
Plaster	B-KEY
of	I-KEY
Paris	I-KEY
and	O
a	O
common	O
building	B-KEY
brick	I-KEY
.	O
All	O
of	O
these	O
data	O
sets	O
are	O
being	O
made	O
available	O
on	O
the	O
Visible	O
Cement	O
Data	O
Set	O
website	O
at	O
http://visiblecement.nist.gov	O
.	O
The	O
website	O
includes	O
the	O
raw	O
3-D	O
datafiles	O
,	O
a	O
description	O
of	O
the	O
material	O
imaged	O
for	O
each	O
data	O
set	O
,	O
example	O
two-dimensional	B-KEY
images	I-KEY
and	O
visualizations	O
for	O
each	O
data	O
set	O
,	O
and	O
a	O
collection	O
of	O
C	O
language	O
computer	O
programs	O
that	O
will	O
be	O
of	O
use	O
in	O
processing	O
and	O
analyzing	O
the	O
3-D	O
microstructural	B-KEY
images	O
.	O
This	O
paper	O
provides	O
the	O
details	O
of	O
the	O
experiments	O
performed	O
at	O
the	O
ESRF	B-KEY
,	O
the	O
analysis	O
procedures	O
utilized	O
in	O
obtaining	O
the	O
data	O
set	O
files	O
,	O
and	O
a	O
few	O
representative	O
example	O
images	O
for	O
each	O
of	O
the	O
three	O
materials	O
investigated	O

Design	O
and	O
implementation	O
of	O
a	O
flexible	O
manufacturing	O
control	O
system	O
using	O
neural	O
network	O
Design	O
and	O
implementation	O
of	O
a	O
sequential	B-KEY
controller	I-KEY
based	O
on	O
the	O
concept	O
of	O
artificial	O
neural	O
networks	O
for	O
a	O
flexible	B-KEY
manufacturing	I-KEY
system	I-KEY
are	O
presented	O
.	O
The	O
recurrent	B-KEY
neural	I-KEY
network	I-KEY
-LRB-	O
RNN	O
-RRB-	O
type	O
is	O
used	O
for	O
such	O
a	O
purpose	O
.	O
Contrary	O
to	O
the	O
programmable	B-KEY
controller	I-KEY
,	O
an	O
RNN-based	O
sequential	B-KEY
controller	I-KEY
is	O
based	O
on	O
a	O
definite	O
mathematical	O
model	O
rather	O
than	O
depending	O
on	O
the	O
experience	O
and	O
trial	O
and	O
error	O
techniques	O
.	O
The	O
proposed	O
controller	O
is	O
also	O
more	O
flexible	O
because	O
it	O
is	O
not	O
limited	O
by	O
the	O
restrictions	O
of	O
the	O
finite	B-KEY
state	I-KEY
automata	I-KEY
theory	O
.	O
Adequate	O
guidelines	O
of	O
how	O
to	O
construct	O
an	O
RNN-based	O
sequential	B-KEY
controller	I-KEY
are	O
presented	O
.	O
These	O
guidelines	O
are	O
applied	O
to	O
different	O
case	O
studies	O
.	O
The	O
proposed	O
controller	O
is	O
tested	O
by	O
simulations	O
and	O
real-time	O
experiments	O
.	O
These	O
tests	O
prove	O
the	O
successfulness	O
of	O
the	O
proposed	O
controller	O
performances	O
.	O
Theoretical	O
as	O
well	O
as	O
experimental	O
results	O
are	O
presented	O
and	O
discussed	O
indicating	O
that	O
the	O
proposed	O
design	O
procedure	O
using	O
Elman	O
's	O
RNN	O
can	O
be	O
effective	O
in	O
designing	O
a	O
sequential	B-KEY
controller	I-KEY
for	O
event-based	O
type	O
manufacturing	O
systems	O
.	O
In	O
addition	O
,	O
the	O
simulation	O
results	O
assure	O
the	O
effectiveness	O
of	O
the	O
proposed	O
controller	O
to	O
overcome	O
the	O
effect	O
of	O
noisy	B-KEY
inputs	I-KEY

The	O
use	O
of	O
subtypes	B-KEY
and	O
stereotypes	B-KEY
in	O
the	O
UML	B-KEY
model	I-KEY
Based	O
on	O
users	O
'	O
experiences	O
of	O
Version	O
1.3	O
of	O
the	O
Unified	B-KEY
Modeling	I-KEY
Language	I-KEY
-LRB-	O
UML	O
-RRB-	O
of	O
the	O
Object	B-KEY
Management	I-KEY
Group	I-KEY
-LRB-	O
OMG	O
-RRB-	O
,	O
a	O
Request	B-KEY
For	I-KEY
Information	I-KEY
in	O
1999	O
elicited	O
several	O
responses	O
which	O
were	O
asked	O
to	O
identify	O
``	O
problems	O
''	O
but	O
not	O
to	O
offer	O
any	O
solutions	O
.	O
One	O
of	O
these	O
responses	O
is	O
examined	O
for	O
``	O
problems	O
''	O
relating	O
to	O
the	O
UML	O
metamodel	O
and	O
here	O
some	O
solutions	O
to	O
the	O
problems	O
identified	O
there	O
are	O
proposed	O
.	O
Specifically	O
,	O
we	O
evaluate	O
the	O
metamodel	O
relating	O
to	O
stereotypes	B-KEY
versus	O
subtypes	B-KEY
;	O
the	O
various	O
kinds	O
of	O
Classifier	B-KEY
-LRB-	O
particularly	O
Types	O
,	O
Interfaces	O
and	O
Classes	O
-RRB-	O
;	O
the	O
introduction	O
of	O
a	O
new	O
subtype	B-KEY
for	O
the	O
whole	B-KEY
part	I-KEY
relationship	I-KEY
;	O
as	O
well	O
as	O
identifying	O
areas	O
in	O
the	O
metamodel	O
where	O
the	O
UML	O
seems	O
to	O
have	O
been	O
used	O
inappropriately	O
in	O
the	O
very	O
definition	O
of	O
the	O
UML	O
's	O
metamodel	O

Optimal	B-KEY
estimation	I-KEY
of	O
a	O
finite	B-KEY
sample	I-KEY
of	O
a	O
discrete	B-KEY
chaotic	I-KEY
process	I-KEY
The	O
synthesis	O
of	O
optimal	O
algorithms	O
for	O
estimating	O
discrete	B-KEY
chaotic	I-KEY
processes	I-KEY
specified	O
by	O
a	O
finite	B-KEY
sample	I-KEY
is	O
considered	O
;	O
various	O
possible	O
approaches	O
are	O
discussed	O
.	O
Expressions	O
determining	O
the	O
potential	O
accuracy	O
in	O
estimating	O
a	O
single	O
value	O
of	O
the	O
chaotic	O
process	O
are	O
derived	O
.	O
An	O
example	O
of	O
the	O
application	O
of	O
the	O
general	O
equations	O
obtained	O
is	O
given	O

An	O
adaptive	B-KEY
sphere-fitting	I-KEY
method	I-KEY
for	O
sequential	B-KEY
tolerance	I-KEY
control	I-KEY
The	O
machining	O
of	O
complex	O
parts	O
typically	O
involves	O
a	O
logical	O
and	O
chronological	O
sequence	O
of	O
n	O
operations	O
on	O
m	O
machine	B-KEY
tools	I-KEY
.	O
Because	O
manufacturing	O
datums	O
can	O
not	O
always	O
match	O
design	B-KEY
constraints	I-KEY
,	O
some	O
of	O
the	O
design	O
specifications	O
imposed	O
on	O
the	O
part	O
are	O
usually	O
satisfied	O
by	O
distinct	O
subsets	O
of	O
the	O
n	O
operations	O
prescribed	O
in	O
the	O
process	O
plan	O
.	O
Conventional	O
tolerance	O
control	O
specifies	O
a	O
fixed	O
set	O
point	O
for	O
each	O
operation	O
and	O
a	O
permissible	O
variation	O
about	O
this	O
set	O
point	O
to	O
insure	O
compliance	B-KEY
with	O
the	O
specifications	O
,	O
whereas	O
sequential	B-KEY
tolerance	I-KEY
control	I-KEY
-LRB-	O
STC	O
-RRB-	O
uses	O
real-time	B-KEY
measurement	I-KEY
information	I-KEY
at	O
the	O
completion	O
of	O
one	O
stage	O
to	O
reposition	O
the	O
set	O
point	O
for	O
subsequent	O
operations	O
.	O
However	O
,	O
it	O
has	O
been	O
shown	O
that	O
earlier	O
sphere-fitting	O
methods	O
for	O
STC	O
can	O
lead	O
to	O
inferior	O
solutions	O
when	O
the	O
process	O
distributions	O
are	O
skewed	O
.	O
This	O
paper	O
introduces	O
an	O
extension	O
of	O
STC	O
that	O
uses	O
an	O
adaptive	B-KEY
sphere-fitting	I-KEY
method	I-KEY
that	O
significantly	O
improves	O
the	O
yield	O
in	O
the	O
presence	O
of	O
skewed	B-KEY
distributions	I-KEY
as	O
well	O
as	O
significantly	O
reducing	O
the	O
computational	B-KEY
effort	I-KEY
required	O
by	O
earlier	O
probabilistic	O
search	O
methods	O

High	O
dynamic	O
control	O
of	O
a	O
three-level	O
voltage-source-converter	O
drive	O
for	O
a	O
main	O
strip	B-KEY
mill	I-KEY
A	O
high	B-KEY
dynamic	I-KEY
control	I-KEY
system	I-KEY
for	O
the	O
Alspa	O
VDM	O
7000	O
medium-voltage	B-KEY
drive	I-KEY
was	O
implemented	O
,	O
which	O
provides	O
fast	O
torque	O
response	O
times	O
of	O
a	O
few	O
milliseconds	O
despite	O
the	O
typically	O
low	O
switching	B-KEY
frequency	I-KEY
of	O
gate-turn-off	B-KEY
thyristors	I-KEY
which	O
is	O
necessary	O
to	O
achieve	O
high	O
efficiency	B-KEY
.	O
The	O
drive	O
system	O
consists	O
of	O
a	O
three-level	B-KEY
voltage-source	I-KEY
converter	I-KEY
with	O
active	O
front	O
end	O
and	O
a	O
synchronous	B-KEY
motor	I-KEY
.	O
The	O
drive	O
has	O
most	O
recently	O
been	O
applied	O
to	O
a	O
main	O
strip	B-KEY
mill	I-KEY
.	O
It	O
provides	O
a	O
maximum	O
of	O
8.3-MW	O
mechanical	B-KEY
power	I-KEY
with	O
a	O
rated	O
motor	B-KEY
voltage	I-KEY
of	O
3	B-KEY
kV	I-KEY
.	O
Besides	O
motor	O
torque	O
as	O
the	O
main	O
control	B-KEY
objective	I-KEY
,	O
the	O
control	O
system	O
has	O
to	O
comply	O
with	O
a	O
number	O
of	O
additional	O
objectives	O
and	O
constraints	O
like	O
DC-link	B-KEY
voltage	I-KEY
regulation	I-KEY
and	O
balancing	O
,	O
current	O
and	O
torque	B-KEY
harmonics	I-KEY
,	O
motor	B-KEY
flux	I-KEY
,	O
and	O
excitation	B-KEY

An	O
approach	O
to	O
developing	O
computational	O
supports	O
for	O
reciprocal	O
tutoring	O
This	O
study	O
presents	O
a	O
novel	O
approach	O
to	O
developing	O
computational	O
supports	O
for	O
reciprocal	O
tutoring	O
.	O
Reciprocal	O
tutoring	O
is	O
a	O
collaborative	B-KEY
learning	I-KEY
activity	O
,	O
where	O
two	O
participants	O
take	O
turns	O
to	O
play	O
the	O
role	O
of	O
a	O
tutor	O
and	O
a	O
tutee	O
.	O
The	O
computational	O
supports	O
include	O
scaffolding	B-KEY
tools	I-KEY
for	O
the	O
tutor	O
and	O
a	O
computer-simulated	B-KEY
virtual	I-KEY
participant	I-KEY
.	O
The	O
approach	O
,	O
including	O
system	B-KEY
architecture	I-KEY
,	O
implementations	O
of	O
scaffolding	B-KEY
tools	I-KEY
for	O
the	O
tutor	O
and	O
of	O
a	O
virtual	O
participant	O
is	O
presented	O
herein	O
.	O
Furthermore	O
,	O
a	O
system	O
for	O
reciprocal	O
tutoring	O
is	O
implemented	O
as	O
an	O
example	O
of	O
the	O
approach	O

Three-dimensional	O
spiral	O
MR	O
imaging	O
:	O
application	O
to	O
renal	B-KEY
multiphase	I-KEY
contrast-enhanced	I-KEY
angiography	I-KEY
A	O
fast	O
MR	O
pulse	O
sequence	O
with	O
spiral	B-KEY
in-plane	I-KEY
readout	I-KEY
and	O
conventional	O
3D	B-KEY
partition	I-KEY
encoding	I-KEY
was	O
developed	O
for	O
multiphase	O
contrast-enhanced	O
magnetic	O
resonance	O
angiography	O
-LRB-	O
CE-MRA	O
-RRB-	O
of	O
the	O
renal	B-KEY
vasculature	I-KEY
.	O
Compared	O
to	O
a	O
standard	O
multiphase	O
3D	O
CE-MRA	O
with	O
FLASH	O
readout	O
,	O
an	O
isotropic	O
in-plane	O
spatial	B-KEY
resolution	I-KEY
of	O
1.4	O
*	O
1.4	O
mm/sup	O
2	O
/	O
over	O
2.0	O
*	O
1.4	O
mm/sup	O
2	O
/	O
could	O
be	O
achieved	O
with	O
a	O
temporal	O
resolution	O
of	O
6	O
sec	O
.	O
The	O
theoretical	O
gain	O
of	O
spatial	B-KEY
resolution	I-KEY
by	O
using	O
the	O
spiral	O
pulse	O
sequence	O
and	O
the	O
performance	O
in	O
the	O
presence	O
of	O
turbulent	O
flow	O
was	O
evaluated	O
in	O
phantom	O
measurements	O
.	O
Multiphase	O
3D	O
CE-MRA	O
of	O
the	O
renal	O
arteries	O
was	O
performed	O
in	O
five	O
healthy	O
volunteers	O
using	O
both	O
techniques	O
.	O
A	O
deblurring	B-KEY
technique	O
was	O
used	O
to	O
correct	O
the	O
spiral	O
raw	O
data	O
.	O
Thereby	O
,	O
the	O
off-resonance	B-KEY
frequencies	I-KEY
were	O
determined	O
by	O
minimizing	O
the	O
imaginary	O
part	O
of	O
the	O
data	O
in	O
image	O
space	O
.	O
The	O
chosen	O
correction	O
algorithm	O
was	O
able	O
to	O
reduce	O
image	O
blurring	O
substantially	O
in	O
all	O
MRA	O
phases	O
.	O
The	O
image	B-KEY
quality	I-KEY
of	O
the	O
spiral	O
CE-MRA	O
pulse	O
sequence	O
was	O
comparable	O
to	O
that	O
of	O
the	O
FLASH	O
CE-MRA	O
with	O
increased	O
spatial	B-KEY
resolution	I-KEY
and	O
a	O
25	O
%	O
reduced	B-KEY
contrast-to-noise	I-KEY
ratio	I-KEY
.	O
Additionally	O
,	O
artifacts	O
specific	O
to	O
spiral	O
MRI	O
could	O
be	O
observed	O
which	O
had	O
no	O
impact	O
on	O
the	O
assessment	O
of	O
the	O
renal	O
arteries	O

Verizon	O
leapfrogs	O
Sprint	B-KEY
PCS	I-KEY
with	O
Q2	O
subscriber	O
numbers	O
The	O
wireless	B-KEY
carrier	I-KEY
industry	I-KEY
's	O
second-quarter	O
results	O
showed	O
a	O
surprising	O
shift	O
in	O
market	O
share	O
as	O
Sprint	B-KEY
PCS	I-KEY
fell	O
from	O
grace	O
after	O
a	O
nearly	O
four-year	O
lead	O
in	O
subscriber	O
additions	O
and	O
Verizon	B-KEY
Wireless	I-KEY
added	O
considerably	O
more	O
customers	O
than	O
analysts	O
expected	O

How	O
does	O
attitude	O
impact	O
IT	O
implementation	O
:	O
a	O
study	O
of	O
small	B-KEY
business	I-KEY
owners	I-KEY
According	O
to	O
previous	O
studies	O
,	O
attitude	O
towards	O
information	O
technology	O
-LRB-	O
IT	O
-RRB-	O
among	O
small	B-KEY
business	I-KEY
owners	I-KEY
appears	O
to	O
be	O
a	O
key	O
factor	O
in	O
achieving	O
high	O
quality	O
IT	O
implementations	O
.	O
In	O
an	O
effort	O
to	O
extend	O
this	O
stream	O
of	O
research	O
,	O
we	O
conducted	O
case	O
studies	O
with	O
small	B-KEY
business	I-KEY
owners	I-KEY
and	O
learned	O
that	O
high	O
quality	O
IT	O
implementations	O
resulted	O
with	O
owners	O
who	O
had	O
positive	O
or	O
negative	B-KEY
attitudes	I-KEY
toward	O
IT	O
,	O
but	O
not	O
with	O
owners	O
who	O
had	O
uncertain	B-KEY
attitudes	I-KEY
.	O
Owners	O
with	O
apolar	O
attitude	O
,	O
either	O
positive	O
or	O
negative	O
,	O
all	O
took	O
action	O
to	O
temper	O
the	O
uncertainty	O
and	O
risk	B-KEY
surrounding	O
the	O
use	O
of	O
new	O
IT	O
in	O
their	O
organization	B-KEY
.	O
In	O
contrast	O
,	O
owners	O
with	O
uncertain	B-KEY
attitudes	I-KEY
did	O
not	O
make	O
mitigating	O
attempts	O
to	O
reduce	O
uncertainty	O
and	O
risk	B-KEY
.	O
A	O
consistent	O
finding	O
among	O
those	O
with	O
high	O
quality	O
IT	O
implementations	O
was	O
an	O
entrepreneurial	O
,	O
or	O
shared	O
,	O
management	B-KEY
style	I-KEY
.	O
It	O
is	O
proposed	O
,	O
based	O
on	O
case	O
study	O
data	O
,	O
that	O
small	B-KEY
business	I-KEY
owners	I-KEY
with	O
an	O
uncertain	B-KEY
attitude	I-KEY
towards	O
IT	O
might	O
experience	O
higher	O
quality	O
IT	O
results	O
in	O
their	O
organizations	B-KEY
through	O
practicing	O
a	O
more	O
entrepreneurial	O
,	O
or	O
shared	O
,	O
management	B-KEY
style	I-KEY
.	O
The	O
study	O
provides	O
insights	O
for	O
both	O
computer	B-KEY
specialists	I-KEY
and	O
small	B-KEY
business	I-KEY
owners	I-KEY
planning	B-KEY
IT	O
implementations	O

Abundance	O
of	O
mosaic	B-KEY
patterns	I-KEY
for	O
CNN	B-KEY
with	O
spatially	B-KEY
variant	I-KEY
templates	I-KEY
This	O
work	O
investigates	O
the	O
complexity	O
of	O
one-dimensional	B-KEY
cellular	I-KEY
neural	I-KEY
network	I-KEY
mosaic	B-KEY
patterns	I-KEY
with	O
spatially	B-KEY
variant	I-KEY
templates	I-KEY
on	O
finite	O
and	O
infinite	B-KEY
lattices	I-KEY
.	O
Various	O
boundary	B-KEY
conditions	I-KEY
are	O
considered	O
for	O
finite	B-KEY
lattices	I-KEY
and	O
the	O
exact	O
number	O
of	O
mosaic	B-KEY
patterns	I-KEY
is	O
computed	O
precisely	O
.	O
The	O
entropy	O
of	O
mosaic	B-KEY
patterns	I-KEY
with	O
periodic	O
templates	O
can	O
also	O
be	O
calculated	O
for	O
infinite	B-KEY
lattices	I-KEY
.	O
Furthermore	O
,	O
we	O
show	O
the	O
abundance	O
of	O
mosaic	B-KEY
patterns	I-KEY
with	O
respect	O
to	O
template	O
periods	O
and	O
,	O
which	O
differ	O
greatly	O
from	O
cases	O
with	O
spatially	O
invariant	O
templates	O

Improved	O
analysis	O
for	O
the	O
nonlinear	B-KEY
performance	I-KEY
of	O
CMOS	B-KEY
current	I-KEY
mirrors	I-KEY
with	O
device	B-KEY
mismatch	I-KEY
The	O
nonlinear	B-KEY
performance	I-KEY
of	O
the	O
simple	O
and	O
complementary	B-KEY
MOSFET	I-KEY
current	I-KEY
mirrors	I-KEY
are	O
analyzed	O
.	O
Closed-form	B-KEY
expressions	I-KEY
are	O
obtained	O
for	O
the	O
harmonic	O
and	O
intermodulation	B-KEY
components	I-KEY
resulting	O
from	O
a	O
multisinusoidal	B-KEY
input	I-KEY
current	I-KEY
.	O
These	O
expressions	O
can	O
be	O
used	O
for	O
predicting	O
the	O
limiting	O
values	O
of	O
the	O
input	B-KEY
current	I-KEY
under	O
prespecified	O
conditions	O
of	O
threshold-voltage	B-KEY
mismatches	I-KEY
and/or	O
transconductance	B-KEY
mismatches	I-KEY
.	O
The	O
case	O
of	O
a	O
single	O
input	O
sinusoid	O
is	O
discussed	O
in	O
detail	O
and	O
the	O
results	O
are	O
compared	O
with	O
SPICE	B-KEY
simulations	I-KEY

A	O
network	B-KEY
simplex	I-KEY
algorithm	I-KEY
with	O
O	O
-LRB-	O
n	O
-RRB-	O
consecutive	O
degenerate	B-KEY
pivots	I-KEY
We	O
suggest	O
a	O
pivot	O
rule	O
for	O
the	O
primal	O
simplex	O
algorithm	O
for	O
the	O
minimum	B-KEY
cost	I-KEY
flow	I-KEY
problem	I-KEY
,	O
known	O
as	O
the	O
network	B-KEY
simplex	I-KEY
algorithm	I-KEY
.	O
Due	O
to	O
degeneracy	B-KEY
,	O
cycling	B-KEY
may	O
occur	O
in	O
the	O
network	B-KEY
simplex	I-KEY
algorithm	I-KEY
.	O
The	O
cycling	B-KEY
can	O
be	O
prevented	O
by	O
maintaining	O
strongly	O
feasible	O
bases	O
proposed	O
by	O
Cunningham	O
-LRB-	O
1976	O
-RRB-	O
;	O
however	O
,	O
if	O
we	O
do	O
not	O
impose	O
any	O
restrictions	O
on	O
the	O
entering	O
variables	O
,	O
the	O
algorithm	O
can	O
still	O
perform	O
an	O
exponentially	O
long	O
sequence	O
of	O
degenerate	B-KEY
pivots	I-KEY
.	O
This	O
phenomenon	O
is	O
known	O
as	O
stalling	B-KEY
.	O
Researchers	O
have	O
suggested	O
several	O
pivot	O
rules	O
with	O
the	O
following	O
bounds	O
on	O
the	O
number	O
of	O
consecutive	O
degenerate	B-KEY
pivots	I-KEY
:	O
m	O
,	O
n/sup	O
2	O
/	O
,	O
k	O
-LRB-	O
k	O
+	O
1	O
-RRB-	O
/	O
2	O
,	O
where	O
n	O
is	O
the	O
number	O
of	O
nodes	O
in	O
the	O
network	O
,	O
m	O
is	O
the	O
number	O
of	O
arcs	O
in	O
the	O
network	O
,	O
and	O
k	O
is	O
the	O
number	O
of	O
degenerate	O
arcs	O
in	O
the	O
basis	O
.	O
-LRB-	O
Observe	O
that	O
k	O
<	O
or	O
=	O
n.	O
-RRB-	O
In	O
this	O
paper	O
,	O
we	O
describe	O
an	O
anti-stalling	B-KEY
pivot	I-KEY
rule	I-KEY
that	O
ensures	O
that	O
the	O
network	B-KEY
simplex	I-KEY
algorithm	I-KEY
performs	O
at	O
most	O
k	O
consecutive	O
degenerate	B-KEY
pivots	I-KEY
.	O
This	O
rule	O
uses	O
a	O
negative	B-KEY
cost	I-KEY
augmenting	I-KEY
cycle	I-KEY
to	O
identify	O
a	O
sequence	O
of	O
entering	O
variables	O

Knowledge	B-KEY
acquisition	I-KEY
and	O
ontology	B-KEY
modelling	I-KEY
for	O
construction	O
of	O
a	O
control	B-KEY
and	I-KEY
monitoring	I-KEY
expert	I-KEY
system	I-KEY
This	O
paper	O
presents	O
the	O
processes	O
of	O
knowledge	B-KEY
acquisition	I-KEY
and	O
ontology	O
development	O
for	O
structuring	O
the	O
knowledge	B-KEY
base	I-KEY
of	O
an	O
expert	O
system	O
.	O
Ontological	B-KEY
engineering	I-KEY
is	O
a	O
process	O
that	O
facilitates	O
construction	O
of	O
the	O
knowledge	B-KEY
base	I-KEY
of	O
an	O
intelligent	B-KEY
system	I-KEY
.	O
Ontology	O
is	O
the	O
study	O
of	O
the	O
organization	O
and	O
classification	O
of	O
knowledge	O
.	O
Ontological	B-KEY
engineering	I-KEY
in	O
artificial	B-KEY
intelligence	I-KEY
has	O
the	O
practical	O
goal	O
of	O
constructing	O
frameworks	O
for	O
knowledge	O
that	O
allow	O
computational	O
systems	O
to	O
tackle	O
knowledge-intensive	O
problems	O
and	O
it	O
supports	O
knowledge	O
sharing	O
and	O
reuse	O
.	O
To	O
illustrate	O
the	O
process	O
of	O
conceptual	O
modelling	O
using	O
the	O
Inferential	B-KEY
Modelling	I-KEY
Technique	I-KEY
as	O
a	O
basis	O
for	O
ontology	O
construction	O
,	O
the	O
tool	O
and	O
processes	O
are	O
applied	O
to	O
build	O
an	O
expert	O
system	O
in	O
the	O
domain	O
of	O
monitoring	O
of	O
a	O
petroleum-production	B-KEY
facility	I-KEY

A	O
case	O
for	O
end	B-KEY
system	I-KEY
multicast	I-KEY
The	O
conventional	O
wisdom	O
has	O
been	O
that	O
Internet	B-KEY
protocol	I-KEY
-LRB-	O
IP	O
-RRB-	O
is	O
the	O
natural	O
protocol	B-KEY
layer	I-KEY
for	O
implementing	O
multicast	O
related	O
functionality	O
.	O
However	O
,	O
more	O
than	O
a	O
decade	O
after	O
its	O
initial	O
proposal	O
,	O
IP	B-KEY
multicast	I-KEY
is	O
still	O
plagued	O
with	O
concerns	O
pertaining	O
to	O
scalability	O
,	O
network	B-KEY
management	I-KEY
,	O
deployment	O
,	O
and	O
support	O
for	O
higher	B-KEY
layer	I-KEY
functionality	I-KEY
such	O
as	O
error	O
,	O
flow	O
,	O
and	O
congestion	B-KEY
control	I-KEY
.	O
We	O
explore	O
an	O
alternative	O
architecture	O
that	O
we	O
term	O
end	B-KEY
system	I-KEY
multicast	I-KEY
,	O
where	O
end	O
systems	O
implement	O
all	O
multicast	O
related	O
functionality	O
including	O
membership	B-KEY
management	I-KEY
and	O
packet	B-KEY
replication	I-KEY
.	O
This	O
shifting	O
of	O
multicast	O
support	O
from	O
routers	O
to	O
end	O
systems	O
has	O
the	O
potential	O
to	O
address	O
most	O
problems	O
associated	O
with	O
IP	B-KEY
multicast	I-KEY
.	O
However	O
,	O
the	O
key	O
concern	O
is	O
the	O
performance	B-KEY
penalty	I-KEY
associated	O
with	O
such	O
a	O
model	O
.	O
In	O
particular	O
,	O
end	B-KEY
system	I-KEY
multicast	I-KEY
introduces	O
duplicate	O
packets	O
on	O
physical	O
links	O
and	O
incurs	O
larger	O
end-to-end	B-KEY
delays	I-KEY
than	O
IP	B-KEY
multicast	I-KEY
.	O
We	O
study	O
these	O
performance	O
concerns	O
in	O
the	O
context	O
of	O
the	O
Narada	B-KEY
protocol	I-KEY
.	O
In	O
Narada	O
,	O
end	O
systems	O
self-organize	O
into	O
an	O
overlay	B-KEY
structure	I-KEY
using	O
a	O
fully	O
distributed	B-KEY
protocol	I-KEY
.	O
Further	O
,	O
end	O
systems	O
attempt	O
to	O
optimize	O
the	O
efficiency	O
of	O
the	O
overlay	O
by	O
adapting	O
to	O
network	B-KEY
dynamics	I-KEY
and	O
by	O
considering	O
application	B-KEY
level	I-KEY
performance	I-KEY
.	O
We	O
present	O
details	O
of	O
Narada	O
and	O
evaluate	O
it	O
using	O
both	O
simulation	B-KEY
and	O
Internet	B-KEY
experiments	I-KEY
.	O
Our	O
results	O
indicate	O
that	O
the	O
performance	B-KEY
penalties	I-KEY
are	O
low	O
both	O
from	O
the	O
application	O
and	O
the	O
network	O
perspectives	O
.	O
We	O
believe	O
the	O
potential	O
benefits	O
of	O
transferring	O
multicast	O
functionality	O
from	O
end	O
systems	O
to	O
routers	O
significantly	O
outweigh	O
the	O
performance	B-KEY
penalty	I-KEY
incurred	O

High-performance	O
numerical	O
pricing	B-KEY
methods	O
The	O
pricing	B-KEY
of	O
financial	O
derivatives	O
is	O
an	O
important	O
field	O
in	O
finance	B-KEY
and	O
constitutes	O
a	O
major	O
component	O
of	O
financial	B-KEY
management	I-KEY
applications	O
.	O
The	O
uncertainty	O
of	O
future	O
events	O
often	O
makes	O
analytic	O
approaches	O
infeasible	O
and	O
,	O
hence	O
,	O
time-consuming	O
numerical	O
simulations	O
are	O
required	O
.	O
In	O
the	O
Aurora	O
Financial	B-KEY
Management	I-KEY
System	O
,	O
pricing	O
is	O
performed	O
on	O
the	O
basis	O
of	O
lattice	O
representations	O
of	O
stochastic	O
multidimensional	O
scenario	O
processes	O
using	O
the	O
Monte	O
Carlo	O
simulation	O
and	O
Backward	O
Induction	O
methods	O
,	O
the	O
latter	O
allowing	O
for	O
the	O
exploitation	O
of	O
shared-memory	O
parallelism	O
.	O
We	O
present	O
the	O
parallelization	O
of	O
a	O
Backward	O
Induction	O
numerical	O
pricing	B-KEY
kernel	O
on	O
a	O
cluster	O
of	O
SMPs	O
using	O
HPF	O
+	O
,	O
an	O
extended	O
version	O
of	O
High-Performance	O
Fortran	O
.	O
Based	O
on	O
language	O
extensions	O
for	O
specifying	O
a	O
hierarchical	O
mapping	O
of	O
data	O
onto	O
an	O
SMP	O
cluster	O
,	O
the	O
compiler	O
generates	O
a	O
hybrid-parallel	O
program	O
combining	O
distributed-memory	O
and	O
shared-memory	O
parallelism	O
.	O
We	O
outline	O
the	O
parallelization	O
strategy	O
adopted	O
by	O
the	O
VFC	O
compiler	O
and	O
present	O
an	O
experimental	O
evaluation	O
of	O
the	O
pricing	B-KEY
kernel	O
on	O
an	O
NEC	O
SX-5	O
vector	O
supercomputer	O
and	O
a	O
Linux	O
SMP	O
cluster	O
,	O
comparing	O
a	O
pure	O
MPI	O
version	O
to	O
a	O
hybrid-parallel	O
MPI/OpenMP	O
version	O

Deterministic	B-KEY
calculations	I-KEY
of	O
photon	O
spectra	O
for	O
clinical	O
accelerator	O
targets	O
A	O
method	O
is	O
proposed	O
to	O
compute	O
photon	B-KEY
energy	I-KEY
spectra	I-KEY
produced	O
in	O
clinical	B-KEY
electron	I-KEY
accelerator	I-KEY
targets	I-KEY
,	O
based	O
on	O
the	O
deterministic	O
solution	O
of	O
the	O
Boltzmann	B-KEY
equation	I-KEY
for	O
coupled	B-KEY
electron-photon	I-KEY
transport	I-KEY
in	O
one-dimensional	O
-LRB-	O
1-D	O
-RRB-	O
slab	O
geometry	O
.	O
It	O
is	O
shown	O
that	O
the	O
deterministic	O
method	O
gives	O
similar	O
results	O
as	O
Monte	O
Carlo	O
calculations	O
over	O
the	O
angular	B-KEY
range	I-KEY
of	I-KEY
interest	I-KEY
for	O
therapy	B-KEY
applications	I-KEY
.	O
Relative	B-KEY
energy	I-KEY
spectra	I-KEY
computed	O
by	O
deterministic	O
and	O
3-D	B-KEY
Monte	I-KEY
Carlo	I-KEY
methods	I-KEY
,	O
respectively	O
,	O
are	O
compared	O
for	O
several	O
realistic	O
target	O
materials	O
and	O
different	O
electron	O
beams	O
,	O
and	O
are	O
found	O
to	O
give	O
similar	O
photon	O
energy	O
distributions	O
and	O
mean	O
energies	O
.	O
The	O
deterministic	B-KEY
calculations	I-KEY
typically	O
require	O
1-2	O
mins	O
of	O
execution	O
time	O
on	O
a	O
Sun	O
Workstation	O
,	O
compared	O
to	O
2-36	O
h	O
for	O
the	O
Monte	O
Carlo	O
runs	O

Multiple	B-KEY
criteria	I-KEY
decision	I-KEY
making	I-KEY
without	O
optimization	O
We	O
present	O
a	O
development	O
intended	O
to	O
make	O
interactive	B-KEY
decision	I-KEY
making	I-KEY
schemes	I-KEY
accessible	O
for	O
a	O
wider	O
spectrum	O
of	O
decision	B-KEY
makers	I-KEY
.	O
To	O
this	O
aim	O
we	O
propose	O
to	O
eliminate	O
the	O
need	O
to	O
solve	O
optimization	O
problems	O
at	O
successive	O
iterations	O
of	O
interactive	O
decision	O
processes	O
.	O
We	O
show	O
that	O
the	O
need	O
for	O
optimization	O
can	O
be	O
eliminated	O
by	O
the	O
ability	O
of	O
establishing	O
sufficiently	O
tight	O
bounds	O
on	O
criteria	O
values	O
for	O
efficient	B-KEY
decisions	I-KEY
prior	O
to	O
explicit	O
identification	O
of	O
such	O
decisions	O
.	O
We	O
present	O
a	O
technique	O
,	O
fully	O
operational	O
and	O
numerically	O
simple	O
,	O
for	O
establishing	O
such	O
bounds	O
.	O
Bounds	O
are	O
dynamic	O
,	O
i.e.	O
,	O
they	O
become	O
stronger	O
with	O
the	O
growing	O
number	O
of	O
decisions	O
explicitly	O
identified	O
.	O
They	O
are	O
also	O
parametric	O
with	O
respect	O
to	O
weighting	O
coefficients	O
.	O
We	O
also	O
point	O
out	O
how	O
this	O
technique	O
can	O
enhance	O
the	O
existing	O
interactive	O
decision	O
making	O
methods	O

X-Rite	B-KEY
:	O
more	O
than	O
a	O
graphic	B-KEY
arts	I-KEY
company	O
Although	O
it	O
is	O
well	O
known	O
as	O
a	O
maker	O
of	O
densitometers	O
and	O
spectrophotometers	O
,	O
X-Rite	B-KEY
is	O
active	O
in	O
measuring	O
light	O
and	O
shape	O
in	O
many	O
industries	O
.	O
Among	O
them	O
are	O
automobile	O
finishes	O
,	O
paint	O
and	O
home	O
improvements	O
,	O
scientific	O
instruments	O
,	O
optical	O
semiconductors	O
and	O
even	O
cosmetic	O
dentistry	O

The	O
Malaysian	O
model	O
Japan	O
's	O
first	O
third	O
generation	O
service	O
,	O
Foma	O
,	O
is	O
unlikely	O
to	O
be	O
truly	O
attractive	O
to	O
consumers	O
until	O
2005	O
.	O
That	O
still	O
falls	O
well	O
within	O
the	O
financial	O
planning	O
of	O
its	O
operator	O
Docomo	O
.	O
But	O
where	O
does	O
that	O
leave	O
European	O
3G	B-KEY
operators	I-KEY
looking	O
for	O
reassurance	O
?	O
Malaysia	B-KEY
,	O
says	O
Simon	O
Marshall	O

A	O
dataflow	B-KEY
computer	I-KEY
which	O
accelerates	O
execution	O
of	O
sequential	B-KEY
programs	I-KEY
by	O
precedent	B-KEY
firing	I-KEY
instructions	I-KEY
In	O
the	O
dataflow	O
machine	O
,	O
it	O
is	O
important	O
to	O
avoid	O
degradation	O
of	O
performance	O
in	O
sequential	O
processing	O
,	O
and	O
it	O
is	O
important	O
from	O
the	O
viewpoint	O
of	O
hardware	B-KEY
scale	I-KEY
to	O
reduce	O
the	O
number	O
of	O
waiting	B-KEY
operands	I-KEY
.	O
This	O
paper	O
demonstrates	O
that	O
processing	B-KEY
performance	I-KEY
is	O
degraded	O
by	O
sequential	O
processing	O
in	O
the	O
switching	B-KEY
process	I-KEY
,	O
and	O
presents	O
a	O
method	O
of	O
remedy	O
.	O
Precedent	B-KEY
firing	I-KEY
control	I-KEY
is	O
proposed	O
as	O
a	O
means	O
of	O
remedy	O
,	O
and	O
it	O
is	O
shown	O
by	O
a	O
simulation	O
that	O
the	O
execution	B-KEY
time	I-KEY
and	O
the	O
total	O
number	O
of	O
waiting	B-KEY
operands	I-KEY
can	O
be	O
reduced	O
by	O
the	O
precedent	B-KEY
firing	I-KEY
control	I-KEY
.	O
Then	O
the	O
hardware	B-KEY
scale	I-KEY
is	O
examined	O
as	O
an	O
evaluation	O
of	O
precedent	B-KEY
firing	I-KEY
control	I-KEY

HEW	O
selects	O
network	B-KEY
management	I-KEY
software	I-KEY
For	O
more	O
than	O
100	O
years	O
,	O
Hamburgische	B-KEY
Electricitats-Werke	I-KEY
AG	O
-LRB-	O
HEW	O
-RRB-	O
has	O
provided	O
a	O
reliable	O
electricity	O
service	O
to	O
the	O
city	O
of	O
Hamburg	B-KEY
,	O
Germany	B-KEY
.	O
Today	O
,	O
the	O
company	O
supplies	O
electricity	O
to	O
some	O
1.7	O
million	O
inhabitants	O
via	O
285000	O
connections	O
.	O
During	O
1999	O
,	O
the	O
year	O
the	O
energy	O
market	O
was	O
started	O
in	O
Germany	B-KEY
,	O
HEW	O
needed	O
to	O
operate	O
and	O
maintain	O
a	O
safe	O
and	O
reliable	O
network	O
cheaply	O
.	O
The	O
development	O
and	O
implementation	O
of	O
a	O
distribution	B-KEY
management	I-KEY
system	I-KEY
-LRB-	O
DMS	O
-RRB-	O
is	O
key	O
to	O
the	O
success	O
of	O
HEW	O
.	O
HEW	O
's	O
strategy	O
was	O
to	O
obtain	O
efficient	O
new	O
software	O
for	O
network	O
management	O
that	O
also	O
offered	O
a	O
good	O
platform	O
for	O
future	O
applications	O
.	O
Following	O
a	O
pilot	O
and	O
prequalification	O
phase	O
,	O
HEW	O
invited	O
several	O
companies	O
to	O
process	O
the	O
requirements	O
catalog	O
and	O
to	O
submit	O
a	O
detailed	O
tender	O
.	O
The	O
network	O
information	O
management	O
system	O
,	O
Xpower	B-KEY
,	O
developed	O
by	O
Tekla	B-KEY
Oyj	I-KEY
,	O
successfully	O
passed	O
HEW	O
's	O
test	O
program	O
and	O
satisfied	O
all	O
the	O
performance	O
and	O
system	O
capacity	O
requirements	O
.	O
The	O
system	O
met	O
all	O
HEW	O
's	O
conditions	O
by	O
presenting	O
the	O
reality	O
of	O
a	O
network	O
with	O
the	O
attributes	O
of	O
the	O
operating	O
resources	O
.	O
Xpower	B-KEY
platform	O
provides	O
the	O
ability	O
to	O
integrate	O
future	O
applications	O

Option	B-KEY
pricing	I-KEY
formulas	I-KEY
based	O
on	O
a	O
non-Gaussian	O
stock	O
price	O
model	O
Options	O
are	O
financial	B-KEY
instruments	I-KEY
that	O
depend	O
on	O
the	O
underlying	O
stock	O
.	O
We	O
explain	O
their	O
non-Gaussian	O
fluctuations	O
using	O
the	O
nonextensive	B-KEY
thermodynamics	I-KEY
parameter	I-KEY
q	O
.	O
A	O
generalized	O
form	O
of	O
the	O
Black-Scholes	O
-LRB-	O
BS	O
-RRB-	O
partial	O
differential	O
equation	O
-LRB-	O
1973	O
-RRB-	O
and	O
some	O
closed-form	B-KEY
solutions	I-KEY
are	O
obtained	O
.	O
The	O
standard	O
BS	O
equation	O
-LRB-	O
q	O
=	O
1	O
-RRB-	O
which	O
is	O
used	O
by	O
economists	O
to	O
calculate	O
option	O
prices	O
requires	O
multiple	O
values	O
of	O
the	O
stock	B-KEY
volatility	I-KEY
-LRB-	O
known	O
as	O
the	O
volatility	B-KEY
smile	I-KEY
-RRB-	O
.	O
Using	O
q	O
=	O
1.5	O
which	O
well	O
models	O
the	O
empirical	B-KEY
distribution	I-KEY
of	O
returns	O
,	O
we	O
get	O
a	O
good	O
description	O
of	O
option	O
prices	O
using	O
a	O
single	O
volatility	O

Training	O
for	O
trouble	O
In	O
a	O
security	O
context	O
,	O
one	O
example	O
of	O
digitized	O
video	O
's	O
integration	O
into	O
a	O
networked	B-KEY
knowledge	I-KEY
base	I-KEY
is	O
found	O
in	O
the	O
Accident	B-KEY
Response	I-KEY
Group	I-KEY
-LRB-	O
ARG	O
-RRB-	O
at	O
Sandia	B-KEY
National	I-KEY
Labs	I-KEY
.	O
A	O
``	O
national	B-KEY
security	I-KEY
laboratory	I-KEY
''	O
headquartered	O
in	O
Albuquerque	O
,	O
New	O
Mexico	O
,	O
Sandia	O
is	O
operated	O
by	O
Lockheed	O
Martin	O
and	O
primarily	O
funded	O
by	O
the	O
U.S.	O
Department	O
of	O
Energy	O
.	O
The	O
organization	O
handles	O
research	O
,	O
design	O
and	O
development	O
of	O
all	O
non-nuclear	O
components	O
used	O
in	O
U.S.	B-KEY
nuclear	I-KEY
weapons	I-KEY
programs	I-KEY
,	I-KEY
and	O
is	O
involved	O
as	O
well	O
in	O
programs	O
related	O
to	O
energy	O
,	O
critical	O
infrastructure	O
,	O
non-proliferation	O
,	O
materials	O
control	O
,	O
and	O
emerging	O
threats	O
.	O
ARG	O
's	O
searchable	B-KEY
video	I-KEY
database	I-KEY
has	O
been	O
implemented	O
using	O
the	O
Screening	B-KEY
Room	I-KEY
package	I-KEY
of	O
applications	O
from	O
Convera	O
in	O
Vienna	O
,	O
Virginia	O
.	O
Formed	O
in	O
December	O
2000	O
from	O
Excalibur	O
Technologies	O
and	O
Intel	O
's	O
Interactive	O
Media	O
Services	O
Division	O
,	O
Convera	O
targets	O
corporate	O
and	O
institutional	O
markets	O
with	O
products	O
for	O
securely	O
accessing	O
,	O
indexing	O
,	O
and	O
searching	O
rich	O
media	O
content-text	O
,	O
images	O
,	O
audio	O
,	O
and	O
video-across	O
interconnected	O
computer	O
networks	O
.	O
Among	O
its	O
public-sector	B-KEY
clients	I-KEY
are	O
the	O
FBI	O
,	O
NASA	O
,	O
the	O
Nuclear	O
Regulatory	O
Commission	O
,	O
U.S.	O
military	O
services	O
,	O
the	O
Departments	O
of	O
Justice	O
and	O
State	O
,	O
and	O
various	O
domestic	O
and	O
foreign	O
intelligence	O
agencies	O

Quantum	B-KEY
learning	I-KEY
and	O
universal	B-KEY
quantum	I-KEY
matching	I-KEY
machine	I-KEY
Suppose	O
that	O
three	O
kinds	O
of	O
quantum	O
systems	O
are	O
given	O
in	O
some	O
unknown	O
states	O
|	O
f	O
>	O
/	O
sup	O
-LRB-	O
X	O
-RRB-	O
N	O
/	O
,	O
|	O
g/sub	O
1	O
/	O
>	O
/	O
sup	O
-LRB-	O
X	O
-RRB-	O
K	O
/	O
,	O
and	O
|	O
g/sub	O
2	O
/	O
>	O
/	O
sup	O
-LRB-	O
X	O
-RRB-	O
K	O
/	O
,	O
and	O
we	O
want	O
to	O
decide	O
which	O
template	O
state	O
|	O
g/sub	O
1	O
/	O
>	O
or	O
|	O
g/sub	O
2	O
/	O
>	O
,	O
each	O
representing	O
the	O
feature	O
of	O
the	O
pattern	B-KEY
class	I-KEY
C/sub	O
1	O
/	O
or	O
C/sub	O
2	O
/	O
,	O
respectively	O
,	O
is	O
closest	O
to	O
the	O
input	O
feature	O
state	O
|	O
f	O
>	O
.	O
This	O
is	O
an	O
extension	O
of	O
the	O
pattern	B-KEY
matching	I-KEY
problem	I-KEY
into	O
the	O
quantum	B-KEY
domain	I-KEY
.	O
Assuming	O
that	O
these	O
states	O
are	O
known	O
a	O
priori	O
to	O
belong	O
to	O
a	O
certain	O
parametric	O
family	O
of	O
pure	O
qubit	B-KEY
systems	I-KEY
,	O
we	O
derive	O
two	O
kinds	O
of	O
matching	B-KEY
strategies	I-KEY
.	O
The	O
first	O
one	O
is	O
a	O
semiclassical	B-KEY
strategy	I-KEY
that	O
is	O
obtained	O
by	O
the	O
natural	O
extension	O
of	O
conventional	O
matching	B-KEY
strategies	I-KEY
and	O
consists	O
of	O
a	O
two-stage	B-KEY
procedure	I-KEY
:	O
identification	O
-LRB-	O
estimation	O
-RRB-	O
of	O
the	O
unknown	O
template	O
states	O
to	O
design	O
the	O
classifier	O
-LRB-	O
learning	B-KEY
process	I-KEY
to	O
train	O
the	O
classifier	O
-RRB-	O
and	O
classification	O
of	O
the	O
input	O
system	O
into	O
the	O
appropriate	O
pattern	B-KEY
class	I-KEY
based	O
on	O
the	O
estimated	O
results	O
.	O
The	O
other	O
is	O
a	O
fully	O
quantum	B-KEY
strategy	I-KEY
without	O
any	O
intermediate	O
measurement	O
,	O
which	O
we	O
might	O
call	O
as	O
the	O
universal	B-KEY
quantum	I-KEY
matching	I-KEY
machine	I-KEY
.	O
We	O
present	O
the	O
Bayes	B-KEY
optimal	I-KEY
solutions	I-KEY
for	O
both	O
strategies	O
in	O
the	O
case	O
of	O
K	O
=	O
1	O
,	O
showing	O
that	O
there	O
certainly	O
exists	O
a	O
fully	O
quantum	B-KEY
matching	I-KEY
procedure	I-KEY
that	O
is	O
strictly	O
superior	O
to	O
the	O
straightforward	O
semiclassical	B-KEY
extension	I-KEY
of	O
the	O
conventional	O
matching	B-KEY
strategy	I-KEY
based	O
on	O
the	O
learning	B-KEY
process	I-KEY

Instability	B-KEY
phenomena	I-KEY
in	O
the	O
gas-metal	B-KEY
arc	I-KEY
welding	I-KEY
self-regulation	B-KEY
process	I-KEY
Arc	B-KEY
instability	I-KEY
is	O
a	O
very	O
important	O
determinant	O
of	O
weld	B-KEY
quality	I-KEY
.	O
The	O
instability	O
behaviour	O
of	O
the	O
gas-metal	B-KEY
arc	I-KEY
welding	I-KEY
-LRB-	O
GMAW	O
-RRB-	O
process	O
is	O
characterized	O
by	O
strong	O
oscillations	O
in	O
arc	O
length	O
and	O
current	O
.	O
In	O
the	O
paper	O
,	O
a	O
model	O
of	O
the	O
GMAW	B-KEY
process	I-KEY
is	O
developed	O
using	O
an	O
exact	B-KEY
arc	I-KEY
voltage	I-KEY
characteristic	I-KEY
.	O
This	O
model	O
is	O
used	O
to	O
study	O
stability	O
of	O
the	O
self-regulation	B-KEY
process	I-KEY
and	O
to	O
develop	O
a	O
simulation	O
program	O
that	O
helps	O
to	O
understand	O
the	O
transient	O
or	O
dynamic	O
nature	O
of	O
the	O
GMAW	B-KEY
process	I-KEY
and	O
relationships	O
among	O
current	O
,	O
electrode	O
extension	O
and	O
contact	O
tube-work	O
distance	O
.	O
The	O
process	O
is	O
shown	O
to	O
exhibit	O
instabilities	O
at	O
both	O
long	O
electrode	O
extension	O
and	O
normal	O
extension	O
.	O
Results	O
obtained	O
from	O
simulation	O
runs	O
of	O
the	O
model	O
were	O
also	O
experimentally	O
confirmed	O
by	O
the	O
present	O
author	O
,	O
as	O
reported	O
in	O
this	O
study	O
.	O
In	O
order	O
to	O
explain	O
the	O
concept	O
of	O
the	O
instability	B-KEY
phenomena	I-KEY
,	O
the	O
metal	B-KEY
transfer	I-KEY
mode	I-KEY
and	O
the	O
arc	O
voltage-current	O
characteristic	O
were	O
examined	O
.	O
Based	O
on	O
this	O
examination	O
,	O
the	O
conclusion	O
of	O
this	O
study	O
is	O
that	O
their	O
combined	O
effects	O
lead	O
to	O
the	O
oscillations	O
in	O
arc	O
current	O
and	O
length	O

Noise	O
and	O
the	O
PSTH	B-KEY
response	O
to	O
current	O
transients	O
:	O
II	O
.	O
Integrate-and-fire	B-KEY
model	I-KEY
with	O
slow	O
recovery	O
and	O
application	O
to	O
motoneuron	B-KEY
data	O
For	O
pt.I	O
see	O
ibid.	O
,	O
vol	O
.11	O
,	O
no.	O
2	O
,	O
p.135-151	O
-LRB-	O
2001	O
-RRB-	O
.	O
A	O
generalized	O
version	O
of	O
the	O
integrate-and-fire	B-KEY
model	I-KEY
is	O
presented	O
that	O
qualitatively	O
reproduces	O
firing	B-KEY
rates	I-KEY
and	O
membrane	B-KEY
trajectories	I-KEY
of	O
motoneurons	B-KEY
.	O
The	O
description	O
is	O
based	O
on	O
the	O
spike-response	B-KEY
model	I-KEY
and	O
includes	O
three	O
different	O
time	O
constants	O
:	O
the	O
passive	B-KEY
membrane	I-KEY
time	I-KEY
constant	I-KEY
,	O
a	O
recovery	B-KEY
time	I-KEY
of	O
the	O
input	O
conductance	O
after	O
each	O
spike	O
,	O
and	O
a	O
time	O
constant	O
of	O
the	O
spike	B-KEY
afterpotential	I-KEY
.	O
The	O
effect	O
of	O
stochastic	O
background	O
input	O
on	O
the	O
peristimulus	O
time	O
histogram	O
-LRB-	O
PSTH	B-KEY
-RRB-	O
response	O
to	O
spike	O
input	O
is	O
calculated	O
analytically	O
.	O
Model	O
results	O
are	O
compared	O
with	O
the	O
experimental	O
data	O
of	O
Poliakov	O
et	O
al.	O
-LRB-	O
1996	O
-RRB-	O
.	O
The	O
linearized	O
theory	O
shows	O
that	O
the	O
PSTH	B-KEY
response	O
to	O
an	O
input	O
spike	O
is	O
proportional	O
to	O
a	O
filtered	O
version	O
of	O
the	O
postsynaptic	O
potential	O
generated	O
by	O
the	O
input	O
spike	O
.	O
The	O
shape	O
of	O
the	O
filter	O
depends	O
on	O
the	O
background	O
activity	O
.	O
The	O
full	O
nonlinear	O
theory	O
is	O
in	O
close	O
agreement	O
with	O
simulated	O
PSTH	B-KEY
data	O

Deterministic	B-KEY
single-photon	I-KEY
source	I-KEY
for	O
distributed	B-KEY
quantum	I-KEY
networking	I-KEY
A	O
sequence	O
of	O
single	O
photons	O
is	O
emitted	O
on	O
demand	O
from	O
a	O
single	B-KEY
three-level	I-KEY
atom	I-KEY
strongly	O
coupled	O
to	O
a	O
high-finesse	B-KEY
optical	I-KEY
cavity	I-KEY
.	O
The	O
photons	O
are	O
generated	O
by	O
an	O
adiabatically	B-KEY
driven	I-KEY
stimulated	I-KEY
Raman	I-KEY
transition	I-KEY
between	O
two	O
atomic	O
ground	O
states	O
,	O
with	O
the	O
vacuum	B-KEY
field	I-KEY
of	O
the	O
cavity	O
stimulating	O
one	O
branch	O
of	O
the	O
transition	O
,	O
and	O
laser	O
pulses	O
deterministically	O
driving	O
the	O
other	O
branch	O
.	O
This	O
process	O
is	O
unitary	O
and	O
therefore	O
intrinsically	O
reversible	O
,	O
which	O
is	O
essential	O
for	O
quantum	B-KEY
communication	I-KEY
and	O
networking	O
,	O
and	O
the	O
photons	O
should	O
be	O
appropriate	O
for	O
all-optical	B-KEY
quantum	I-KEY
information	I-KEY
processing	I-KEY

Semidefinite	O
programming	O
vs.	O
LP	B-KEY
relaxations	I-KEY
for	O
polynomial	B-KEY
programming	I-KEY
We	O
consider	O
the	O
global	B-KEY
minimization	I-KEY
of	O
a	O
multivariate	B-KEY
polynomial	I-KEY
on	O
a	O
semi-algebraic	B-KEY
set	I-KEY
Omega	O
defined	O
with	O
polynomial	B-KEY
inequalities	I-KEY
.	O
We	O
then	O
compare	O
two	O
hierarchies	O
of	O
relaxations	O
,	O
namely	O
,	O
LP	B-KEY
relaxations	I-KEY
based	O
on	O
products	O
of	O
the	O
original	O
constraints	O
,	O
in	O
the	O
spirit	O
of	O
the	O
RLT	B-KEY
procedure	I-KEY
of	O
Sherali	O
and	O
Adams	O
-LRB-	O
1990	O
-RRB-	O
,	O
and	O
recent	O
semidefinite	O
programming	O
-LRB-	O
SDP	O
-RRB-	O
relaxations	O
introduced	O
by	O
the	O
author	O
.	O
The	O
comparison	O
is	O
analyzed	O
in	O
light	O
of	O
recent	O
results	O
in	O
real	B-KEY
algebraic	I-KEY
geometry	I-KEY
on	O
various	O
representations	O
of	O
polynomials	O
,	O
positive	O
on	O
a	O
compact	O
semi-algebraic	B-KEY
set	I-KEY

Voltage	B-KEY
control	I-KEY
methods	O
with	O
grid	B-KEY
connected	I-KEY
wind	I-KEY
turbines	I-KEY
:	O
a	O
tutorial	O
review	O
Within	O
electricity	B-KEY
grid	I-KEY
networks	I-KEY
it	O
is	O
conventional	O
for	O
large-scale	B-KEY
central	I-KEY
generators	I-KEY
to	O
both	O
provide	O
power	O
and	O
control	O
grid	O
node	O
voltage	O
.	O
Therefore	O
when	O
wind	O
turbines	O
replace	O
conventional	O
power	O
stations	O
on	O
a	O
substantial	O
scale	O
,	O
they	O
must	O
not	O
only	O
generate	O
power	O
,	O
but	O
also	O
control	O
grid	O
node	O
voltages	O
.	O
This	O
paper	O
reviews	O
the	O
basic	O
principles	O
of	O
voltage	B-KEY
control	I-KEY
for	O
tutorial	O
benefit	O
and	O
then	O
considers	O
application	O
of	O
grid-connected	O
wind	O
turbines	O
for	O
voltage	B-KEY
control	I-KEY
.	O
The	O
most	O
widely	O
used	O
contemporary	O
wind	O
turbine	O
types	O
are	O
considered	O
and	O
further	O
detail	O
is	O
given	O
for	O
determining	O
the	O
range	O
of	O
variables	O
that	O
allow	O
control	O

Robust	O
output-feedback	B-KEY
control	I-KEY
for	O
linear	O
continuous	O
uncertain	O
state	B-KEY
delayed	I-KEY
systems	I-KEY
with	O
unknown	O
time	B-KEY
delay	I-KEY
The	O
state-delayed	O
time	O
often	O
is	O
unknown	O
and	O
independent	O
of	O
other	O
variables	O
in	O
most	O
real	O
physical	O
systems	O
.	O
A	O
new	O
stability	O
criterion	O
for	O
uncertain	B-KEY
systems	I-KEY
with	O
a	O
state	B-KEY
time-varying	I-KEY
delay	I-KEY
is	O
proposed	O
.	O
Then	O
,	O
a	O
robust	O
observer-based	B-KEY
control	I-KEY
law	I-KEY
based	O
on	O
this	O
criterion	O
is	O
constructed	O
via	O
the	O
sequential	B-KEY
quadratic	I-KEY
programming	I-KEY
method	O
.	O
We	O
also	O
develop	O
a	O
separation	O
property	O
so	O
that	O
the	O
state	B-KEY
feedback	I-KEY
control	I-KEY
law	I-KEY
and	O
observer	O
can	O
be	O
independently	O
designed	O
and	O
maintain	O
closed-loop	B-KEY
system	I-KEY
stability	I-KEY
.	O
An	O
example	O
illustrates	O
the	O
availability	O
of	O
the	O
proposed	O
design	O
method	O

Supersampling	O
multiframe	B-KEY
blind	I-KEY
deconvolution	I-KEY
resolution	O
enhancement	O
of	O
adaptive	O
optics	O
compensated	O
imagery	O
of	O
low	O
earth	O
orbit	O
satellites	O
We	O
describe	O
a	O
postprocessing	B-KEY
methodology	I-KEY
for	O
reconstructing	O
undersampled	O
image	O
sequences	O
with	O
randomly	B-KEY
varying	I-KEY
blur	I-KEY
that	O
can	O
provide	O
image	B-KEY
enhancement	I-KEY
beyond	O
the	O
sampling	O
resolution	O
of	O
the	O
sensor	O
.	O
This	O
method	O
is	O
demonstrated	O
on	O
simulated	B-KEY
imagery	I-KEY
and	O
on	O
adaptive-optics	O
-	O
-LRB-	O
AO	O
-RRB-	O
-	O
compensated	O
imagery	O
taken	O
by	O
the	O
Starfire	O
Optical	O
Range	O
3.5-m	O
telescope	O
that	O
has	O
been	O
artificially	O
undersampled	O
.	O
Also	O
shown	O
are	O
the	O
results	O
of	O
multiframe	B-KEY
blind	I-KEY
deconvolution	I-KEY
of	O
some	O
of	O
the	O
highest	O
quality	O
optical	O
imagery	O
of	O
low	B-KEY
earth	I-KEY
orbit	I-KEY
satellites	I-KEY
collected	O
with	O
a	O
ground-based	B-KEY
telescope	I-KEY
to	O
date	O
.	O
The	O
algorithm	O
used	O
is	O
a	O
generalization	O
of	O
multiframe	B-KEY
blind	I-KEY
deconvolution	I-KEY
techniques	O
that	O
include	O
a	O
representation	O
of	O
spatial	B-KEY
sampling	I-KEY
by	O
the	O
focal	B-KEY
plane	I-KEY
array	I-KEY
elements	I-KEY
based	O
on	O
a	O
forward	B-KEY
stochastic	I-KEY
model	I-KEY
.	O
This	O
generalization	O
enables	O
the	O
random	B-KEY
shifts	I-KEY
and	O
shape	O
of	O
the	O
AO-compensated	B-KEY
point	I-KEY
spread	I-KEY
function	I-KEY
-LRB-	O
PSF	O
-RRB-	O
to	O
be	O
used	O
to	O
partially	O
eliminate	O
the	O
aliasing	B-KEY
effects	I-KEY
associated	O
with	O
sub-Nyquist	B-KEY
sampling	I-KEY
of	O
the	O
image	O
by	O
the	O
focal	O
plane	O
array	O
.	O
The	O
method	O
could	O
be	O
used	O
to	O
reduce	O
resolution	B-KEY
loss	I-KEY
that	O
occurs	O
when	O
imaging	O
in	O
wide-field-of-view	O
-LRB-	O
FOV	O
-RRB-	O
modes	O

Attribute	B-KEY
generation	I-KEY
based	O
on	O
association	B-KEY
rules	I-KEY
A	O
decision	B-KEY
tree	I-KEY
is	O
considered	O
to	O
be	O
appropriate	O
-LRB-	O
1	O
-RRB-	O
if	O
the	O
tree	O
can	O
classify	O
the	O
unseen	O
data	O
accurately	O
,	O
and	O
-LRB-	O
2	O
-RRB-	O
if	O
the	O
size	O
of	O
the	O
tree	O
is	O
small	O
.	O
One	O
of	O
the	O
approaches	O
to	O
induce	O
such	O
a	O
good	O
decision	B-KEY
tree	I-KEY
is	O
to	O
add	O
new	O
attributes	O
and	O
their	O
values	O
to	O
enhance	O
the	O
expressiveness	O
of	O
the	O
training	B-KEY
data	I-KEY
at	O
the	O
data	O
pre-processing	O
stage	O
.	O
There	O
are	O
many	O
existing	O
methods	O
for	O
attribute	B-KEY
extraction	I-KEY
and	O
construction	O
,	O
but	O
constructing	O
new	O
attributes	O
is	O
still	O
an	O
art	O
.	O
These	O
methods	O
are	O
very	O
time	O
consuming	O
,	O
and	O
some	O
of	O
them	O
need	O
a	O
priori	O
knowledge	O
of	O
the	O
data	O
domain	O
.	O
They	O
are	O
not	O
suitable	O
for	O
data	B-KEY
mining	I-KEY
dealing	O
with	O
large	O
volumes	O
of	O
data	O
.	O
We	O
propose	O
a	O
novel	O
approach	O
that	O
the	O
knowledge	O
on	O
attributes	O
relevant	O
to	O
the	O
class	O
is	O
extracted	O
as	O
association	B-KEY
rules	I-KEY
from	O
the	O
training	B-KEY
data	I-KEY
.	O
The	O
new	O
attributes	O
and	O
the	O
values	O
are	O
generated	O
from	O
the	O
association	B-KEY
rules	I-KEY
among	O
the	O
originally	O
given	O
attributes	O
.	O
We	O
elaborate	O
on	O
the	O
method	O
and	O
investigate	O
its	O
feature	O
.	O
The	O
effectiveness	O
of	O
our	O
approach	O
is	O
demonstrated	O
through	O
some	O
experiments	B-KEY

Virtual	B-KEY
Development	I-KEY
Center	I-KEY
The	O
Virtual	B-KEY
Development	I-KEY
Center	I-KEY
of	O
the	O
Institute	O
for	O
Women	B-KEY
and	O
Technology	O
seeks	O
to	O
significantly	O
enhance	O
the	O
impact	O
of	O
women	B-KEY
on	O
technology	O
.	O
It	O
addresses	O
this	O
goal	O
by	O
increasing	O
the	O
number	O
of	O
women	B-KEY
who	O
have	O
input	O
on	O
created	O
technology	O
,	O
enhancing	O
the	O
ways	O
people	O
teach	B-KEY
and	O
develop	O
technology	O
,	O
and	O
developing	O
need-based	O
technology	O
that	O
serves	O
the	O
community	O
.	O
Through	O
activities	O
of	O
the	O
Virtual	B-KEY
Development	I-KEY
Center	I-KEY
,	O
a	O
pattern	O
is	O
emerging	O
regarding	O
how	O
computing	O
technologies	O
do	O
or	O
do	O
not	O
satisfy	O
the	O
needs	O
of	O
community	B-KEY
groups	I-KEY
,	O
particularly	O
those	O
communities	O
serving	O
women	B-KEY
.	O
This	O
paper	O
describes	O
the	O
Virtual	B-KEY
Development	I-KEY
Center	I-KEY
program	O
and	O
offers	O
observations	O
on	O
the	O
impact	O
of	O
computing	O
technology	O
on	O
non-technical	O
communities	O

Recognizing	O
groups	O
G/sub	O
2	O
/	O
-LRB-	O
3/sup	O
n	O
/	O
-RRB-	O
by	O
their	O
element	B-KEY
orders	I-KEY
It	O
is	O
proved	O
that	O
a	O
finite	B-KEY
group	I-KEY
that	O
is	O
isomorphic	B-KEY
to	O
a	O
simple	O
non-Abelian	O
group	O
G	O
=	O
G/sub	O
2	O
/	O
-LRB-	O
3/sup	O
n	O
/	O
-RRB-	O
is	O
,	O
up	O
to	O
isomorphism	B-KEY
,	O
recognized	O
by	O
a	O
set	O
omega	O
-LRB-	O
G	O
-RRB-	O
of	O
its	O
element	B-KEY
orders	I-KEY
,	O
that	O
is	O
,	O
H	O
approximately	O
=	O
G	O
if	O
omega	O
-LRB-	O
H	O
-RRB-	O
=	O
omega	O
-LRB-	O
G	O
-RRB-	O
for	O
some	O
finite	B-KEY
group	I-KEY
H	O

Calibrated	B-KEY
initials	I-KEY
for	O
an	O
EM	B-KEY
applied	O
to	O
recursive	B-KEY
models	I-KEY
of	O
categorical	B-KEY
variables	I-KEY
The	O
estimates	O
from	O
an	O
EM	B-KEY
,	O
when	O
it	O
is	O
applied	O
to	O
a	O
large	B-KEY
causal	I-KEY
model	I-KEY
of	O
10	O
or	O
more	O
categorical	B-KEY
variables	I-KEY
,	O
are	O
often	O
subject	O
to	O
the	O
initial	B-KEY
values	I-KEY
for	O
the	O
estimates	O
.	O
This	O
phenomenon	O
becomes	O
more	O
serious	O
as	O
the	O
model	O
structure	O
becomes	O
more	O
complicated	O
involving	O
more	O
variables	O
.	O
As	O
a	O
measure	O
of	O
compensation	O
for	O
this	O
,	O
it	O
has	O
been	O
recommended	O
in	O
literature	O
that	O
EMs	B-KEY
are	O
implemented	O
several	O
times	O
with	O
different	O
sets	O
of	O
initial	B-KEY
values	I-KEY
to	O
obtain	O
more	O
appropriate	O
estimates	O
.	O
We	O
propose	O
an	O
improved	O
approach	O
for	O
initial	B-KEY
values	I-KEY
.	O
The	O
main	O
idea	O
is	O
that	O
we	O
use	O
initials	O
that	O
are	O
calibrated	O
to	O
data	O
.	O
A	O
simulation	B-KEY
result	O
strongly	O
indicates	O
that	O
the	O
calibrated	B-KEY
initials	I-KEY
give	O
rise	O
to	O
the	O
estimates	O
that	O
are	O
far	O
closer	O
to	O
the	O
true	O
values	O
than	O
the	O
initials	O
that	O
are	O
not	O
calibrated	O

The	O
network	B-KEY
society	I-KEY
as	O
seen	O
from	O
Italy	B-KEY
Italy	B-KEY
was	O
behind	O
the	O
European	B-KEY
average	I-KEY
in	O
Internet	B-KEY
development	I-KEY
for	O
many	O
years	O
,	O
but	O
a	O
new	O
trend	O
,	O
which	O
has	O
brought	O
considerable	O
change	O
,	O
emerged	O
at	O
the	O
end	O
of	O
1998	O
and	O
showed	O
its	O
effects	O
in	O
2000	O
and	O
the	O
following	O
years	O
.	O
Now	O
Italy	B-KEY
is	O
one	O
of	O
the	O
top	O
ten	O
countries	O
worldwide	O
in	O
Internet	B-KEY
hostcount	I-KEY
and	O
the	O
fourth	O
largest	O
in	O
Europe	B-KEY
.	O
The	O
density	O
of	O
Internet	B-KEY
activity	I-KEY
in	O
Italy	B-KEY
in	O
proportion	O
to	O
the	O
population	O
is	O
still	O
below	O
the	O
average	O
in	O
the	O
European	B-KEY
Union	I-KEY
,	O
but	O
is	O
growing	O
faster	O
than	O
Germany	B-KEY
,	O
the	O
UK	B-KEY
and	O
France	B-KEY
,	O
and	O
faster	O
than	O
the	O
worldwide	O
or	O
European	B-KEY
average	I-KEY
.	O
From	O
the	O
point	O
of	O
view	O
of	O
media	B-KEY
control	I-KEY
there	O
are	O
several	O
problems	O
.	O
Italy	B-KEY
has	O
democratic	B-KEY
institutions	I-KEY
and	O
freedom	B-KEY
of	I-KEY
speech	I-KEY
,	O
but	O
there	O
is	O
an	O
alarming	O
concentration	O
in	O
the	O
control	O
of	O
mainstream	B-KEY
media	I-KEY
-LRB-	O
especially	O
broadcast	O
-RRB-	O
.	O
There	O
are	O
no	O
officially	O
declared	O
restrictions	O
in	O
the	O
use	O
of	O
the	O
Internet	O
,	O
but	O
several	O
legal	O
and	O
regulatory	B-KEY
decisions	I-KEY
reveal	O
a	O
desire	O
to	O
limit	O
freedom	O
of	O
opinion	O
and	O
dialogue	O
and/or	O
gain	O
centralized	B-KEY
control	I-KEY
of	O
the	O
Net	O

Transcripts	B-KEY
:	O
bane	O
or	O
boon	O
?	O
-LSB-	O
law	B-KEY
reporting	I-KEY
-RSB-	O
Because	O
judge-made	B-KEY
law	I-KEY
,	O
by	O
its	O
very	O
nature	O
,	O
is	O
less	O
immediately	O
accessible	O
than	O
the	O
law	O
of	O
codified	O
,	O
statutory	O
systems	O
,	O
it	O
calls	O
for	O
an	O
efficient	O
system	O
of	O
law	B-KEY
reporting	I-KEY
.	O
Of	O
necessity	O
,	O
any	O
such	O
system	O
will	O
be	O
selective	O
,	O
the	O
majority	O
of	O
decisions	O
going	O
unreported	O
.	O
Considerable	O
power	O
thereby	O
comes	O
to	O
repose	O
in	O
the	O
hands	O
of	O
the	O
law	B-KEY
reporters	I-KEY
.	O
The	O
author	O
shares	O
his	O
invaluable	O
perception	O
and	O
extensive	O
research	O
on	O
the	O
difficulties	O
which	O
arise	O
from	O
the	O
excess	O
of	O
access	O
to	O
judgments	B-KEY

Mathematical	B-KEY
models	I-KEY
of	O
functioning	O
of	O
an	O
insurance	O
company	O
with	O
allowance	O
for	O
the	O
rate	O
of	O
return	O
Models	O
of	O
the	O
functioning	O
of	O
insurance	O
companies	O
are	O
suggested	O
,	O
when	O
the	O
free	B-KEY
capital	I-KEY
increases	I-KEY
from	O
interest	B-KEY
at	O
a	O
certain	O
rate	O
.	O
The	O
basic	O
characteristics	O
of	O
the	O
capital	O
of	O
a	O
company	O
are	O
studied	O
in	O
the	O
stationary	B-KEY
regime	I-KEY

Nonlinear	B-KEY
control	I-KEY
of	O
a	O
shape	B-KEY
memory	I-KEY
alloy	I-KEY
actuated	O
manipulator	B-KEY
This	O
paper	O
presents	O
a	O
nonlinear	O
,	O
robust	O
control	O
algorithm	O
for	O
accurate	O
positioning	B-KEY
of	O
a	O
single	O
degree	O
of	O
freedom	O
rotary	O
manipulator	B-KEY
actuated	O
by	O
Shape	B-KEY
Memory	I-KEY
Alloy	I-KEY
-LRB-	O
SMA	O
-RRB-	O
.	O
A	O
model	O
for	O
an	O
SMA	O
actuated	O
manipulator	B-KEY
is	O
presented	O
.	O
The	O
model	O
includes	O
nonlinear	B-KEY
dynamics	I-KEY
of	O
the	O
manipulator	B-KEY
,	O
a	O
constitutive	O
model	O
of	O
Shape	B-KEY
Memory	I-KEY
Alloy	I-KEY
,	O
and	O
electrical	O
and	O
heat	O
transfer	O
behavior	O
of	O
SMA	O
wire	O
.	O
This	O
model	O
is	O
used	O
for	O
open	O
and	O
closed	B-KEY
loop	I-KEY
motion	O
simulations	O
of	O
the	O
manipulator	B-KEY
.	O
Experiments	O
are	O
presented	O
that	O
show	O
results	O
similar	O
to	O
both	O
closed	O
and	O
open	B-KEY
loop	I-KEY
simulation	O
results	O
.	O
Due	O
to	O
modeling	O
uncertainty	O
and	O
nonlinear	O
behavior	O
of	O
the	O
system	O
,	O
classic	O
control	O
methods	O
such	O
as	O
Proportional-Integral-Derivative	O
control	O
are	O
not	O
able	O
to	O
present	O
fast	O
and	O
accurate	O
performance	O
.	O
Hence	O
a	O
nonlinear	O
,	O
robust	O
control	O
algorithm	O
is	O
presented	O
based	O
on	O
Variable	B-KEY
Structure	I-KEY
Control	I-KEY
.	O
This	O
algorithm	O
is	O
a	O
control	B-KEY
gain	I-KEY
switching	I-KEY
technique	O
based	O
on	O
the	O
weighted	O
average	O
of	O
position	B-KEY
and	O
velocity	O
feedbacks	B-KEY
.	O
This	O
method	O
has	O
been	O
designed	O
through	O
simulation	O
and	O
tested	O
experimentally	O
.	O
Results	O
show	O
fast	O
,	O
accurate	O
,	O
and	O
robust	O
performance	O
of	O
the	O
control	O
system	O
.	O
Computer	O
simulation	O
and	O
experimental	O
results	O
for	O
different	O
stabilization	B-KEY
and	O
tracking	B-KEY
situations	O
are	O
also	O
presented	O

Design	O
patterns	O
for	O
high	O
availability	O
It	O
is	O
possible	O
to	O
achieve	O
five-nines	O
reliability	O
with	O
everyday	O
commercial-quality	O
hardware	O
and	O
software	O
.	O
The	O
key	O
is	O
the	O
way	O
in	O
which	O
these	O
components	O
are	O
combined	O
.	O
The	O
design	O
of	O
high	B-KEY
availability	I-KEY
systems	I-KEY
is	O
based	O
on	O
a	O
combination	O
of	O
redundant	B-KEY
hardware	I-KEY
components	I-KEY
and	O
software	O
to	O
manage	O
fault	B-KEY
detection	I-KEY
and	O
correction	O
without	O
human	O
intervention	O
.	O
The	O
author	O
quickly	O
reviews	O
some	O
definitions	O
tied	O
to	O
high	O
availability	O
and	O
fault	O
management	O
,	O
and	O
then	O
goes	O
on	O
to	O
discuss	O
some	O
hardware	O
and	O
software	B-KEY
design	I-KEY
patterns	I-KEY
for	O
fault	B-KEY
tolerant	I-KEY
systems	I-KEY

A	O
comprehensive	O
chatter	B-KEY
prediction	I-KEY
model	I-KEY
for	O
face	B-KEY
turning	I-KEY
operation	I-KEY
including	O
tool	B-KEY
wear	I-KEY
effect	I-KEY
Presents	O
a	O
three-dimensional	B-KEY
mechanistic	I-KEY
frequency	I-KEY
domain	I-KEY
chatter	I-KEY
model	I-KEY
for	O
face	O
turning	O
processes	O
,	O
that	O
can	O
account	O
for	O
the	O
effects	O
of	O
tool	O
wear	O
including	O
process	B-KEY
damping	I-KEY
.	O
New	O
formulations	O
are	O
presented	O
to	O
model	O
the	O
variation	O
in	O
process	B-KEY
damping	I-KEY
forces	O
along	O
nonlinear	O
tool	O
geometries	O
such	O
as	O
the	O
nose	O
radius	O
.	O
The	O
underlying	O
dynamic	O
force	O
model	O
simulates	O
the	O
variation	O
in	O
the	O
chip	O
cross-sectional	O
area	O
by	O
accounting	O
for	O
the	O
displacements	O
in	O
the	O
axial	O
and	O
radial	B-KEY
directions	I-KEY
.	O
The	O
model	O
can	O
be	O
used	O
to	O
determine	O
stability	B-KEY
boundaries	I-KEY
under	O
various	O
cutting	O
conditions	O
and	O
different	O
states	O
of	O
flank	B-KEY
wear	I-KEY
.	O
Experimental	O
results	O
for	O
different	O
amounts	O
of	O
wear	O
are	O
provided	O
as	O
a	O
validation	O
for	O
the	O
model	O

Gender	O
,	O
software	B-KEY
design	I-KEY
,	O
and	O
occupational	B-KEY
equity	I-KEY
After	O
reviewing	O
the	O
work	O
on	O
gender	B-KEY
bias	I-KEY
in	O
software	B-KEY
design	I-KEY
,	O
a	O
model	O
of	O
gender-role	O
influenced	O
achievement	O
choice	O
taken	O
from	O
Eccles	O
-LRB-	O
1994	O
-RRB-	O
is	O
presented	O
.	O
The	O
paper	O
concludes	O
that	O
-LRB-	O
1	O
-RRB-	O
though	O
laudable	O
,	O
reduction	O
of	O
gender	B-KEY
bias	I-KEY
in	O
software	B-KEY
design	I-KEY
is	O
not	O
the	O
most	O
straightforward	O
way	O
to	O
reduce	O
gender	O
inequity	O
in	O
the	O
choice	O
of	O
computing	O
as	O
a	O
career	O
,	O
-LRB-	O
2	O
-RRB-	O
the	O
model	O
itself	O
makes	O
more	O
clear	O
some	O
of	O
the	O
ethical	B-KEY
issues	I-KEY
involved	O
in	O
attempting	O
to	O
achieve	O
gender	O
equity	O
on	O
computing	O
,	O
and	O
-LRB-	O
3	O
-RRB-	O
efforts	O
to	O
reduce	O
gender	O
inequity	O
in	O
the	O
choice	O
of	O
computing	O
as	O
a	O
career	O
need	O
to	O
be	O
evaluated	O
in	O
the	O
light	O
of	O
this	O
model	O

Approximation	O
and	O
complexity	O
.	O
II	O
.	O
Iterated	O
integration	B-KEY
For	O
pt	O
.	O
I.	O
see	O
ibid.	O
,	O
no.	O
1	O
,	O
p.	O
289-95	O
-LRB-	O
2001	O
-RRB-	O
.	O
We	O
introduce	O
two	O
classes	O
of	O
real	B-KEY
analytic	I-KEY
functions	I-KEY
W	O
contained	O
in/implied	O
by	O
U	O
on	O
an	O
interval	O
.	O
Starting	O
with	O
rational	B-KEY
functions	I-KEY
to	O
construct	O
functions	O
in	O
W	O
we	O
allow	O
the	O
application	O
of	O
three	O
types	O
of	O
operations	O
:	O
addition	B-KEY
,	O
integration	B-KEY
,	O
and	O
multiplication	B-KEY
by	O
a	O
polynomial	B-KEY
with	O
rational	O
coefficients	O
.	O
In	O
a	O
similar	O
way	O
,	O
to	O
construct	O
functions	O
in	O
U	O
we	O
allow	O
integration	B-KEY
,	O
addition	B-KEY
,	O
and	O
multiplication	B-KEY
of	O
functions	O
already	O
constructed	O
in	O
U	O
and	O
multiplication	B-KEY
by	O
rational	O
numbers	O
.	O
Thus	O
,	O
U	O
is	O
a	O
subring	O
of	O
the	O
ring	O
of	O
Pfaffian	B-KEY
functions	I-KEY
.	O
Two	O
lower	B-KEY
bounds	I-KEY
on	O
the	O
L/sub	O
infinity	O
/	O
-	O
norm	O
are	O
proved	O
on	O
a	O
function	O
f	O
from	O
W	O
-LRB-	O
or	O
from	O
U	O
,	O
respectively	O
-RRB-	O
in	O
terms	O
of	O
the	O
complexity	O
of	O
constructing	O
f	O

Abacus	O
,	O
EFI	O
and	O
anti-virus	B-KEY
The	O
Extensible	O
Firmware	O
Interface	O
-LRB-	O
EFI	O
-RRB-	O
standard	O
emerged	O
as	O
a	O
logical	O
step	O
to	O
provide	O
flexibility	O
and	O
extensibility	O
to	O
boot	O
sequence	O
processes	O
,	O
enabling	O
the	O
complete	O
abstraction	O
of	O
a	O
system	O
's	O
BIOS	O
interface	O
from	O
the	O
system	O
's	O
hardware	O
.	O
In	O
doing	O
so	O
,	O
this	O
provided	O
the	O
means	O
of	O
standardizing	O
a	O
boot-up	O
sequence	O
,	O
extending	O
device	O
drivers	O
and	O
boot	O
time	O
applications	O
'	O
portability	O
to	O
non	O
PC-AT-based	O
architectures	O
,	O
including	O
embedded	B-KEY
systems	I-KEY
like	O
Internet	O
appliances	O
,	O
TV	O
Internet	O
set-top	O
boxes	O
and	O
64-bit	O
Itanium	O
platforms	O

Cataloguing	B-KEY
to	O
help	O
law	B-KEY
library	I-KEY
users	I-KEY
The	O
author	O
takes	O
a	O
broader	O
view	O
of	O
the	O
catalogue	B-KEY
than	O
is	O
usual	O
;	O
we	O
can	O
include	O
within	O
it	O
items	O
that	O
have	O
locations	O
other	O
than	O
the	O
office/library	O
itself	O
.	O
This	O
may	O
well	O
start	O
with	O
Internet	B-KEY
resources	I-KEY
,	O
but	O
can	O
perfectly	O
appropriately	O
continue	O
with	O
standard	O
works	O
not	O
held	O
in	O
the	O
immediate	O
collection	O
but	O
available	O
in	O
some	O
other	O
accessible	O
collection	O
,	O
such	O
as	O
the	O
local	O
reference	B-KEY
library	I-KEY
.	O
The	O
essential	O
feature	O
is	O
to	O
include	O
entries	O
for	O
the	O
kind	O
of	O
material	O
sought	O
by	O
users	O
,	O
with	O
the	O
addition	O
of	O
a	O
location	B-KEY
mark	I-KEY
indicating	O
where	O
they	O
can	O
find	O
it	O

A	O
Blog	O
in	O
every	O
law	B-KEY
firm	I-KEY
?	O
You	O
do	O
n't	O
know	O
today	O
what	O
you	O
'll	O
want	O
to	O
know	O
next	O
year	O
.	O
Rather	O
than	O
trying	O
to	O
solve	O
that	O
problem	O
,	O
focus	O
on	O
providing	O
simple	O
tools	O
to	O
users	O
that	O
create	O
valuable	O
content	O
across	O
the	O
firm	O
.	O
Individual	O
contributions	O
will	O
be	O
more	O
visible	O
,	O
and	O
you	O
will	O
have	O
a	O
searchable	O
archive	O
of	O
your	O
institutional	B-KEY
memory	I-KEY
and	O
a	O
simplified	O
process	O
for	O
ensuring	O
everyone	O
is	O
up	O
to	O
speed	O
.	O
Whether	O
you	O
embrace	O
weblogs	B-KEY
for	O
their	O
individual	O
or	O
institutional	O
benefits	O
,	O
one	O
thing	O
is	O
certain	O
:	O
They	O
will	O
become	O
powerful	O
tools	O
for	O
those	O
who	O
seek	O
ways	O
to	O
more	O
efficiently	O
and	O
intelligently	O
manage	O
information	O

eLeaders	O
make	O
the	O
Web	O
work	O
Some	O
companies	O
are	O
making	O
the	O
most	O
of	O
back-office/Web	B-KEY
integration	I-KEY
.	O
Here	O
are	O
some	O
winners	O

Data	B-KEY
extraction	I-KEY
from	O
the	O
Web	O
based	O
on	O
pre-defined	O
schema	B-KEY
With	O
the	O
development	O
of	O
the	O
Internet	B-KEY
,	O
the	O
World	O
Wide	O
Web	O
has	O
become	O
an	O
invaluable	O
information	B-KEY
source	I-KEY
for	O
most	O
organizations	O
.	O
However	O
,	O
most	O
documents	O
available	O
from	O
the	O
Web	O
are	O
in	O
HTML	B-KEY
form	O
which	O
is	O
originally	O
designed	O
for	O
document	O
formatting	O
with	O
little	O
consideration	O
of	O
its	O
contents	O
.	O
Effectively	O
extracting	O
data	O
from	O
such	O
documents	O
remains	O
a	O
nontrivial	O
task	O
.	O
In	O
this	O
paper	O
,	O
we	O
present	O
a	O
schema-guided	O
approach	O
to	O
extracting	O
data	O
from	O
HTML	O
pages	O
.	O
Under	O
the	O
approach	O
,	O
the	O
user	O
defines	O
a	O
schema	B-KEY
specifying	O
what	O
to	O
be	O
extracted	O
and	O
provides	O
sample	O
mappings	O
between	O
the	O
schema	B-KEY
and	O
the	O
HTML	B-KEY
page	O
.	O
The	O
system	O
will	O
induce	O
the	O
mapping	O
rules	O
and	O
generate	O
a	O
wrapper	O
that	O
takes	O
the	O
HTML	B-KEY
page	O
as	O
input	O
and	O
produces	O
the	O
required	O
data	O
in	O
the	O
form	O
of	O
XML	O
conforming	O
to	O
the	O
user-defined	O
schema	B-KEY
.	O
A	O
prototype	O
system	O
implementing	O
the	O
approach	O
has	O
been	O
developed	O
.	O
The	O
preliminary	O
experiments	O
indicate	O
that	O
the	O
proposed	O
semi-automatic	O
approach	O
is	O
not	O
only	O
easy	O
to	O
use	O
but	O
also	O
able	O
to	O
produce	O
a	O
wrapper	O
that	O
extracts	O
required	O
data	O
from	O
inputted	O
pages	O
with	O
high	O
accuracy	O

Modification	O
for	O
synchronization	B-KEY
of	O
Rossler	O
and	O
Chen	B-KEY
chaotic	I-KEY
systems	I-KEY
Active	B-KEY
control	I-KEY
is	O
an	O
effective	O
method	O
for	O
making	O
two	O
identical	O
Rossler	O
and	O
Chen	O
systems	O
be	O
synchronized	B-KEY
.	O
However	O
,	O
this	O
method	O
works	O
only	O
for	O
a	O
certain	O
class	O
of	O
chaotic	O
systems	O
with	O
known	O
parameters	O
both	O
in	O
drive	O
systems	O
and	O
response	B-KEY
systems	I-KEY
.	O
Modification	O
based	O
on	O
Lyapunov	B-KEY
stability	I-KEY
theory	I-KEY
is	O
proposed	O
in	O
order	O
to	O
overcome	O
this	O
limitation	O
.	O
An	O
adaptive	O
synchronization	B-KEY
controller	O
,	O
which	O
can	O
make	O
the	O
states	O
of	O
two	O
identical	O
Rossler	O
and	O
Chen	O
systems	O
globally	O
asymptotically	O
synchronized	O
in	O
the	O
presence	O
of	O
system	O
's	O
unknown	O
constant	O
parameters	O
,	O
is	O
derived	O
.	O
Especially	O
,	O
when	O
some	O
unknown	O
parameters	O
are	O
positive	O
,	O
we	O
can	O
make	O
the	O
controller	O
more	O
simple	O
,	O
besides	O
,	O
the	O
controller	O
is	O
independent	O
of	O
those	O
positive	O
uncertain	O
parameters	O
.	O
At	O
last	O
,	O
when	O
the	O
condition	O
that	O
arbitrary	O
unknown	O
parameters	O
in	O
two	O
systems	O
are	O
identical	O
constants	O
is	O
cancelled	O
,	O
we	O
demonstrate	O
that	O
it	O
is	O
possible	O
to	O
synchronize	B-KEY
two	O
chaotic	O
systems	O
.	O
All	O
results	O
are	O
proved	O
using	O
a	O
well-known	O
Lyapunov	O
stability	O
theorem	O
.	O
Numerical	O
simulations	O
are	O
given	O
to	O
validate	O
the	O
proposed	O
synchronization	B-KEY
approach	O

Virtual-reality-based	O
multidimensional	B-KEY
therapy	I-KEY
for	O
the	O
treatment	O
of	O
body	B-KEY
image	I-KEY
disturbances	I-KEY
in	O
binge	B-KEY
eating	I-KEY
disorders	I-KEY
:	O
a	O
preliminary	O
controlled	O
study	O
The	O
main	O
goal	O
of	O
this	O
paper	O
is	O
to	O
preliminarily	O
evaluate	O
the	O
efficacy	O
of	O
a	O
virtual-reality	O
-LRB-	O
VR	O
-RRB-	O
-	O
based	O
multidimensional	O
approach	O
in	O
the	O
treatment	O
of	O
body	O
image	O
attitudes	O
and	O
related	O
constructs	O
.	O
The	O
female	O
binge	B-KEY
eating	I-KEY
disorder	I-KEY
-LRB-	O
BED	O
-RRB-	O
patients	O
-LRB-	O
n	O
=	O
20	O
-RRB-	O
,	O
involved	O
in	O
a	O
residential	B-KEY
weight	I-KEY
control	I-KEY
treatment	I-KEY
including	O
low-calorie	O
diet	O
-LRB-	O
1200	O
cal/day	O
-RRB-	O
and	O
physical	O
training	O
,	O
were	O
randomly	O
assigned	O
either	O
to	O
the	O
multidimensional	O
VR	O
treatment	O
or	O
to	O
psychonutritional	B-KEY
groups	I-KEY
based	O
on	O
the	O
cognitive-behavior	B-KEY
approach	I-KEY
.	O
Patients	O
were	O
administered	O
a	O
battery	O
of	O
outcome	O
measures	O
assessing	O
eating	O
disorders	O
symptomathology	O
,	O
attitudes	O
toward	O
food	O
,	O
body	O
dissatisfaction	O
,	O
level	O
of	O
anxiety	B-KEY
,	O
motivation	O
for	O
change	O
,	O
level	O
of	O
assertiveness	O
,	O
and	O
general	O
psychiatric	B-KEY
symptoms	I-KEY
.	O
In	O
the	O
short	O
term	O
,	O
the	O
VR	O
treatment	O
was	O
more	O
effective	O
than	O
the	O
traditional	O
cognitive-behavioral	O
psychonutritional	B-KEY
groups	I-KEY
in	O
improving	O
the	O
overall	O
psychological	O
state	O
of	O
the	O
patients	O
.	O
In	O
particular	O
,	O
the	O
therapy	O
was	O
more	O
effective	O
in	O
improving	O
body	O
satisfaction	O
,	O
self-efficacy	O
,	O
and	O
motivation	O
for	O
change	O
.	O
No	O
significant	O
differences	O
were	O
found	O
in	O
the	O
reduction	O
of	O
the	O
binge	O
eating	O
behavior	O
.	O
The	O
possibility	O
of	O
inducing	O
a	O
significant	O
change	O
in	O
body	O
image	O
and	O
its	O
associated	O
behaviors	O
using	O
a	O
VR-based	O
short-term	O
therapy	O
can	O
be	O
useful	O
to	O
improve	O
the	O
body	O
satisfaction	O
in	O
traditional	O
weight	O
reduction	O
programs	O
.	O
However	O
,	O
given	O
the	O
nature	O
of	O
this	O
research	O
that	O
does	O
not	O
include	O
a	O
followup	O
study	O
,	O
the	O
obtained	O
results	O
are	O
preliminary	O
only	O

Image	B-KEY
reconstruction	I-KEY
of	O
simulated	B-KEY
specimens	I-KEY
using	O
convolution	B-KEY
back	I-KEY
projection	I-KEY
This	O
paper	O
reports	O
the	O
reconstruction	O
of	O
cross-sections	O
of	O
composite	B-KEY
structures	I-KEY
.	O
The	O
convolution	B-KEY
back	I-KEY
projection	I-KEY
-LRB-	O
CBP	O
-RRB-	O
algorithm	O
has	O
been	O
used	O
to	O
capture	O
the	O
attenuation	B-KEY
field	I-KEY
over	O
the	O
specimen	O
.	O
Five	O
different	O
test	O
cases	O
have	O
been	O
taken	O
up	O
for	O
evaluation	O
.	O
These	O
cases	O
represent	O
varying	O
degrees	O
of	O
complexity	O
.	O
In	O
addition	O
,	O
the	O
role	O
of	O
filters	B-KEY
on	O
the	O
nature	O
of	O
the	O
reconstruction	B-KEY
errors	I-KEY
has	O
also	O
been	O
discussed	O
.	O
Numerical	O
results	O
obtained	O
in	O
the	O
study	O
reveal	O
that	O
CBP	B-KEY
algorithm	I-KEY
is	O
a	O
useful	O
tool	O
for	O
qualitative	O
as	O
well	O
as	O
quantitative	O
assessment	O
of	O
composite	B-KEY
regions	I-KEY
encountered	O
in	O
engineering	B-KEY
applications	I-KEY

Reconstructing	O
surfaces	O
by	O
volumetric	B-KEY
regularization	I-KEY
using	O
radial	B-KEY
basis	I-KEY
functions	I-KEY
We	O
present	O
a	O
new	O
method	O
of	O
surface	B-KEY
reconstruction	I-KEY
that	O
generates	O
smooth	O
and	O
seamless	O
models	O
from	O
sparse	O
,	O
noisy	O
,	O
nonuniform	O
,	O
and	O
low	B-KEY
resolution	I-KEY
range	I-KEY
data	I-KEY
.	O
Data	B-KEY
acquisition	I-KEY
techniques	I-KEY
from	O
computer	B-KEY
vision	I-KEY
,	O
such	O
as	O
stereo	B-KEY
range	I-KEY
images	I-KEY
and	O
space	B-KEY
carving	I-KEY
,	O
produce	O
3D	B-KEY
point	I-KEY
sets	I-KEY
that	O
are	O
imprecise	O
and	O
nonuniform	O
when	O
compared	O
to	O
laser	O
or	O
optical	O
range	O
scanners	O
.	O
Traditional	O
reconstruction	O
algorithms	O
designed	O
for	O
dense	O
and	O
precise	O
data	O
do	O
not	O
produce	O
smooth	O
reconstructions	O
when	O
applied	O
to	O
vision-based	B-KEY
data	I-KEY
sets	I-KEY
.	O
Our	O
method	O
constructs	O
a	O
3D	B-KEY
implicit	I-KEY
surface	I-KEY
,	O
formulated	O
as	O
a	O
sum	O
of	O
weighted	B-KEY
radial	I-KEY
basis	I-KEY
functions	I-KEY
.	O
We	O
achieve	O
three	O
primary	O
advantages	O
over	O
existing	O
algorithms	O
:	O
-LRB-	O
1	O
-RRB-	O
the	O
implicit	O
functions	O
we	O
construct	O
estimate	O
the	O
surface	O
well	O
in	O
regions	O
where	O
there	O
is	O
little	O
data	O
,	O
-LRB-	O
2	O
-RRB-	O
the	O
reconstructed	O
surface	O
is	O
insensitive	O
to	O
noise	O
in	O
data	O
acquisition	O
because	O
we	O
can	O
allow	O
the	O
surface	O
to	O
approximate	O
,	O
rather	O
than	O
exactly	O
interpolate	O
,	O
the	O
data	O
,	O
and	O
-LRB-	O
3	O
-RRB-	O
the	O
reconstructed	O
surface	O
is	O
locally	O
detailed	O
,	O
yet	O
globally	O
smooth	O
,	O
because	O
we	O
use	O
radial	B-KEY
basis	I-KEY
functions	I-KEY
that	O
achieve	O
multiple	O
orders	O
of	O
smoothness	O

Efficient	O
simplicial	B-KEY
reconstructions	I-KEY
of	O
manifolds	O
from	O
their	O
samples	O
An	O
algorithm	O
for	O
manifold	B-KEY
learning	I-KEY
is	O
presented	O
.	O
Given	O
only	O
samples	O
of	O
a	O
finite-dimensional	B-KEY
differentiable	I-KEY
manifold	I-KEY
and	O
no	O
a	O
priori	O
knowledge	O
of	O
the	O
manifold	O
's	O
geometry	O
or	O
topology	O
except	O
for	O
its	O
dimension	O
,	O
the	O
goal	O
is	O
to	O
find	O
a	O
description	O
of	O
the	O
manifold	O
.	O
The	O
learned	B-KEY
manifold	I-KEY
must	O
approximate	O
the	O
true	B-KEY
manifold	I-KEY
well	O
,	O
both	O
geometrically	O
and	O
topologically	O
,	O
when	O
the	O
sampling	B-KEY
density	I-KEY
is	O
sufficiently	O
high	O
.	O
The	O
proposed	O
algorithm	O
constructs	O
a	O
simplicial	B-KEY
complex	I-KEY
based	O
on	O
approximations	O
to	O
the	O
tangent	O
bundle	O
of	O
the	O
manifold	O
.	O
An	O
important	O
property	O
of	O
the	O
algorithm	O
is	O
that	O
its	O
complexity	O
depends	O
on	O
the	O
dimension	O
of	O
the	O
manifold	O
,	O
rather	O
than	O
that	O
of	O
the	O
embedding	O
space	O
.	O
Successful	O
examples	O
are	O
presented	O
in	O
the	O
cases	O
of	O
learning	O
curves	O
in	O
the	O
plane	O
,	O
curves	O
in	O
space	O
,	O
and	O
surfaces	O
in	O
space	O
;	O
in	O
addition	O
,	O
a	O
case	O
when	O
the	O
algorithm	O
fails	O
is	O
analyzed	O

Technology	B-KEY
CAD	I-KEY
of	O
SiGe-heterojunction	O
field	O
effect	O
transistors	O
A	O
2D	O
virtual	O
wafer	O
fabrication	O
simulation	O
suite	O
has	O
been	O
employed	O
for	O
the	O
technology	B-KEY
CAD	I-KEY
of	O
SiGe	B-KEY
channel	O
heterojunction	B-KEY
field	I-KEY
effect	I-KEY
transistors	I-KEY
-LRB-	O
HFETs	O
-RRB-	O
.	O
Complete	O
fabrication	B-KEY
process	I-KEY
of	O
SiGe	B-KEY
p-HFETs	O
has	O
been	O
simulated	O
.	O
The	O
SiGe	B-KEY
material	B-KEY
parameters	I-KEY
and	O
mobility	B-KEY
model	I-KEY
were	O
incorporated	O
to	O
simulate	O
Si/SiGe	O
p-HFETs	O
with	O
a	O
uniform	O
germanium	O
channel	O
having	O
an	O
L/sub	O
eff	O
/	O
of	O
0.5	O
mu	O
m	O
.	O
A	O
significant	O
improvement	O
in	O
linear	B-KEY
transconductance	I-KEY
is	O
observed	O
when	O
compared	O
to	O
control-silicon	O
p-MOSFETs	O

Variety	B-KEY
identification	I-KEY
of	O
wheat	O
using	O
mass	O
spectrometry	O
with	O
neural	O
networks	O
and	O
the	O
influence	O
of	O
mass	B-KEY
spectra	I-KEY
processing	I-KEY
prior	O
to	O
neural	B-KEY
network	I-KEY
analysis	I-KEY
The	O
performance	O
of	O
matrix-assisted	B-KEY
laser	I-KEY
desorption/ionisation	I-KEY
time-of-flight	I-KEY
mass	I-KEY
spectrometry	I-KEY
with	O
neural	O
networks	O
in	O
wheat	B-KEY
variety	I-KEY
classification	I-KEY
is	O
further	O
evaluated	O
.	O
Two	O
principal	O
issues	O
were	O
studied	O
:	O
-LRB-	O
a	O
-RRB-	O
the	O
number	O
of	O
varieties	O
that	O
could	O
be	O
classified	O
correctly	O
;	O
and	O
-LRB-	O
b	O
-RRB-	O
various	O
means	O
of	O
preprocessing	O
mass	B-KEY
spectrometric	I-KEY
data	I-KEY
.	O
The	O
number	O
of	O
wheat	O
varieties	O
tested	O
was	O
increased	O
from	O
10	O
to	O
30	O
.	O
The	O
main	O
pre-processing	O
method	O
investigated	O
was	O
based	O
on	O
Gaussian	B-KEY
smoothing	I-KEY
of	O
the	O
spectra	O
,	O
but	O
other	O
methods	O
based	O
on	O
normalisation	B-KEY
procedures	I-KEY
and	O
multiplicative	B-KEY
scatter	I-KEY
correction	I-KEY
of	O
data	O
were	O
also	O
used	O
.	O
With	O
the	O
final	O
method	O
,	O
it	O
was	O
possible	O
to	O
classify	O
30	O
wheat	O
varieties	O
with	O
87	O
%	O
correctly	B-KEY
classified	I-KEY
mass	I-KEY
spectra	I-KEY
and	O
a	O
correlation	B-KEY
coefficient	I-KEY
of	O
0.90	O

The	O
free	O
lunch	O
is	O
over	O
:	O
online	B-KEY
content	I-KEY
subscriptions	I-KEY
on	O
the	O
rise	O
High	O
need	O
,	O
rather	O
than	O
high	O
use	O
,	O
may	O
be	O
what	O
really	O
determines	O
a	O
user	O
's	O
willingness	O
to	O
pay	O
.	O
Retooling	O
and	O
targeting	O
content	O
may	O
be	O
a	O
sharper	O
strategy	O
than	O
trying	O
to	O
re-educate	O
users	O
that	O
it	O
is	O
time	O
to	O
pay	O
up	O
for	O
material	O
that	O
has	O
been	O
free	O
.	O
Waiting	O
for	O
a	O
paradigm	O
shift	O
in	O
general	O
user	O
attitudes	O
about	O
paying	O
for	O
online	O
content	O
Could	O
be	O
a	O
fool	O
's	O
errand	O

Do	O
you	O
see	O
what	O
I	O
see	O
?	O
-LSB-	O
visual	B-KEY
technology	I-KEY
in	O
law	B-KEY
firms	I-KEY
-RSB-	O
Think	O
of	O
how	O
well-done	O
computer	B-KEY
presentations	I-KEY
can	O
aid	O
in	O
the	O
learning	O
experience	O
.	O
They	O
are	O
,	O
however	O
,	O
less	O
common	O
in	O
client	O
meetings	O
,	O
settlement	O
conferences	O
and	O
the	O
courtroom	O
.	O
And	O
you	O
have	O
to	O
wonder	O
why	O
,	O
when	O
the	O
same	O
benefits	O
of	O
attention	O
focus	O
and	O
visual	O
learning	O
apply	O
in	O
those	O
legal	O
communication	O
settings	O
.	O
The	O
software	O
and	O
hardware	O
components	O
are	O
easy	O
to	O
use	O
,	O
and	O
they	O
're	O
increasingly	O
affordable	O
to	O
boot	O
.	O
The	O
next	O
time	O
you	O
need	O
to	O
convey	O
a	O
point	O
to	O
an	O
audience	O
-LRB-	O
be	O
it	O
one	O
person	O
or	O
many	O
-RRB-	O
,	O
think	O
of	O
how	O
you	O
might	O
benefit	O
from	O
the	O
visual	O
impact	O
available	O
through	O
presentation	O
software	O
like	O
PowerPoint	B-KEY
.	O
Anyone	O
will	O
understand	O
you	O
more	O
easily	O
when	O
assisted	O
by	O
visual	O
input	O
,	O
and	O
it	O
may	O
make	O
all	O
the	O
difference	O
in	O
reaching	O
visual-focused	O
learners	O

On	O
generalized	B-KEY
Gaussian	I-KEY
quadratures	I-KEY
for	O
exponentials	O
and	O
their	O
applications	O
We	O
introduce	O
new	O
families	O
of	O
Gaussian-type	O
quadratures	O
for	O
weighted	B-KEY
integrals	I-KEY
of	O
exponential	O
functions	O
and	O
consider	O
their	O
applications	O
to	O
integration	O
and	O
interpolation	O
of	O
bandlimited	O
functions	O
.	O
We	O
use	O
a	O
generalization	O
of	O
a	O
representation	O
theorem	O
due	O
to	O
Caratheodory	O
to	O
derive	O
these	O
quadratures	O
.	O
For	O
each	O
positive	O
measure	O
,	O
the	O
quadratures	O
are	O
parameterized	O
by	O
eigenvalues	B-KEY
of	O
the	O
Toeplitz	B-KEY
matrix	I-KEY
constructed	O
from	O
the	O
trigonometric	B-KEY
moments	I-KEY
of	O
the	O
measure	O
.	O
For	O
a	O
given	O
accuracy	O
epsilon	O
,	O
selecting	O
an	O
eigenvalue	B-KEY
close	O
to	O
epsilon	O
yields	O
an	O
approximate	B-KEY
quadrature	O
with	O
that	O
accuracy	O
.	O
To	O
compute	O
its	O
weights	O
and	O
nodes	O
,	O
we	O
present	O
a	O
new	O
fast	O
algorithm	O
.	O
These	O
new	O
quadratures	O
can	O
be	O
used	O
to	O
approximate	B-KEY
and	O
integrate	B-KEY
bandlimited	B-KEY
functions	I-KEY
,	O
such	O
as	O
prolate	B-KEY
spheroidal	I-KEY
wave	I-KEY
functions	I-KEY
,	O
and	O
essentially	O
bandlimited	B-KEY
functions	I-KEY
,	O
such	O
as	O
Bessel	B-KEY
functions	I-KEY
.	O
We	O
also	O
develop	O
,	O
for	O
a	O
given	O
precision	O
,	O
an	O
interpolating	B-KEY
basis	O
for	O
bandlimited	B-KEY
functions	I-KEY
on	O
an	O
interval	O

Structure	O
of	O
weakly	B-KEY
invertible	I-KEY
semi-input-memory	O
finite	O
automata	O
with	O
delay	O
1	O
Semi-input-memory	B-KEY
finite	B-KEY
automata	I-KEY
,	O
a	O
kind	O
of	O
finite	O
automata	O
introduced	O
by	O
the	O
first	O
author	O
of	O
this	O
paper	O
for	O
studying	O
error	O
propagation	O
,	O
are	O
a	O
generalization	O
of	O
input	O
memory	O
finite	O
automata	O
by	O
appending	O
an	O
autonomous	O
finite	O
automaton	O
component	O
.	O
In	O
this	O
paper	O
,	O
we	O
give	O
a	O
characterization	O
of	O
the	O
structure	O
of	O
weakly	B-KEY
invertible	I-KEY
semi-input-memory	O
finite	O
automata	O
with	O
delay	O
1	O
,	O
in	O
which	O
the	O
state	O
graph	O
of	O
each	O
autonomous	O
finite	O
automaton	O
is	O
a	O
cycle	O
.	O
From	O
a	O
result	O
on	O
mutual	O
invertibility	B-KEY
of	O
finite	B-KEY
automata	I-KEY
obtained	O
by	O
the	O
authors	O
recently	O
,	O
it	O
leads	O
to	O
a	O
characterization	O
of	O
the	O
structure	O
of	O
feedforward	B-KEY
inverse	I-KEY
finite	I-KEY
automata	I-KEY
with	O
delay	O
1	O

A	O
switching	B-KEY
synchronization	I-KEY
scheme	I-KEY
for	O
a	O
class	O
of	O
chaotic	B-KEY
systems	I-KEY
In	O
this	O
Letter	O
,	O
we	O
propose	O
an	O
observer-based	O
synchronization	O
scheme	O
for	O
a	O
class	O
of	O
chaotic	B-KEY
systems	I-KEY
.	O
This	O
class	O
of	O
systems	O
are	O
given	O
by	O
piecewise-linear	B-KEY
dynamics	I-KEY
.	O
By	O
using	O
some	O
properties	O
of	O
such	O
systems	O
,	O
we	O
give	O
a	O
procedure	O
to	O
construct	O
the	O
gain	O
of	O
the	O
observer	O
.	O
We	O
prove	O
various	O
stability	O
results	O
and	O
comment	O
on	O
the	O
robustness	B-KEY
of	O
the	O
proposed	O
scheme	O
.	O
We	O
also	O
present	O
some	O
simulation	O
results	O

Ventilation-perfusion	B-KEY
ratio	I-KEY
of	O
signal	B-KEY
intensity	I-KEY
in	O
human	B-KEY
lung	I-KEY
using	O
oxygen-enhanced	O
and	O
arterial	B-KEY
spin	I-KEY
labeling	I-KEY
techniques	I-KEY
This	O
study	O
investigates	O
the	O
distribution	O
of	O
ventilation-perfusion	O
-LRB-	O
V/Q	O
-RRB-	O
signal	B-KEY
intensity	I-KEY
-LRB-	O
SI	O
-RRB-	O
ratios	O
using	O
oxygen-enhanced	O
and	O
arterial	O
spin	O
labeling	O
-LRB-	O
ASL	O
-RRB-	O
techniques	O
in	O
the	O
lungs	O
of	O
10	O
healthy	O
volunteers	O
.	O
Ventilation	O
and	O
perfusion	B-KEY
images	I-KEY
were	O
simultaneously	O
acquired	O
using	O
the	O
flow-sensitive	B-KEY
alternating	I-KEY
inversion	I-KEY
recovery	I-KEY
-LRB-	O
FAIR	O
-RRB-	O
method	O
as	O
volunteers	O
alternately	O
inhaled	O
room	O
air	O
and	O
100	O
%	O
oxygen	O
.	O
Images	O
of	O
the	O
T/sub	O
1	O
/	O
distribution	O
were	O
calculated	O
for	O
five	O
volunteers	O
for	O
both	O
selective	O
-LRB-	O
T/sub	O
1f	O
/	O
-RRB-	O
and	O
nonselective	O
-LRB-	O
T/sub	O
1	O
/	O
-RRB-	O
inversion	O
.	O
The	O
average	O
T/sub	O
1	O
/	O
was	O
1360	O
ms	O
+	O
or-116	O
ms	O
,	O
and	O
the	O
average	O
T/sub	O
1f	O
/	O
was	O
1012	O
ms	O
+	O
or-112	O
ms	O
,	O
yielding	O
a	O
difference	O
that	O
is	O
statistically	O
significant	O
-LRB-	O
P	O
<	O
0.002	O
-RRB-	O
.	O
Excluding	O
large	O
pulmonary	O
vessels	O
,	O
the	O
average	O
V/Q	O
SI	O
ratios	O
were	O
0.355	O
+	O
or-0	O
.073	O
for	O
the	O
left	O
lung	O
and	O
0.371	O
+	O
or-0	O
.093	O
for	O
the	O
right	O
lung	O
,	O
which	O
are	O
in	O
agreement	O
with	O
the	O
theoretical	O
V/Q	O
SI	O
ratio	O
.	O
Plots	O
of	O
the	O
WO	O
SI	O
ratio	O
are	O
similar	O
to	O
the	O
logarithmic	B-KEY
normal	I-KEY
distribution	I-KEY
obtained	O
by	O
multiple	B-KEY
inert	I-KEY
gas	I-KEY
elimination	I-KEY
techniques	O
,	O
with	O
a	O
range	O
of	O
ratios	O
matching	O
ventilation	O
and	O
perfusion	O
.	O
This	O
MRI	B-KEY
V/Q	O
technique	O
is	O
completely	O
noninvasive	O
and	O
does	O
not	O
involve	O
ionized	O
radiation	O
.	O
A	O
limitation	O
of	O
this	O
method	O
is	O
the	O
nonsimultaneous	B-KEY
acquisition	I-KEY
of	O
perfusion	O
and	O
ventilation	O
data	O
,	O
with	O
oxygen	O
administered	O
only	O
for	O
the	O
ventilation	O
data	O

On	O
the	O
discretization	O
of	O
double-bracket	O
flows	O
This	O
paper	O
extends	O
the	O
method	O
of	O
Magnus	B-KEY
series	I-KEY
to	O
Lie-algebraic	B-KEY
equations	I-KEY
originating	O
in	O
double-bracket	O
flows	O
.	O
We	O
show	O
that	O
the	O
solution	O
of	O
the	O
isospectral	B-KEY
flow	I-KEY
Y	O
'	O
=	O
-LSB-	O
-LSB-	O
Y	O
,	O
N	O
-RSB-	O
,	O
Y	O
-RSB-	O
,	O
Y	O
-LRB-	O
O	O
-RRB-	O
=	O
Y/sub	O
0	O
/	O
in	O
Sym	O
-LRB-	O
n	O
-RRB-	O
,	O
can	O
be	O
represented	O
in	O
the	O
form	O
Y	O
-LRB-	O
t	O
-RRB-	O
=	O
e/sup	O
Omega	O
-LRB-	O
t	O
-RRB-	O
/	O
Y/sub	O
0/e/sup	O
-	O
Omega	O
-LRB-	O
1	O
-RRB-	O
/	O
,	O
where	O
the	O
Taylor	B-KEY
expansion	I-KEY
of	O
Omega	O
can	O
be	O
constructed	O
explicitly	O
,	O
term-by-term	O
,	O
identifying	O
individual	O
expansion	O
terms	O
with	O
certain	O
rooted	O
trees	O
with	O
bicolor	B-KEY
leaves	I-KEY
.	O
This	O
approach	O
is	O
extended	O
to	O
other	O
Lie-algebraic	B-KEY
equations	I-KEY
that	O
can	O
be	O
appropriately	O
expressed	O
in	O
terms	O
of	O
a	O
finite	O
``	O
alphabet	O
''	O

Estimation	O
of	O
the	O
Poisson	B-KEY
stream	I-KEY
intensity	I-KEY
in	O
a	O
multilinear	O
queue	O
with	O
an	O
exponential	O
job	O
queue	O
decay	O
Times	O
the	O
busy	B-KEY
queue	I-KEY
periods	I-KEY
start	I-KEY
are	O
found	O
for	O
a	O
multilinear	B-KEY
queue	I-KEY
with	O
an	O
exponential	B-KEY
job	I-KEY
queue	I-KEY
decay	I-KEY
and	O
uniform	B-KEY
resource	I-KEY
allocation	I-KEY
to	O
individual	B-KEY
servers	I-KEY
.	O
The	O
stream	B-KEY
intensity	I-KEY
and	O
the	O
average	O
job	O
are	O
estimated	O
from	O
observations	O
of	O
the	O
times	O
the	O
queue	O
busy	O
periods	O
start	O

A	O
stochastic	B-KEY
averaging	I-KEY
approach	O
for	O
feedback	B-KEY
control	I-KEY
design	O
of	O
nonlinear	B-KEY
systems	I-KEY
under	O
random	B-KEY
excitations	I-KEY
This	O
paper	O
presents	O
a	O
method	O
for	O
designing	O
and	O
quantifying	O
the	O
performance	O
of	O
feedback	B-KEY
stochastic	I-KEY
controls	I-KEY
for	O
nonlinear	B-KEY
systems	I-KEY
.	O
The	O
design	O
makes	O
use	O
of	O
the	O
method	O
of	O
stochastic	B-KEY
averaging	I-KEY
to	O
reduce	O
the	O
dimension	O
of	O
the	O
state	O
space	O
and	O
to	O
derive	O
the	O
Ito	B-KEY
stochastic	I-KEY
differential	I-KEY
equation	I-KEY
for	O
the	O
response	O
amplitude	O
process	O
.	O
The	O
moment	O
equation	O
of	O
the	O
amplitude	O
process	O
closed	O
by	O
the	O
Rayleigh	B-KEY
approximation	I-KEY
is	O
used	O
as	O
a	O
means	O
to	O
characterize	O
the	O
transient	O
performance	O
of	O
the	O
feedback	B-KEY
control	I-KEY
.	O
The	O
steady	B-KEY
state	I-KEY
and	O
transient	B-KEY
response	I-KEY
of	O
the	O
amplitude	O
process	O
are	O
used	O
as	O
the	O
design	O
criteria	O
for	O
choosing	O
the	O
feedback	B-KEY
control	I-KEY
gains	O
.	O
Numerical	O
examples	O
are	O
studied	O
to	O
demonstrate	O
the	O
performance	O
of	O
the	O
control	O

Marble	B-KEY
cutting	I-KEY
with	O
single	B-KEY
point	I-KEY
cutting	I-KEY
tool	I-KEY
and	O
diamond	B-KEY
segments	I-KEY
An	O
investigation	O
has	O
been	O
undertaken	O
into	O
the	O
frame	B-KEY
sawing	I-KEY
with	O
diamond	O
blades	O
.	O
The	O
kinematic	B-KEY
behaviour	I-KEY
of	O
the	O
frame	B-KEY
sawing	I-KEY
process	O
is	O
discussed	O
.	O
Under	O
different	O
cutting	O
conditions	O
,	O
cutting	O
and	O
indenting-cutting	B-KEY
tests	I-KEY
are	O
carried	O
out	O
by	O
single	O
point	O
cutting	O
tools	O
and	O
single	O
diamond	O
segments	O
.	O
The	O
results	O
indicate	O
that	O
the	O
depth	O
of	O
cut	O
per	O
diamond	O
grit	O
increases	O
as	O
the	O
blades	O
move	O
forward	O
.	O
Only	O
a	O
few	O
grits	O
per	O
segment	O
can	O
remove	O
the	O
material	O
in	O
the	O
cutting	O
process	O
.	O
When	O
the	O
direction	O
of	O
the	O
stroke	O
changes	O
,	O
the	O
cutting	O
forces	O
do	O
not	O
decrease	O
to	O
zero	O
because	O
of	O
the	O
residual	B-KEY
plastic	I-KEY
deformation	I-KEY
beneath	O
the	O
diamond	O
grits	O
.	O
The	O
plastic	O
deformation	O
and	O
fracture	B-KEY
chipping	I-KEY
of	O
material	O
are	O
the	O
dominant	O
removal	B-KEY
processes	I-KEY
,	O
which	O
can	O
be	O
explained	O
by	O
the	O
fracture	B-KEY
theory	I-KEY
of	O
brittle	B-KEY
material	I-KEY
indentation	I-KEY

Women	B-KEY
in	O
computing	B-KEY
history	I-KEY
Exciting	O
inventions	O
,	O
innovative	O
technology	O
,	O
human	O
interaction	O
,	O
and	O
intriguing	O
politics	O
fill	O
computing	B-KEY
history	I-KEY
.	O
However	O
,	O
the	O
recorded	O
history	O
is	O
mainly	O
composed	O
of	O
male	O
achievements	O
and	O
involvements	O
,	O
even	O
though	O
women	B-KEY
have	O
played	O
substantial	O
roles	O
.	O
This	O
situation	O
is	O
not	O
unusual	O
.	O
Most	O
science	O
fields	O
are	O
notorious	O
for	O
excluding	O
,	O
undervaluing	O
,	O
or	O
overlooking	O
the	O
accomplishments	O
of	O
their	O
female	O
scientists	O
.	O
As	O
Lee	O
points	O
out	O
,	O
it	O
is	O
up	O
to	O
the	O
historians	O
and	O
others	O
to	O
remedy	O
this	O
imbalance	O
.	O
Steps	O
have	O
been	O
taken	O
towards	O
this	O
goal	O
through	O
publishing	O
biographies	O
on	O
women	B-KEY
in	O
technology	O
,	O
and	O
through	O
honoring	O
the	O
pioneers	O
with	O
various	O
awards	O
such	O
as	O
the	O
GHC	O
'97	O
Pioneering	O
Awards	O
,	O
the	O
WITI	O
Hall	O
of	O
Fame	O
,	O
and	O
the	O
AWC	O
Lovelace	O
Award	O
.	O
A	O
few	O
online	O
sites	O
contain	O
biographies	O
of	O
women	B-KEY
in	O
technology	O
.	O
However	O
,	O
even	O
with	O
these	O
resources	O
,	O
many	O
women	B-KEY
who	O
have	O
contributed	O
significantly	O
to	O
computer	O
science	O
are	O
still	O
to	O
be	O
discovered	O

Rank	B-KEY
tests	I-KEY
of	O
association	B-KEY
for	O
exchangeable	O
paired	O
data	O
We	O
describe	O
two	O
rank	B-KEY
tests	I-KEY
of	O
association	B-KEY
for	O
paired	B-KEY
exchangeable	I-KEY
data	I-KEY
motivated	O
by	O
the	O
study	O
of	O
lifespans	O
in	O
twins	O
.	O
The	O
pooled	B-KEY
sample	I-KEY
is	O
ranked	O
.	O
The	O
nonparametric	B-KEY
test	I-KEY
of	O
association	B-KEY
is	O
based	O
on	O
R/sup	O
+	O
/	O
,	O
the	O
sum	O
of	O
the	O
smaller	O
within-pair	B-KEY
ranks	I-KEY
.	O
A	O
second	O
measure	O
L/sup	O
+	O
/	O
is	O
the	O
sum	O
of	O
within-pair	B-KEY
rank	I-KEY
products	O
.	O
Under	O
the	O
null	B-KEY
hypothesis	I-KEY
of	O
within-pair	B-KEY
independence	I-KEY
,	O
the	O
two	O
test	B-KEY
statistics	I-KEY
are	O
approximately	O
normally	O
distributed	O
.	O
Expressions	O
for	O
the	O
exact	B-KEY
means	I-KEY
and	O
variances	O
of	O
R/sup	O
+	O
/	O
and	O
L/sup	O
+	O
/	O
are	O
given	O
.	O
We	O
describe	O
the	O
power	O
of	O
these	O
two	O
statistics	O
under	O
a	O
close	O
alternative	O
hypothesis	O
to	O
that	O
of	O
independence	O
.	O
Both	O
the	O
R/sup	O
+	O
/	O
and	O
L/sup	O
+	O
/	O
tests	O
indicate	O
nonparametric	B-KEY
statistical	I-KEY
evidence	I-KEY
of	O
positive	O
association	B-KEY
of	O
longevity	B-KEY
in	O
identical	B-KEY
twins	I-KEY
and	O
a	O
negligible	O
relationship	O
between	O
the	O
lifespans	O
of	O
fraternal	B-KEY
twins	I-KEY
listed	O
in	O
the	O
Danish	B-KEY
twin	I-KEY
registry	I-KEY
.	O
The	O
statistics	O
are	O
also	O
applied	O
to	O
the	O
analysis	O
of	O
a	O
clinical	B-KEY
trial	I-KEY
studying	O
the	O
time	O
to	O
failure	O
of	O
ventilation	O
tubes	O
in	O
children	O
with	O
bilateral	B-KEY
otitis	I-KEY
media	I-KEY

Community	B-KEY
technology	I-KEY
and	O
democratic	B-KEY
rationalization	I-KEY
The	O
objective	O
of	O
the	O
paper	O
is	O
to	O
explore	O
questions	O
of	O
human	B-KEY
agency	I-KEY
and	O
democratic	B-KEY
process	I-KEY
in	O
the	O
technical	B-KEY
sphere	I-KEY
through	O
the	O
example	O
of	O
``	O
virtual	B-KEY
community	I-KEY
.	O
''	O
The	O
formation	O
of	O
relatively	O
stable	B-KEY
long-term	I-KEY
group	I-KEY
associations	I-KEY
-LRB-	O
community	O
in	O
the	O
broad	O
sense	O
of	O
the	O
term	O
-RRB-	O
,	O
is	O
the	O
scene	O
on	O
which	O
a	O
large	O
share	O
of	O
human	B-KEY
development	I-KEY
occurs	O
.	O
As	O
such	O
it	O
is	O
a	O
fundamental	O
human	B-KEY
value	I-KEY
mobilizing	O
diverse	B-KEY
ideologies	I-KEY
and	O
sensitivities	O
.	O
The	O
promise	O
of	O
realizing	O
this	O
value	O
in	O
a	O
new	O
domain	O
naturally	O
stirs	O
up	O
much	O
excitement	O
among	O
optimistic	B-KEY
observers	I-KEY
of	O
the	O
Internet	O
.	O
At	O
the	O
same	O
time	O
,	O
the	O
eagerness	O
to	O
place	O
hopes	O
for	O
community	O
in	O
a	O
technical	B-KEY
system	I-KEY
flies	O
in	O
the	O
face	O
of	O
an	O
influential	O
intellectual	B-KEY
tradition	I-KEY
of	O
technology	B-KEY
criticism	I-KEY
.	O
This	O
eagerness	O
seems	O
even	O
more	O
naive	O
in	O
the	O
light	O
of	O
the	O
recent	O
commercialization	O
of	O
so	O
much	O
Internet	B-KEY
activity	I-KEY
.	O
Despite	O
the	O
widespread	O
skepticism	O
,	O
we	O
believe	O
the	O
growth	O
of	O
virtual	B-KEY
community	I-KEY
is	O
significant	O
for	O
an	O
inquiry	O
into	O
the	O
democratization	O
of	O
technology	O
.	O
We	O
show	O
that	O
conflicting	B-KEY
answers	I-KEY
to	O
the	O
central	O
question	O
of	O
the	O
present	O
theoretical	O
debate	O
-	O
Is	O
community	O
possible	O
on	O
computer	B-KEY
networks	I-KEY
?	O
epsilon	O
neralize	O
from	O
particular	O
features	O
of	O
systems	O
and	O
software	O
prevalent	O
at	O
different	O
stages	O
in	O
the	O
development	O
of	O
computer	B-KEY
networking	I-KEY
.	O
We	O
conclude	O
that	O
research	O
should	O
focus	O
instead	O
on	O
how	O
to	O
design	O
computer	B-KEY
networks	I-KEY
to	O
better	O
support	O
community	B-KEY
activities	I-KEY
and	O
values	O

SPTL/BIALL	B-KEY
academic	B-KEY
law	I-KEY
library	I-KEY
survey	B-KEY
2000/2001	O
The	O
paper	O
outlines	O
the	O
activities	O
and	O
funding	B-KEY
of	O
academic	B-KEY
law	I-KEY
libraries	I-KEY
in	O
the	O
UK	B-KEY
and	O
Ireland	B-KEY
in	O
the	O
academic	O
year	O
2000/2001	O
.	O
The	O
figures	O
have	O
been	O
taken	O
from	O
the	O
results	O
of	O
a	O
postal	B-KEY
questionnaire	I-KEY
undertaken	O
by	O
information	B-KEY
services	I-KEY
staff	O
at	O
Cardiff	B-KEY
University	I-KEY
on	O
behalf	O
of	O
BIALL	O

Error-probability	O
analysis	O
of	O
MIL-STD-1773	O
optical	B-KEY
fiber	I-KEY
data	I-KEY
buses	I-KEY
We	O
have	O
analyzed	O
the	O
error	B-KEY
probabilities	I-KEY
of	O
MIL-STD-1773	O
optical	B-KEY
fiber	I-KEY
data	I-KEY
buses	I-KEY
with	O
three	O
modulation	B-KEY
schemes	I-KEY
,	O
namely	O
,	O
original	O
Manchester	O
II	O
bi-phase	O
coding	O
,	O
PTMBC	O
,	O
and	O
EMBC-BSF	O
.	O
Using	O
these	O
derived	O
expressions	O
of	O
error	B-KEY
probabilities	I-KEY
,	O
we	O
can	O
also	O
compare	O
the	O
receiver	B-KEY
sensitivities	I-KEY
of	O
such	O
optical	B-KEY
fiber	I-KEY
data	I-KEY
buses	I-KEY

Efficient	O
transitive	B-KEY
closure	I-KEY
reasoning	I-KEY
in	O
a	O
combined	O
class/part/containment	O
hierarchy	O
Class	B-KEY
hierarchies	I-KEY
form	O
the	O
backbone	O
of	O
many	O
implemented	O
knowledge	B-KEY
representation	I-KEY
and	O
reasoning	O
systems	O
.	O
They	O
are	O
used	O
for	O
inheritance	B-KEY
,	O
classification	B-KEY
and	O
transitive	B-KEY
closure	I-KEY
reasoning	I-KEY
.	O
Part	B-KEY
hierarchies	I-KEY
are	O
also	O
important	O
in	O
artificial	B-KEY
intelligence	I-KEY
.	O
Other	O
hierarchies	O
,	O
e.g.	O
containment	B-KEY
hierarchies	I-KEY
,	O
have	O
received	O
less	O
attention	O
in	O
artificial	B-KEY
intelligence	I-KEY
.	O
This	O
paper	O
presents	O
an	O
architecture	O
and	O
an	O
implementation	O
of	O
a	O
hierarchy	O
reasoner	O
that	O
integrates	O
a	O
class	B-KEY
hierarchy	I-KEY
,	O
a	O
part	B-KEY
hierarchy	I-KEY
,	O
and	O
a	O
containment	B-KEY
hierarchy	I-KEY
into	O
one	O
structure	O
.	O
In	O
order	O
to	O
make	O
an	O
implemented	O
reasoner	O
useful	O
,	O
it	O
needs	O
to	O
operate	O
at	O
least	O
at	O
speeds	O
comparable	O
to	O
human	O
reasoning	O
.	O
As	O
real-world	O
hierarchies	O
are	O
always	O
large	O
,	O
special	O
techniques	O
need	O
to	O
be	O
used	O
to	O
achieve	O
this	O
.	O
We	O
have	O
developed	O
a	O
set	O
of	O
parallel	B-KEY
algorithms	I-KEY
and	O
a	O
data	B-KEY
representation	I-KEY
called	O
maximally	B-KEY
reduced	I-KEY
tree	I-KEY
cover	I-KEY
for	O
that	O
purpose	O
.	O
The	O
maximally	B-KEY
reduced	I-KEY
tree	I-KEY
cover	I-KEY
is	O
an	O
improvement	O
of	O
a	O
materialized	B-KEY
transitive	I-KEY
closure	I-KEY
representation	I-KEY
which	O
has	O
appeared	O
in	O
the	O
literature	O
.	O
Our	O
experiments	B-KEY
with	O
a	O
medical	B-KEY
vocabulary	I-KEY
show	O
that	O
transitive	B-KEY
closure	I-KEY
reasoning	I-KEY
for	O
combined	O
class/part/containment	O
hierarchies	O
in	O
near	O
constant	O
time	O
is	O
possible	O
for	O
a	O
fixed	O
hardware	O
configuration	O

Building	O
an	O
effective	O
computer	B-KEY
science	I-KEY
student	I-KEY
organization	I-KEY
:	O
the	O
Carnegie	O
Mellon	O
Women@SCS	B-KEY
action	I-KEY
plan	I-KEY
This	O
paper	O
aims	O
to	O
provide	O
a	O
practical	O
guide	O
for	O
building	O
a	O
student	O
organization	O
and	O
designing	O
activities	O
and	O
events	O
that	O
can	O
encourage	O
and	O
support	O
a	O
community	O
of	O
women	B-KEY
in	O
computer	O
science	O
.	O
This	O
guide	O
is	O
based	O
on	O
our	O
experience	O
in	O
building	O
Women@SCS,	O
a	O
community	O
of	O
women	B-KEY
in	O
the	O
School	O
of	O
Computer	O
Science	O
-LRB-	O
SCS	O
-RRB-	O
at	O
Carnegie	B-KEY
Mellon	I-KEY
University	I-KEY
.	O
Rather	O
than	O
provide	O
an	O
abstract	O
``	O
to-do	O
''	O
or	O
``	O
must-do	O
''	O
list	O
,	O
we	O
present	O
a	O
sampling	O
of	O
concrete	O
activities	O
and	O
events	O
in	O
the	O
hope	O
that	O
these	O
might	O
suggest	O
possibilities	O
for	O
a	O
likeminded	O
student	O
organization	O
.	O
However	O
,	O
since	O
we	O
have	O
found	O
it	O
essential	O
to	O
have	O
a	O
core	O
group	O
of	O
activist	O
students	O
at	O
the	O
helm	O
,	O
we	O
provide	O
a	O
``	O
to-do	O
''	O
list	O
of	O
features	O
that	O
we	O
feel	O
are	O
essential	O
for	O
forming	O
,	O
supporting	O
and	O
sustaining	O
creative	O
and	O
effective	O
student	B-KEY
leadership	I-KEY

Friedberg	B-KEY
numberings	I-KEY
of	O
families	B-KEY
of	I-KEY
n-computably	I-KEY
enumerable	I-KEY
sets	I-KEY
We	O
establish	O
a	O
number	O
of	O
results	O
on	O
numberings	O
,	O
in	O
particular	O
,	O
on	O
Friedberg	B-KEY
numberings	I-KEY
,	O
of	O
families	O
of	O
d.c.e.	O
sets	O
.	O
First	O
,	O
it	O
is	O
proved	O
that	O
there	O
exists	O
a	O
Friedberg	B-KEY
numbering	I-KEY
of	O
the	O
family	O
of	O
all	O
d.c.e.	O
sets	O
.	O
We	O
also	O
show	O
that	O
this	O
result	O
,	O
patterned	O
on	O
Friedberg	O
's	O
famous	O
theorem	O
for	O
the	O
family	O
of	O
all	O
c.e.	O
sets	O
,	O
holds	O
for	O
the	O
family	O
of	O
all	O
n-c.e.	O
sets	O
for	O
any	O
n	O
>	O
2	O
.	O
Second	O
,	O
it	O
is	O
stated	O
that	O
there	O
exists	O
an	O
infinite	B-KEY
family	I-KEY
of	O
d.c.e.	O
sets	O
without	O
a	O
Friedberg	B-KEY
numbering	I-KEY
.	O
Third	O
,	O
it	O
is	O
shown	O
that	O
there	O
exists	O
an	O
infinite	B-KEY
family	I-KEY
of	O
c.e.	O
sets	O
-LRB-	O
treated	O
as	O
a	O
family	O
of	O
d.c.e.	O
sets	O
-RRB-	O
with	O
a	O
numbering	O
which	O
is	O
unique	O
up	O
to	O
equivalence	O
.	O
Fourth	O
,	O
it	O
is	O
proved	O
that	O
there	O
exists	O
a	O
family	O
of	O
d.c.e.	O
sets	O
with	O
a	O
least	O
numbering	O
-LRB-	O
under	O
reducibility	O
-RRB-	O
which	O
is	O
Friedberg	O
but	O
is	O
not	O
the	O
only	O
numbering	O
-LRB-	O
modulo	O
reducibility	O
-RRB-	O

Robust	B-KEY
Kalman	I-KEY
filter	I-KEY
design	O
for	O
discrete	B-KEY
time-delay	I-KEY
systems	I-KEY
The	O
problem	O
of	O
finite	O
-	O
and	O
infinite-horizon	O
robust	B-KEY
Kalman	I-KEY
filtering	I-KEY
for	O
uncertain	O
discrete-time	O
systems	O
with	O
state	B-KEY
delay	I-KEY
is	O
addressed	O
.	O
The	O
system	O
under	O
consideration	O
is	O
subject	O
to	O
time-varying	O
norm-bounded	B-KEY
parameter	I-KEY
uncertainty	I-KEY
in	O
both	O
the	O
state	O
and	O
output	B-KEY
matrices	I-KEY
.	O
We	O
develop	O
a	O
new	O
methodology	O
for	O
designing	O
a	O
linear	B-KEY
filter	I-KEY
such	O
that	O
the	O
error	O
variance	O
of	O
the	O
filter	O
is	O
guaranteed	O
to	O
be	O
within	O
a	O
certain	O
upper	O
bound	O
for	O
any	O
allowed	O
uncertainty	O
and	O
time	O
delay	O
.	O
The	O
solution	O
is	O
given	O
in	O
terms	O
of	O
two	O
Riccati	B-KEY
equations	I-KEY
.	O
Multiple	O
time-delay	O
systems	O
are	O
also	O
investigated	O

Tools	O
for	O
the	O
analysis	O
of	O
dose	O
optimization	O
.	O
I.	O
Effect-volume	B-KEY
histogram	I-KEY
With	O
the	O
advent	O
of	O
dose	B-KEY
optimization	I-KEY
algorithms	I-KEY
,	O
predominantly	O
for	O
intensity-modulated	B-KEY
radiotherapy	I-KEY
-LRB-	O
IMRT	O
-RRB-	O
,	O
computer	B-KEY
software	I-KEY
has	O
progressed	O
beyond	O
the	O
point	O
of	O
being	O
merely	O
a	O
tool	O
at	O
the	O
hands	O
of	O
an	O
expert	O
and	O
has	O
become	O
an	O
active	O
,	O
independent	O
mediator	O
of	O
the	O
dosimetric	B-KEY
conflicts	I-KEY
between	O
treatment	B-KEY
goals	I-KEY
and	O
risks	O
.	O
To	O
understand	O
and	O
control	O
the	O
internal	O
decision	O
finding	O
as	O
well	O
as	O
to	O
provide	O
means	O
to	O
influence	O
it	O
,	O
a	O
tool	O
for	O
the	O
analysis	O
of	O
the	O
dose	O
distribution	O
is	O
presented	O
which	O
reveals	O
the	O
decision-making	B-KEY
process	I-KEY
performed	O
by	O
the	O
algorithm	O
.	O
The	O
internal	O
trade-offs	O
between	O
partial	B-KEY
volumes	I-KEY
receiving	O
high	O
or	O
low	B-KEY
doses	I-KEY
are	O
driven	O
by	O
functions	O
which	O
attribute	O
a	O
weight	O
to	O
each	O
volume	O
element	O
.	O
The	O
statistics	O
of	O
the	O
distribution	O
of	O
these	O
weights	O
is	O
cast	O
into	O
an	O
effect-volume	B-KEY
histogram	I-KEY
-LRB-	O
EVH	O
-RRB-	O
in	O
analogy	O
to	O
dose-volume	O
histograms	O
.	O
The	O
analysis	O
of	O
the	O
EVH	O
reveals	O
which	O
traits	O
of	O
the	O
optimum	O
dose	O
distribution	O
result	O
from	O
the	O
defined	O
objectives	O
,	O
and	O
which	O
are	O
a	O
random	O
consequence	O
of	O
under	O
-	O
or	O
misspecification	O
of	O
treatment	B-KEY
goals	I-KEY
.	O
The	O
EVH	O
can	O
further	O
assist	O
in	O
the	O
process	O
of	O
finding	O
suitable	O
objectives	O
and	O
balancing	O
conflicting	O
objectives	O
.	O
If	O
biologically	O
inspired	O
objectives	O
are	O
used	O
,	O
the	O
EVH	O
shows	O
the	O
distribution	O
of	O
local	O
dose	O
effect	O
relative	O
to	O
the	O
prescribed	O
level	O

Groove	B-KEY
Networks	I-KEY
.	O
Matching	O
technology	O
with	O
human	O
needs	O
If	O
what	O
has	O
been	O
occurring	O
in	O
information	O
technology	O
during	O
the	O
past	O
decade	O
or	O
so	O
can	O
be	O
classified	O
as	O
the	O
``	O
Information	O
Age	O
,	O
''	O
then	O
going	O
forward	O
,	O
I	O
believe	O
it	O
's	O
going	O
to	O
be	O
viewed	O
more	O
as	O
the	O
``	O
connection	O
age	O
''	O
,	O
''	O
says	O
Ray	O
Ozzie	O
,	O
CEO	O
and	O
chairman	O
of	O
Groove	B-KEY
Networks	I-KEY
,	O
the	O
Beverly	O
,	O
Massachusetts	O
company	O
that	O
produces	O
collaboration	O
software	O
.	O
``	O
We	O
're	O
all	O
going	O
to	O
be	O
thinking	O
more	O
about	O
the	O
connections	O
between	O
people	O
and	O
the	O
connections	O
between	O
companies	O
,	O
''	O
Ozzie	O
says	O
.	O
``	O
Our	O
mission	O
has	O
two	O
parts	O
:	O
to	O
help	O
businesses	B-KEY
achieve	O
a	O
greater	O
``	O
return	O
on	O
connection	O
''	O
from	O
their	O
relationships	O
with	O
customers	O
,	O
vendors	O
,	O
and	O
partners	O
;	O
and	O
to	O
help	O
individuals	O
strengthen	O
online	B-KEY
connections	I-KEY
with	O
the	O
people	O
with	O
whom	O
they	O
interact	O
.	O
''	O

Delayed-choice	B-KEY
entanglement	I-KEY
swapping	O
with	O
vacuum-one-photon	B-KEY
quantum	I-KEY
states	I-KEY
We	O
report	O
the	O
experimental	O
realization	O
of	O
a	O
recently	O
discovered	O
quantum-information	B-KEY
protocol	O
by	O
Peres	O
implying	O
an	O
apparent	O
nonlocal	B-KEY
quantum	I-KEY
mechanical	I-KEY
retrodiction	I-KEY
effect	I-KEY
.	O
The	O
demonstration	O
is	O
carried	O
out	O
by	O
a	O
quantum	B-KEY
optical	I-KEY
method	I-KEY
by	O
which	O
each	O
singlet	B-KEY
entangled	I-KEY
state	I-KEY
is	O
physically	O
implemented	O
by	O
a	O
two-dimensional	B-KEY
subspace	I-KEY
of	O
Fock	B-KEY
states	I-KEY
of	O
a	O
mode	O
of	O
the	O
electromagnetic	O
field	O
,	O
specifically	O
the	O
space	O
spanned	O
by	O
the	O
vacuum	O
and	O
the	O
one-photon	B-KEY
state	I-KEY
,	O
along	O
lines	O
suggested	O
recently	O
by	O
E.	O
Knill	O
et	O
al.	O
-LSB-	O
Nature	O
-LRB-	O
London	O
-RRB-	O
409	O
,	O
46	O
-LRB-	O
2001	O
-RRB-	O
-RSB-	O
and	O
by	O
M.	O
Duan	O
et	O
al.	O
-LSB-	O
ibid	O
.	O
414	O
,	O
413	O
-LRB-	O
2001	O
-RRB-	O
-RSB-	O

Coarse-grained	B-KEY
reduction	O
and	O
analysis	O
of	O
a	O
network	O
model	O
of	O
cortical	O
response	O
:	O
I.	O
Drifting	O
grating	O
stimuli	O
We	O
present	O
a	O
reduction	O
of	O
a	O
large-scale	B-KEY
network	I-KEY
model	I-KEY
of	O
visual	B-KEY
cortex	I-KEY
developed	O
by	O
McLaughlin	O
,	O
Shapley	O
,	O
Shelley	O
,	O
and	O
Wielaard	O
.	O
The	O
reduction	O
is	O
from	O
many	O
integrate-and-fire	O
neurons	O
to	O
a	O
spatially	O
coarse-grained	B-KEY
system	O
for	O
firing	O
rates	O
of	O
neuronal	O
subpopulations	O
.	O
It	O
accounts	O
explicitly	O
for	O
spatially	O
varying	O
architecture	O
,	O
ordered	O
cortical	O
maps	O
-LRB-	O
such	O
as	O
orientation	O
preference	O
-RRB-	O
that	O
vary	O
regularly	O
across	O
the	O
cortical	O
layer	O
,	O
and	O
disordered	O
cortical	O
maps	O
-LRB-	O
such	O
as	O
spatial	O
phase	O
preference	O
or	O
stochastic	O
input	O
conductances	O
-RRB-	O
that	O
may	O
vary	O
widely	O
from	O
cortical	O
neuron	O
to	O
cortical	O
neuron	O
.	O
The	O
result	O
of	O
the	O
reduction	O
is	O
a	O
set	O
of	O
nonlinear	B-KEY
spatiotemporal	I-KEY
integral	I-KEY
equations	I-KEY
for	O
``	O
phase-averaged	O
''	O
firing	O
rates	O
of	O
neuronal	O
subpopulations	O
across	O
the	O
model	O
cortex	O
,	O
derived	O
asymptotically	O
from	O
the	O
full	O
model	O
without	O
the	O
addition	O
of	O
any	O
extra	O
phenomological	O
constants	O
.	O
This	O
reduced	O
system	O
is	O
used	O
to	O
study	O
the	O
response	O
of	O
the	O
model	O
to	O
drifting	O
grating	O
stimuli	O
-	O
where	O
it	O
is	O
shown	O
to	O
be	O
useful	O
for	O
numerical	O
investigations	O
that	O
reproduce	O
,	O
at	O
far	O
less	O
computational	O
cost	O
,	O
the	O
salient	O
features	O
of	O
the	O
point-neuron	B-KEY
network	I-KEY
and	O
for	O
analytical	O
investigations	O
that	O
unveil	O
cortical	O
mechanisms	O
behind	O
the	O
responses	O
observed	O
in	O
the	O
simulations	O
of	O
the	O
large-scale	O
computational	O
model	O
.	O
For	O
example	O
,	O
the	O
reduced	O
equations	O
clearly	O
show	O
-LRB-	O
1	O
-RRB-	O
phase	O
averaging	O
as	O
the	O
source	O
of	O
the	O
time-invariance	O
of	O
cortico-cortical	O
conductances	O
,	O
-LRB-	O
2	O
-RRB-	O
the	O
mechanisms	O
in	O
the	O
model	O
for	O
higher	O
firing	O
rates	O
and	O
better	O
orientation	B-KEY
selectivity	I-KEY
of	O
simple	O
cells	O
which	O
are	O
near	O
pinwheel	O
centers	O
,	O
-LRB-	O
3	O
-RRB-	O
the	O
effects	O
of	O
the	O
length-scales	O
of	O
cortico-cortical	O
coupling	O
,	O
and	O
-LRB-	O
4	O
-RRB-	O
the	O
role	O
of	O
noise	O
in	O
improving	O
the	O
contrast	O
invariance	O
of	O
orientation	B-KEY
selectivity	I-KEY

WebCAD	B-KEY
:	O
A	O
computer	B-KEY
aided	I-KEY
design	I-KEY
tool	I-KEY
constrained	O
with	O
explicit	O
`	O
design	O
for	O
manufacturability	O
'	O
rules	O
for	O
computer	O
numerical	O
control	O
milling	O
A	O
key	O
element	O
in	O
the	O
overall	O
efficiency	O
of	O
a	O
manufacturing	O
enterprise	O
is	O
the	O
compatibility	O
between	O
the	O
features	O
that	O
have	O
been	O
created	O
in	O
a	O
newly	O
designed	O
part	O
,	O
and	O
the	O
capabilities	O
of	O
the	O
downstream	O
manufacturing	O
processes	O
.	O
With	O
this	O
in	O
mind	O
,	O
a	O
process-aware	O
computer	O
aided	O
design	O
-LRB-	O
CAD	O
-RRB-	O
system	O
called	O
WebCAD	B-KEY
has	O
been	O
developed	O
.	O
The	O
system	O
restricts	O
the	O
freedom	O
of	O
the	O
designer	O
in	O
such	O
a	O
way	O
that	O
the	O
designed	O
parts	O
can	O
be	O
manufactured	O
on	O
a	O
three-axis	O
computer	B-KEY
numerical	I-KEY
control	I-KEY
milling	I-KEY
machine	O
.	O
This	O
paper	O
discusses	O
the	O
vision	O
of	O
WebCAD	B-KEY
and	O
explains	O
the	O
rationale	O
for	O
its	O
development	O
in	O
comparison	O
with	O
commercial	O
CAD/CAM	O
-LRB-	O
computer	O
aided	O
design/manufacture	O
-RRB-	O
systems	O
.	O
The	O
paper	O
then	O
goes	O
on	O
to	O
describe	O
the	O
implementation	O
issues	O
that	O
enforce	O
the	O
manufacturability	B-KEY
rules	I-KEY
.	O
Finally	O
,	O
certain	O
design	B-KEY
tools	I-KEY
are	O
described	O
that	O
aid	O
a	O
user	O
during	O
the	O
design	O
process	O
.	O
Some	O
examples	O
are	O
given	O
of	O
the	O
parts	O
designed	O
and	O
manufactured	O
with	O
WebCAD	B-KEY

Quantum	B-KEY
Zeno	I-KEY
subspaces	I-KEY
The	O
quantum	O
Zeno	O
effect	O
is	O
recast	O
in	O
terms	O
of	O
an	O
adiabatic	B-KEY
theorem	I-KEY
when	O
the	O
measurement	B-KEY
is	O
described	O
as	O
the	O
dynamical	B-KEY
coupling	I-KEY
to	O
another	O
quantum	O
system	O
that	O
plays	O
the	O
role	O
of	O
apparatus	O
.	O
A	O
few	O
significant	O
examples	O
are	O
proposed	O
and	O
their	O
practical	O
relevance	O
discussed	O
.	O
We	O
also	O
focus	O
on	O
decoherence-free	B-KEY
subspaces	I-KEY

An	O
analytic	B-KEY
center	I-KEY
cutting	I-KEY
plane	I-KEY
method	I-KEY
for	O
semidefinite	B-KEY
feasibility	I-KEY
problems	I-KEY
Semidefinite	B-KEY
feasibility	I-KEY
problems	I-KEY
arise	O
in	O
many	O
areas	O
of	O
operations	B-KEY
research	I-KEY
.	O
The	O
abstract	O
form	O
of	O
these	O
problems	O
can	O
be	O
described	O
as	O
finding	O
a	O
point	O
in	O
a	O
nonempty	B-KEY
bounded	I-KEY
convex	I-KEY
body	I-KEY
Gamma	O
in	O
the	O
cone	O
of	O
symmetric	B-KEY
positive	I-KEY
semidefinite	I-KEY
matrices	I-KEY
.	O
Assume	O
that	O
Gamma	O
is	O
defined	O
by	O
an	O
oracle	B-KEY
,	O
which	O
for	O
any	O
given	O
m	O
*	O
m	O
symmetric	O
positive	O
semidefinite	O
matrix	O
Gamma	O
either	O
confirms	O
that	O
Y	O
epsilon	O
Gamma	O
or	O
returns	O
a	O
cut	O
,	O
i.e.	O
,	O
a	O
symmetric	O
matrix	O
A	O
such	O
that	O
Gamma	O
is	O
in	O
the	O
half-space	O
-LCB-	O
Y	O
:	O
A	O
.	O
Y	O
<	O
or	O
=	O
A	O
.	O
Y	O
-RCB-	O
.	O
We	O
study	O
an	O
analytic	O
center	O
cutting	O
plane	O
algorithm	O
for	O
this	O
problem	O
.	O
At	O
each	O
iteration	B-KEY
,	O
the	O
algorithm	O
computes	O
an	O
approximate	B-KEY
analytic	I-KEY
center	I-KEY
of	O
a	O
working	B-KEY
set	I-KEY
defined	O
by	O
the	O
cutting	O
plane	O
system	O
generated	O
in	O
the	O
previous	O
iterations	B-KEY
.	O
If	O
this	O
approximate	B-KEY
analytic	I-KEY
center	I-KEY
is	O
a	O
solution	O
,	O
then	O
the	O
algorithm	O
terminates	O
;	O
otherwise	O
the	O
new	O
cutting	O
plane	O
returned	O
by	O
the	O
oracle	B-KEY
is	O
added	O
into	O
the	O
system	O
.	O
As	O
the	O
number	O
of	O
iterations	B-KEY
increases	O
,	O
the	O
working	B-KEY
set	I-KEY
shrinks	O
and	O
the	O
algorithm	O
eventually	O
finds	O
a	O
solution	O
to	O
the	O
problem	O
.	O
All	O
iterates	B-KEY
generated	O
by	O
the	O
algorithm	O
are	O
positive	O
definite	O
matrices	O
.	O
The	O
algorithm	O
has	O
a	O
worst-case	B-KEY
complexity	I-KEY
of	O
O	O
*	O
-LRB-	O
m/sup	O
3	O
/	O
/	O
epsilon	O
/	O
sup	O
2	O
/	O
-RRB-	O
on	O
the	O
total	O
number	O
of	O
cuts	O
to	O
be	O
used	O
,	O
where	O
epsilon	O
is	O
the	O
maximum	O
radius	O
of	O
a	O
ball	O
contained	O
by	O
Gamma	O

Active	B-KEY
pitch	I-KEY
control	I-KEY
in	O
larger	O
scale	O
fixed	O
speed	O
horizontal	O
axis	O
wind	B-KEY
turbine	I-KEY
systems	O
.	O
I.	O
linear	B-KEY
controller	I-KEY
design	I-KEY
This	O
paper	O
reviews	O
and	O
addresses	O
the	O
principles	O
of	O
linear	B-KEY
controller	I-KEY
design	I-KEY
of	O
the	O
fixed	O
speed	O
wind	B-KEY
turbine	I-KEY
system	O
in	O
above	O
rated	O
wind	O
speed	O
,	O
using	O
pitch	O
angle	O
control	O
of	O
the	O
blades	O
and	O
applying	O
modern	O
control	O
theory	O
.	O
First	O
,	O
the	O
nonlinear	B-KEY
equations	I-KEY
of	O
the	O
system	O
are	O
built	O
in	O
under	O
some	O
reasonable	O
suppositions	O
.	O
Then	O
,	O
the	O
nonlinear	B-KEY
equations	I-KEY
are	O
linearised	O
at	O
set	O
operating	O
point	O
and	O
digital	B-KEY
simulation	I-KEY
results	O
are	O
shown	O
in	O
this	O
paper	O
.	O
Finally	O
,	O
a	O
linear	B-KEY
quadratic	I-KEY
optimal	I-KEY
feedback	I-KEY
controller	I-KEY
is	O
designed	O
and	O
the	O
dynamics	O
of	O
the	O
closed	B-KEY
circle	I-KEY
system	I-KEY
are	O
simulated	O
with	O
digital	O
calculation	O
.	O
The	O
advantages	O
and	O
disadvantages	O
of	O
the	O
assumptions	O
and	O
design	O
method	O
are	O
also	O
discussed	O
.	O
Because	O
of	O
the	O
inherent	O
characteristics	O
of	O
the	O
linear	B-KEY
system	I-KEY
control	I-KEY
theory	I-KEY
,	O
the	O
performance	O
of	O
the	O
linear	O
controller	O
is	O
not	O
sufficient	O
for	O
operating	O
wind	O
turbines	O
,	O
as	O
is	O
discussed	O

A	O
look	O
at	O
MonacoProfiler	B-KEY
4	I-KEY
The	O
newest	O
profiling	O
program	O
from	O
Monaco	O
Software	O
adds	O
some	O
valuable	O
features	O
:	O
support	O
for	O
up	O
to	O
8-color	O
printing	O
,	O
profiling	O
for	O
digital	O
cameras	O
,	O
fine-tuning	O
of	O
black	O
generation	O
and	O
tweaking	O
of	O
profile	O
transforms	O
.	O
We	O
tested	O
its	O
ease	O
of	O
use	O
and	O
a	O
few	O
of	O
the	O
advanced	O
functions	O
.	O
In	O
all	O
,	O
it	O
's	O
pretty	O
good	O

Mothball	B-KEY
mania	O
-LSB-	O
3G	O
licences	O
-RSB-	O
Telefonica	O
Moviles	O
has	O
frozen	O
its	O
3G	O
operations	O
in	O
Germany	O
,	O
Austria	O
,	O
Italy	O
and	O
Switzerland	O
.	O
With	O
other	O
3G	B-KEY
licence	I-KEY
holders	I-KEY
questioning	O
the	O
logic	O
of	O
entering	O
already	O
saturated	B-KEY
markets	I-KEY
with	O
unproven	O
technology	O
,	O
Emma	O
McClune	O
asks	O
if	O
the	O
mothball	B-KEY
effect	O
is	O
set	O
to	O
snowball	O
any	O
further	O

Topology-adaptive	B-KEY
modeling	I-KEY
of	O
objects	O
using	O
surface	B-KEY
evolutions	I-KEY
based	O
on	O
3D	B-KEY
mathematical	I-KEY
morphology	I-KEY
Level	B-KEY
set	I-KEY
methods	I-KEY
were	O
proposed	O
mainly	O
by	O
mathematicians	O
for	O
constructing	O
a	O
model	O
of	O
a	O
3D	B-KEY
object	I-KEY
of	O
arbitrary	B-KEY
topology	I-KEY
.	O
However	O
,	O
those	O
methods	O
are	O
computationally	O
inefficient	O
due	O
to	O
repeated	B-KEY
distance	I-KEY
transformations	I-KEY
and	O
increased	O
dimensions	O
.	O
In	O
the	O
paper	O
,	O
we	O
propose	O
a	O
new	O
method	O
of	O
modeling	O
fast	O
objects	O
of	O
arbitrary	B-KEY
topology	I-KEY
by	O
using	O
a	O
surface	B-KEY
evolution	I-KEY
approach	O
based	O
on	O
mathematical	O
morphology	O
.	O
Given	O
sensor	O
data	O
covering	O
the	O
whole	O
object	O
surface	O
,	O
the	O
method	O
begins	O
with	O
an	O
initial	B-KEY
approximation	I-KEY
of	O
the	O
object	O
by	O
evolving	O
a	O
closed	O
surface	O
into	O
a	O
model	O
topologically	O
equivalent	O
to	O
the	O
real	O
object	O
.	O
The	O
refined	B-KEY
approximation	I-KEY
is	O
then	O
performed	O
using	O
energy	B-KEY
minimization	I-KEY
.	O
The	O
method	O
has	O
been	O
applied	O
in	O
several	O
experiments	O
using	O
range	B-KEY
data	I-KEY
,	O
and	O
the	O
results	O
are	O
reported	O
in	O
the	O
paper	O

SRP	O
rolls	O
out	O
reliability	O
and	O
asset	O
management	O
initiative	O
Reliability	B-KEY
planning	I-KEY
analysis	I-KEY
at	O
the	O
Salt	B-KEY
River	I-KEY
Project	I-KEY
-LRB-	O
SRP	O
,	O
Tempe	B-KEY
,	O
Arizona	B-KEY
,	O
US	O
-RRB-	O
prioritizes	O
geographic	B-KEY
areas	I-KEY
for	O
preventive	B-KEY
inspections	I-KEY
based	O
on	O
a	O
cost	B-KEY
benefit	I-KEY
model	I-KEY
.	O
However	O
,	O
SRP	O
wanted	O
a	O
new	O
application	O
system	O
to	O
prioritize	O
inspections	O
and	O
to	O
predict	O
when	O
direct	B-KEY
buried	I-KEY
cable	I-KEY
would	O
fail	O
using	O
the	O
same	O
cost	B-KEY
benefit	I-KEY
model	I-KEY
.	O
In	O
the	O
business	O
cases	O
,	O
the	O
represented	O
type	O
of	O
kilowatt	O
load-residential	O
,	O
commercial	O
or	O
critical	O
circuit-determines	O
the	O
cost	O
benefit	O
per	O
circuit	O
.	O
The	O
preferred	O
solution	O
was	O
to	O
develop	O
a	O
geographical	B-KEY
information	I-KEY
system	I-KEY
-LRB-	O
GIS	B-KEY
-RRB-	O
application	O
allowing	O
for	O
a	O
circuit	O
query	O
for	O
the	O
specific	O
geographic	B-KEY
areas	I-KEY
it	O
crosses	O
and	O
the	O
density	O
of	O
load	O
points	O
of	O
a	O
given	O
type	O
within	O
those	O
areas	O
.	O
The	O
query	O
returns	O
results	O
based	O
on	O
the	O
type	O
of	O
equipment	B-KEY
analysis	I-KEY
execution	I-KEY
:	O
wood	B-KEY
pole	I-KEY
,	O
preventive	O
maintenance	O
for	O
a	O
line	O
or	O
cable	B-KEY
replacement	I-KEY
.	O
This	O
differentiation	O
insures	O
that	O
all	O
the	O
facilities	O
relevant	O
to	O
a	O
specific	O
analysis	O
type	O
influence	O
prioritization	O
of	O
the	O
geographic	B-KEY
areas	I-KEY

Benchmarking	B-KEY
of	O
the	O
Dose	O
Planning	O
Method	O
-LRB-	O
DPM	O
-RRB-	O
Monte	O
Carlo	O
code	O
using	O
electron	O
beams	O
from	O
a	O
racetrack	B-KEY
microtron	I-KEY
A	O
comprehensive	O
set	O
of	O
measurements	O
and	O
calculations	O
has	O
been	O
conducted	O
to	O
investigate	O
the	O
accuracy	O
of	O
the	O
Dose	O
Planning	O
Method	O
-LRB-	O
DPM	O
-RRB-	O
Monte	O
Carlo	O
code	O
for	O
dose	O
calculations	O
from	O
10	O
and	O
50	B-KEY
MeV	I-KEY
scanned	B-KEY
electron	I-KEY
beams	I-KEY
produced	O
from	O
a	O
racetrack	B-KEY
microtron	I-KEY
.	O
Central	B-KEY
axis	I-KEY
depth	I-KEY
dose	I-KEY
measurements	I-KEY
and	O
a	O
series	O
of	O
profile	B-KEY
scans	I-KEY
at	O
various	O
depths	O
were	O
acquired	O
in	O
a	O
water	B-KEY
phantom	I-KEY
using	O
a	O
Scanditronix	O
type	O
RK	O
ion	B-KEY
chamber	I-KEY
.	O
Source	B-KEY
spatial	I-KEY
distributions	I-KEY
for	O
the	O
Monte	O
Carlo	O
calculations	O
were	O
reconstructed	O
from	O
in-air	O
ion	B-KEY
chamber	I-KEY
measurements	O
carried	O
out	O
across	O
the	O
two-dimensional	B-KEY
beam	I-KEY
profile	I-KEY
at	O
100	O
cm	O
downstream	O
from	O
the	O
source	O
.	O
The	O
in-air	B-KEY
spatial	I-KEY
distributions	I-KEY
were	O
found	O
to	O
have	O
full	O
width	O
at	O
half	O
maximum	O
of	O
4.7	O
and	O
1.3	O
cm	O
,	O
at	O
100	O
cm	O
from	O
the	O
source	O
,	O
for	O
the	O
10	O
and	O
50	B-KEY
MeV	I-KEY
beams	O
,	O
respectively	O
.	O
Energy	O
spectra	O
for	O
the	O
10	O
and	O
50	B-KEY
MeV	I-KEY
beams	O
were	O
determined	O
by	O
simulating	O
the	O
components	O
of	O
the	O
microtron	O
treatment	O
head	O
using	O
the	O
code	O
MCNP4B	B-KEY
.	O
DPM	O
calculations	O
are	O
on	O
average	O
within	O
+	O
or-2	O
%	O
agreement	O
with	O
measurement	O
for	O
all	O
depth	O
dose	O
and	O
profile	O
comparisons	O
conducted	O
in	O
this	O
study	O
.	O
The	O
accuracy	O
of	O
the	O
DPM	O
code	O
illustrated	O
in	O
this	O
work	O
suggests	O
that	O
DPM	O
may	O
be	O
used	O
as	O
a	O
valuable	O
tool	O
for	O
electron	B-KEY
beam	I-KEY
dose	I-KEY
calculations	I-KEY

Establishing	O
the	O
discipline	O
of	O
physics-based	O
CMP	B-KEY
modeling	O
For	O
the	O
past	O
decade	O
,	O
a	O
physically	O
based	O
comprehensive	O
process	O
model	O
for	O
chemical	B-KEY
mechanical	I-KEY
polishing	I-KEY
has	O
eluded	O
the	O
semiconductor	O
industry	O
.	O
However	O
,	O
a	O
long-term	O
collaborative	O
effort	O
has	O
now	O
resulted	O
in	O
a	O
workable	O
version	O
of	O
that	O
approach	O
.	O
The	O
highly	O
fundamental	O
model	O
is	O
based	O
on	O
advanced	O
finite	B-KEY
element	I-KEY
analysis	I-KEY
and	O
is	O
beginning	O
to	O
show	O
promise	O
in	O
CMP	B-KEY
process	O
development	O

Algorithmic	B-KEY
results	I-KEY
for	O
ordered	B-KEY
median	I-KEY
problems	I-KEY
In	O
a	O
series	O
of	O
papers	O
a	O
new	O
type	O
of	O
objective	B-KEY
function	I-KEY
in	O
location	B-KEY
theory	I-KEY
,	O
called	O
ordered	B-KEY
median	I-KEY
function	I-KEY
,	O
has	O
been	O
introduced	O
and	O
analyzed	O
.	O
This	O
objective	B-KEY
function	I-KEY
unifies	O
and	O
generalizes	O
most	O
common	O
objective	B-KEY
functions	I-KEY
used	O
in	O
location	B-KEY
theory	I-KEY
.	O
In	O
this	O
paper	O
we	O
identify	O
finite	B-KEY
dominating	I-KEY
sets	I-KEY
for	O
these	O
models	O
and	O
develop	O
polynomial	B-KEY
time	I-KEY
algorithms	I-KEY
together	O
with	O
a	O
detailed	B-KEY
complexity	I-KEY
analysis	I-KEY

Agent-based	B-KEY
product-support	I-KEY
logistics	I-KEY
system	I-KEY
using	O
XML	B-KEY
and	O
RDF	B-KEY
The	O
capability	O
of	O
the	O
timely	O
provision	O
of	O
maintenance	B-KEY
services	I-KEY
and	O
service	B-KEY
parts	I-KEY
is	O
critical	O
to	O
the	O
competitiveness	O
of	O
industrial	B-KEY
systems	I-KEY
.	O
To	O
enhance	O
the	O
timely	O
operations	O
in	O
a	O
product-support	O
logistics	O
chain	O
,	O
business	O
partners	O
-LRB-	O
equipment	O
manufacturers	O
,	O
parts	O
distributors	O
,	O
customers	O
-RRB-	O
may	O
have	O
to	O
collaborate	O
for	O
the	O
efficient	O
exchange	O
of	O
relevant	O
information	O
.	O
We	O
propose	O
the	O
architecture	O
of	O
an	O
agent-based	B-KEY
product-support	I-KEY
logistics	I-KEY
system	I-KEY
.	O
Emphasis	O
is	O
placed	O
on	O
the	O
problems	O
of	O
sharing	O
and	O
exchanging	O
information	O
through	O
agent	O
communication	O
.	O
We	O
adopt	O
the	O
Resource	O
Description	O
Framework	O
-LRB-	O
RDF	B-KEY
-RRB-	O
schema	O
for	O
information	B-KEY
modelling	I-KEY
in	O
product-support	O
logistics	O
domain	O
.	O
The	O
eXtensible	B-KEY
Markup	I-KEY
Language	I-KEY
-LRB-	O
XML	B-KEY
-RRB-	O
serialization	O
generates	O
messages	O
for	O
agent	O
communication	O
.	O
The	O
use	O
of	O
XML	B-KEY
and	O
RDF	B-KEY
enables	O
software	B-KEY
agents	I-KEY
to	O
understand	O
the	O
contents	O
of	O
messages	O
correctly	O
and	O
consistently	O
.	O
We	O
demonstrate	O
the	O
feasibility	O
of	O
our	O
agent	O
architecture	O
using	O
a	O
scenario	O
in	O
logistical	O
support	O
processes	O
.	O
We	O
believe	O
that	O
the	O
approach	O
can	O
provide	O
a	O
promising	O
way	O
to	O
the	O
automation	O
of	O
business	O
processes	O
in	O
product-support	O
logistics	O
through	O
seamless	O
communication	O
among	O
the	O
partners	O

Indexing-neglected	O
and	O
poorly	O
understood	O
The	O
growth	O
of	O
the	O
Internet	B-KEY
has	O
highlighted	O
the	O
use	O
of	O
machine	B-KEY
indexing	I-KEY
.	O
The	O
difficulties	O
in	O
using	O
the	O
Internet	B-KEY
as	O
a	O
searching	B-KEY
device	O
can	O
be	O
frustrating	O
.	O
The	O
use	O
of	O
the	O
term	O
``	O
python	O
''	O
is	O
given	O
as	O
an	O
example	O
.	O
Machine	B-KEY
indexing	I-KEY
is	O
noted	O
as	O
``	O
rotten	O
''	O
and	O
human	B-KEY
indexing	I-KEY
as	O
``	O
capricious	O
.	O
''	O
The	O
problem	O
seems	O
to	O
be	O
a	O
lack	O
of	O
a	O
theoretical	O
foundation	O
for	O
the	O
art	O
of	O
indexing	O
.	O
What	O
librarians	O
have	O
learned	O
over	O
the	O
last	O
hundred	O
years	O
has	O
yet	O
to	O
yield	O
a	O
consistent	O
approach	O
to	O
what	O
really	O
works	O
best	O
in	O
preparing	O
index	B-KEY
terms	I-KEY
and	O
in	O
the	O
ability	O
of	O
our	O
customers	O
to	O
search	B-KEY
the	O
various	O
indexes	O
.	O
An	O
attempt	O
is	O
made	O
to	O
consider	O
the	O
elements	O
of	O
indexing	O
,	O
their	O
pros	O
and	O
cons	O
.	O
The	O
argument	O
is	O
made	O
that	O
machine	B-KEY
indexing	I-KEY
is	O
far	O
too	O
prolific	O
in	O
its	O
production	O
of	O
index	B-KEY
terms	I-KEY
.	O
Neither	O
librarians	O
nor	O
computer	O
programmers	O
have	O
made	O
much	O
progress	O
to	O
improve	O
Internet	B-KEY
indexing	O
.	O
Human	B-KEY
indexing	I-KEY
has	O
had	O
the	O
same	O
problems	O
for	O
over	O
fifty	O
years	O

Compatibility	O
comparison	O
and	O
performance	O
evaluation	O
for	O
Japanese	O
HPF	B-KEY
compilers	B-KEY
using	O
scientific	O
applications	O
The	O
lack	O
of	O
compatibility	O
of	O
High-Performance	B-KEY
Fortran	I-KEY
-LRB-	O
HPF	B-KEY
-RRB-	O
between	O
vender	O
implementations	O
has	O
been	O
disheartening	O
scientific	O
application	O
users	O
so	O
as	O
to	O
hinder	O
the	O
development	O
of	O
portable	B-KEY
programs	I-KEY
.	O
Thus	O
parallel	O
computing	O
is	O
still	O
unpopular	O
in	O
the	O
computational	O
science	O
community	O
,	O
even	O
though	O
parallel	B-KEY
programming	I-KEY
is	O
common	O
to	O
the	O
computer	O
science	O
community	O
.	O
As	O
users	O
would	O
like	O
to	O
run	O
the	O
same	O
source	O
code	O
on	O
parallel	O
machines	O
with	O
different	O
architectures	O
as	O
fast	O
as	O
possible	O
,	O
we	O
have	O
investigated	O
the	O
compatibility	O
of	O
source	O
codes	O
for	O
Japanese	O
HPF	B-KEY
compilers	B-KEY
-LRB-	O
NEC	O
,	O
Fujitsu	O
and	O
Hitachi	O
-RRB-	O
with	O
two	O
real-world	O
applications	O
:	O
a	O
3D	O
fluid	O
code	O
and	O
a	O
2D	O
particle	O
code	O
.	O
We	O
have	O
found	O
that	O
the	O
source-level	O
compatibility	O
between	O
Japanese	O
HPF	B-KEY
compilers	B-KEY
is	O
almost	O
preserved	O
,	O
but	O
more	O
effort	O
will	O
be	O
needed	O
to	O
sustain	O
complete	O
compatibility	O
.	O
We	O
have	O
also	O
evaluated	O
parallel	B-KEY
performance	I-KEY
and	O
found	O
that	O
HPF	B-KEY
can	O
achieve	O
good	O
performance	O
for	O
the	O
3D	O
fluid	O
code	O
with	O
almost	O
the	O
same	O
source	O
code	O
.	O
For	O
the	O
2D	O
particle	O
code	O
,	O
good	O
results	O
have	O
also	O
been	O
obtained	O
with	O
a	O
small	O
number	O
of	O
processors	O
,	O
but	O
some	O
changes	O
in	O
the	O
original	O
source	O
code	O
and	O
the	O
addition	O
of	O
interface	O
blocks	O
is	O
required	O

Santera	O
targets	O
independents	O
in	O
major	O
strategy	O
overhaul	O
-LSB-	O
telecom	O
-RSB-	O
With	O
big	O
carriers	O
slashing	O
capital	O
expense	O
budgets	O
,	O
Santera	B-KEY
Systems	I-KEY
is	O
broadening	O
the	O
reach	O
of	O
its	O
next-generation	O
switching	B-KEY
platform	O
to	O
include	O
independent	O
telcos	O
.	O
This	O
week	O
,	O
the	O
vendor	O
will	O
announce	O
that	O
it	O
has	O
signed	O
a	O
deal	O
with	O
Kerman	O
,	O
Calif-based	O
Kerman	B-KEY
Telephone	I-KEY
Co.	O
.	O
Furthermore	O
,	O
the	O
company	O
is	O
angling	O
for	O
inclusion	O
in	O
the	O
Rural	B-KEY
Utilities	I-KEY
Service	I-KEY
's	O
approved	O
equipment	O
list	O
,	O
hoping	O
to	O
sell	O
its	O
Class	O
5	O
replacement	O
boxes	O
to	O
the	O
smallest	O
carriers	O
.	O
The	O
move	O
is	O
almost	O
a	O
complete	O
reversal	O
for	O
the	O
Plano	O
,	O
Texas-based	O
vendor	O
,	O
which	O
previously	O
focused	O
solely	O
on	O
large	O
carriers	O
,	O
including	O
the	O
RBOCs	O

Limitations	O
of	O
delayed	B-KEY
state	I-KEY
feedback	I-KEY
:	O
a	O
numerical	O
study	O
Stabilization	O
of	O
a	O
class	O
of	O
linear	B-KEY
time-delay	I-KEY
systems	I-KEY
can	O
be	O
achieved	O
by	O
a	O
numerical	O
procedure	O
,	O
called	O
the	O
continuous	B-KEY
pole	I-KEY
placement	I-KEY
method	I-KEY
-LSB-	O
Michiels	O
et	O
al.	O
,	O
2000	O
-RSB-	O
.	O
This	O
method	O
can	O
be	O
seen	O
as	O
an	O
extension	O
of	O
the	O
classical	O
pole	O
placement	O
algorithm	O
for	O
ordinary	O
differential	O
equations	O
to	O
a	O
class	O
of	O
delay	B-KEY
differential	I-KEY
equations	I-KEY
.	O
In	O
-LSB-	O
Michiels	O
et	O
al.	O
,	O
2000	O
-RSB-	O
it	O
was	O
applied	O
to	O
the	O
stabilization	O
of	O
a	O
linear	O
time-invariant	O
system	O
with	O
an	O
input	O
delay	O
using	O
static	B-KEY
state	I-KEY
feedback	I-KEY
.	O
In	O
this	O
paper	O
we	O
study	O
the	O
limitations	O
of	O
such	O
delayed	B-KEY
state	I-KEY
feedback	I-KEY
laws	O
.	O
More	O
precisely	O
we	O
completely	O
characterize	O
the	O
class	O
of	O
stabilizable	O
plants	O
in	O
the	O
2D-case	O
.	O
For	O
that	O
purpose	O
we	O
make	O
use	O
of	O
numerical	B-KEY
continuation	I-KEY
techniques	O
.	O
The	O
use	O
of	O
delayed	B-KEY
state	I-KEY
feedback	I-KEY
in	O
various	O
control	O
applications	O
and	O
the	O
effect	O
of	O
its	O
limitations	O
are	O
briefly	O
discussed	O

A	O
three-tier	B-KEY
technology	I-KEY
training	I-KEY
strategy	I-KEY
in	O
a	O
dynamic	B-KEY
business	I-KEY
environment	I-KEY
As	O
end-user	B-KEY
training	I-KEY
becomes	O
increasingly	O
important	O
in	O
today	O
's	O
technology-intensive	B-KEY
business	I-KEY
environment	I-KEY
,	O
progressive	O
companies	B-KEY
remain	O
alert	O
to	O
find	O
ways	O
to	O
provide	O
their	O
end	O
users	O
with	O
timely	O
training	O
and	O
resources	O
.	O
This	O
paper	O
describes	O
an	O
innovative	B-KEY
training	I-KEY
strategy	I-KEY
adopted	O
by	O
one	O
midsize	B-KEY
organization	I-KEY
to	O
provide	O
its	O
end	O
users	O
with	O
adequate	O
,	O
flexible	O
,	O
and	O
responsive	O
training	O
.	O
The	O
paper	O
then	O
compares	O
the	O
three-tier	O
strategy	O
with	O
other	O
models	O
described	O
in	O
technology	O
training	O
literature	O
.	O
Managers	O
who	O
supervise	O
technology	O
end	O
users	O
in	O
organizations	B-KEY
comparable	O
to	O
the	O
one	O
in	O
the	O
study	O
may	O
find	O
the	O
three-tier	O
strategy	O
workable	O
and	O
may	O
want	O
to	O
use	O
it	O
in	O
their	O
own	O
training	O
programs	O
to	O
facilitate	O
training	O
and	O
improve	O
end-user	O
skills	O
.	O
Researchers	O
and	O
scholars	O
may	O
find	O
that	O
the	O
idea	O
of	O
three-tier	O
training	O
generates	O
new	O
opportunities	O
for	O
research	O

Six	O
common	O
enterprise	B-KEY
programming	I-KEY
mistakes	I-KEY
Instead	O
of	O
giving	O
you	O
tips	O
to	O
use	O
in	O
your	O
programming	O
-LRB-	O
at	O
least	O
directly	O
-RRB-	O
,	O
I	O
want	O
to	O
look	O
at	O
some	O
common	O
mistakes	O
made	O
in	O
enterprise	O
programming	O
.	O
Instead	O
of	O
focusing	O
on	O
what	O
to	O
do	O
,	O
I	O
want	O
to	O
look	O
at	O
what	O
you	O
should	O
not	O
do	O
.	O
Most	O
programmers	O
take	O
books	O
like	O
mine	O
and	O
add	O
in	O
the	O
good	O
things	O
,	O
but	O
they	O
leave	O
their	O
mistakes	O
in	O
the	O
very	O
same	O
programs	O
!	O
So	O
I	O
touch	O
on	O
several	O
common	B-KEY
errors	I-KEY
I	O
see	O
in	O
enterprise	O
programming	O
,	O
and	O
then	O
briefly	O
mention	O
how	O
to	O
avoid	O
those	O
mistakes	O

Deadlock-free	B-KEY
scheduling	I-KEY
in	O
flexible	B-KEY
manufacturing	I-KEY
systems	I-KEY
using	O
Petri	B-KEY
nets	I-KEY
This	O
paper	O
addresses	O
the	O
deadlock-free	B-KEY
scheduling	I-KEY
problem	O
in	O
Flexible	B-KEY
Manufacturing	I-KEY
Systems	I-KEY
.	O
An	O
efficient	O
deadlock-free	B-KEY
scheduling	I-KEY
algorithm	O
was	O
developed	O
,	O
using	O
timed	O
Petri	B-KEY
nets	I-KEY
,	O
for	O
a	O
class	O
of	O
FMSs	O
called	O
Systems	B-KEY
of	I-KEY
Sequential	I-KEY
Systems	I-KEY
with	I-KEY
Shared	I-KEY
Resources	I-KEY
-LRB-	O
S/sup	O
4	O
/	O
R	O
-RRB-	O
.	O
The	O
algorithm	O
generates	O
a	O
partial	B-KEY
reachability	I-KEY
graph	I-KEY
to	O
find	O
the	O
optimal	O
or	O
near-optimal	O
deadlock-free	B-KEY
schedule	I-KEY
in	O
terms	O
of	O
the	O
firing	O
sequence	O
of	O
the	O
transitions	O
of	O
the	O
Petri	O
net	O
model	O
.	O
The	O
objective	O
is	O
to	O
minimize	O
the	O
mean	O
flow	O
time	O
-LRB-	O
MFT	O
-RRB-	O
.	O
An	O
efficient	O
truncation	O
technique	O
,	O
based	O
on	O
the	O
siphon	O
concept	O
,	O
has	O
been	O
developed	O
and	O
used	O
to	O
generate	O
the	O
minimum	O
necessary	O
portion	O
of	O
the	O
reachability	O
graph	O
to	O
be	O
searched	O
.	O
It	O
has	O
been	O
shown	O
experimentally	O
that	O
the	O
developed	O
siphon	B-KEY
truncation	I-KEY
technique	I-KEY
enhances	O
the	O
ability	O
to	O
develop	O
deadlock-free	B-KEY
schedules	I-KEY
of	O
systems	O
with	O
a	O
high	O
number	O
of	O
deadlocks	O
,	O
which	O
can	O
not	O
be	O
achieved	O
using	O
standard	O
Petri	B-KEY
net	I-KEY
scheduling	O
approaches	O
.	O
It	O
may	O
be	O
necessary	O
,	O
in	O
some	O
cases	O
,	O
to	O
relax	O
the	O
optimality	O
condition	O
for	O
large	O
FMSs	O
in	O
order	O
to	O
make	O
the	O
search	O
effort	O
reasonable	O
.	O
Hence	O
,	O
a	O
User	B-KEY
Control	I-KEY
Factor	I-KEY
-LRB-	O
UCF	O
-RRB-	O
was	O
defined	O
and	O
used	O
in	O
the	O
scheduling	O
algorithm	O
.	O
The	O
objective	O
of	O
using	O
the	O
UCF	O
is	O
to	O
achieve	O
an	O
acceptable	O
trade-off	O
between	O
the	O
solution	O
quality	O
and	O
the	O
search	O
effort	O
.	O
Its	O
effect	O
on	O
the	O
MFT	O
and	O
the	O
CPU	B-KEY
time	I-KEY
has	O
been	O
investigated	O
.	O
Randomly	B-KEY
generated	I-KEY
examples	I-KEY
are	O
used	O
for	O
illustration	O
and	O
comparison	O
.	O
Although	O
the	O
effect	O
of	O
UCF	O
did	O
not	O
affect	O
the	O
mean	O
flow	O
time	O
,	O
it	O
was	O
shown	O
that	O
increasing	O
it	O
reduces	O
the	O
search	O
effort	O
-LRB-	O
CPU	B-KEY
time	I-KEY
-RRB-	O
significantly	O

Digital	B-KEY
stochastic	I-KEY
realization	I-KEY
of	O
complex	B-KEY
analog	I-KEY
controllers	I-KEY
Stochastic	B-KEY
logic	I-KEY
is	O
based	O
on	O
digital	O
processing	O
of	O
a	O
random	B-KEY
pulse	I-KEY
stream	I-KEY
,	O
where	O
the	O
information	O
is	O
codified	O
as	O
the	O
probability	O
of	O
a	O
high	O
level	O
in	O
a	O
finite	O
sequence	O
.	O
This	O
binary	B-KEY
pulse	I-KEY
sequence	I-KEY
can	O
be	O
digitally	O
processed	O
exploiting	O
the	O
similarity	O
between	O
Boolean	B-KEY
algebra	I-KEY
and	O
statistical	B-KEY
algebra	I-KEY
.	O
Given	O
a	O
random	B-KEY
pulse	I-KEY
sequence	I-KEY
,	O
any	O
Boolean	B-KEY
operation	I-KEY
among	O
individual	O
pulses	O
will	O
correspond	O
to	O
an	O
algebraic	O
expression	O
among	O
the	O
variables	O
represented	O
by	O
their	O
respective	O
average	B-KEY
pulse	I-KEY
rates	I-KEY
.	O
Subsequently	O
,	O
this	O
pulse	B-KEY
stream	I-KEY
can	O
be	O
digitally	O
processed	O
to	O
perform	O
analog	O
operations	O
.	O
In	O
this	O
paper	O
,	O
we	O
propose	O
a	O
stochastic	B-KEY
approach	I-KEY
to	O
the	O
digital	O
implementation	O
of	O
complex	O
controllers	O
using	O
programmable	B-KEY
devices	I-KEY
as	O
an	O
alternative	O
to	O
traditional	O
digital	O
signal	O
processors	O
.	O
As	O
an	O
example	O
,	O
a	O
practical	O
realization	O
of	O
nonlinear	B-KEY
dissipative	I-KEY
controllers	I-KEY
for	O
a	O
series	B-KEY
resonant	I-KEY
converter	I-KEY
is	O
presented	O

Modeling	O
discourse	O
in	O
collaborative	B-KEY
work	I-KEY
support	I-KEY
systems	I-KEY
:	O
a	O
knowledge	B-KEY
representation	I-KEY
and	O
configuration	O
perspective	O
Collaborative	O
work	O
processes	O
usually	O
raise	O
a	O
lot	O
of	O
intricate	O
debates	O
and	O
negotiations	O
among	O
participants	O
,	O
whereas	O
conflicts	B-KEY
of	I-KEY
interest	I-KEY
are	O
inevitable	O
and	O
support	O
for	O
achieving	O
consensus	B-KEY
and	O
compromise	B-KEY
is	O
required	O
.	O
Individual	O
contributions	O
,	O
brought	O
up	O
by	O
parties	O
with	O
different	O
backgrounds	O
and	O
interests	O
,	O
need	O
to	O
be	O
appropriately	O
structured	O
and	O
maintained	O
.	O
This	O
paper	O
presents	O
a	O
model	O
of	O
discourse	O
acts	O
that	O
participants	O
use	O
to	O
communicate	O
their	O
attitudes	O
to	O
each	O
other	O
,	O
or	O
affect	O
the	O
attitudes	O
of	O
others	O
,	O
in	O
such	O
environments	O
.	O
The	O
first	O
part	O
deals	O
with	O
the	O
knowledge	B-KEY
representation	I-KEY
and	O
communication	O
aspects	O
of	O
the	O
problem	O
,	O
while	O
the	O
second	O
one	O
,	O
in	O
the	O
context	O
of	O
an	O
already	O
implemented	O
system	O
,	O
namely	O
HERMES	B-KEY
,	O
with	O
issues	O
related	O
to	O
the	O
configuration	O
of	O
the	O
contributions	O
asserted	O
at	O
each	O
discourse	O
instance	O
.	O
The	O
overall	O
work	O
focuses	O
on	O
the	O
machinery	O
needed	O
in	O
a	O
computer-assisted	O
collaborative	O
work	O
environment	O
,	O
the	O
aim	O
being	O
to	O
further	O
enhance	O
the	O
human-computer	B-KEY
interaction	I-KEY

Oxygen-enhanced	B-KEY
MRI	I-KEY
of	O
the	O
brain	B-KEY
Blood	O
oxygenation	O
level-dependent	O
-LRB-	O
BOLD	O
-RRB-	O
contrast	O
MRI	O
is	O
a	O
potential	O
method	O
for	O
a	O
physiological	O
characterization	O
of	O
tissue	O
beyond	O
mere	O
morphological	O
representation	O
.	O
The	O
purpose	O
of	O
this	O
study	O
was	O
to	O
develop	O
evaluation	O
techniques	O
for	O
such	O
examinations	O
using	O
a	O
hyperoxia	B-KEY
challenge	O
.	O
Administration	O
of	O
pure	O
oxygen	O
was	O
applied	O
to	O
test	O
these	O
techniques	O
,	O
as	O
pure	O
oxygen	O
can	O
be	O
expected	O
to	O
induce	O
relatively	O
small	O
signal	O
intensity	O
-LRB-	O
SI	O
-RRB-	O
changes	O
compared	O
to	O
CO/sub	O
2	O
/	O
-	O
containing	O
gases	O
and	O
thus	O
requires	O
very	O
sensitive	O
evaluation	O
methods	O
.	O
Fourteen	O
volunteers	O
were	O
investigated	O
by	O
alternating	O
between	O
breathing	O
100	O
%	O
O/sub	O
2	O
/	O
and	O
normal	O
air	O
,	O
using	O
two	O
different	O
paradigms	O
of	O
administration	O
.	O
Changes	O
ranged	O
from	O
>	O
30	O
%	O
in	O
large	O
veins	O
to	O
1.71	O
%	O
+	O
or-0	O
.14	O
%	O
in	O
basal	O
ganglia	O
and	O
0.82	O
%	O
+	O
or-0	O
.08	O
%	O
in	O
white	O
matter	O
.	O
To	O
account	O
for	O
a	O
slow	O
physiological	B-KEY
response	I-KEY
function	I-KEY
,	O
a	O
reference	O
for	O
correlation	B-KEY
analysis	I-KEY
was	O
derived	O
from	O
the	O
venous	B-KEY
reaction	I-KEY
.	O
An	O
objective	O
method	O
is	O
presented	O
that	O
allows	O
the	O
adaptation	O
of	O
the	O
significance	B-KEY
threshold	I-KEY
to	O
the	O
complexity	O
of	O
the	O
paradigm	O
used	O
.	O
Reference	O
signal	O
characteristics	O
in	O
representative	O
brain	B-KEY
tissue	O
regions	O
were	O
established	O
.	O
As	O
the	O
presented	O
evaluation	O
scheme	O
proved	O
its	O
applicability	O
to	O
small	O
SI	O
changes	O
induced	O
by	O
pure	O
oxygen	O
,	O
it	O
can	O
readily	O
be	O
used	O
for	O
similar	O
experiments	O
with	O
other	O
gases	O

The	O
existence	B-KEY
condition	I-KEY
of	O
gamma	O
-	O
acyclic	O
database	O
schemes	O
with	O
MVDs	B-KEY
constraints	I-KEY
It	O
is	O
very	O
important	O
to	O
use	O
database	B-KEY
technology	I-KEY
for	O
a	O
large-scale	B-KEY
system	I-KEY
such	O
as	O
ERP	O
and	O
MIS	O
.	O
A	O
good	O
database	O
design	O
may	O
improve	O
the	O
performance	O
of	O
the	O
system	O
.	O
Some	O
research	O
shows	O
that	O
a	O
gamma	O
-	O
acyclic	O
database	O
scheme	O
has	O
many	O
good	O
properties	O
,	O
e.g.	O
,	O
each	O
connected	B-KEY
join	I-KEY
expression	I-KEY
is	O
monotonous	O
,	O
which	O
helps	O
to	O
improve	O
query	B-KEY
performance	I-KEY
of	O
the	O
database	O
system	O
.	O
Thus	O
what	O
conditions	O
are	O
needed	O
to	O
generate	O
a	O
gamma	O
-	O
acyclic	O
database	O
scheme	O
for	O
a	O
given	O
relational	O
scheme	O
?	O
In	O
this	O
paper	O
,	O
the	O
sufficient	B-KEY
and	I-KEY
necessary	I-KEY
condition	I-KEY
of	O
the	O
existence	O
of	O
gamma	O
-	O
acyclic	O
,	O
join-lossless	O
and	O
dependencies-preserved	O
database	O
schemes	O
meeting	O
4NF	O
is	O
given	O

Speech	O
enhancement	O
using	O
a	O
mixture-maximum	B-KEY
model	I-KEY
We	O
present	O
a	O
spectral	B-KEY
domain	I-KEY
,	O
speech	B-KEY
enhancement	I-KEY
algorithm	I-KEY
.	O
The	O
new	O
algorithm	O
is	O
based	O
on	O
a	O
mixture	B-KEY
model	I-KEY
for	O
the	O
short	B-KEY
time	I-KEY
spectrum	I-KEY
of	O
the	O
clean	B-KEY
speech	I-KEY
signal	I-KEY
,	O
and	O
on	O
a	O
maximum	O
assumption	O
in	O
the	O
production	O
of	O
the	O
noisy	B-KEY
speech	I-KEY
spectrum	I-KEY
.	O
In	O
the	O
past	O
this	O
model	O
was	O
used	O
in	O
the	O
context	O
of	O
noise	B-KEY
robust	I-KEY
speech	I-KEY
recognition	I-KEY
.	O
In	O
this	O
paper	O
we	O
show	O
that	O
this	O
model	O
is	O
also	O
effective	O
for	O
improving	O
the	O
quality	O
of	O
speech	O
signals	O
corrupted	O
by	O
additive	B-KEY
noise	I-KEY
.	O
The	O
computational	O
requirements	O
of	O
the	O
algorithm	O
can	O
be	O
significantly	O
reduced	O
,	O
essentially	O
without	O
paying	O
performance	B-KEY
penalties	I-KEY
,	O
by	O
incorporating	O
a	O
dual	B-KEY
codebook	I-KEY
scheme	O
with	O
tied	B-KEY
variances	I-KEY
.	O
Experiments	O
,	O
using	O
recorded	B-KEY
speech	I-KEY
signals	I-KEY
and	O
actual	O
noise	B-KEY
sources	I-KEY
,	O
show	O
that	O
in	O
spite	O
of	O
its	O
low	B-KEY
computational	I-KEY
requirements	I-KEY
,	O
the	O
algorithm	O
shows	O
improved	O
performance	O
compared	O
to	O
alternative	O
speech	B-KEY
enhancement	I-KEY
algorithms	I-KEY

Vehicle	B-KEY
travel	I-KEY
time	I-KEY
models	O
for	O
AGV	B-KEY
systems	O
under	O
various	O
dispatching	O
rules	O
The	O
design	O
and	O
evaluation	O
of	O
AGV-based	O
material	O
handling	O
systems	O
are	O
highly	O
complex	O
because	O
of	O
the	O
randomness	O
and	O
the	O
large	O
number	O
of	O
variables	O
involved	O
.	O
Vehicle	B-KEY
travel	I-KEY
time	I-KEY
is	O
a	O
fundamental	O
parameter	O
for	O
solving	O
various	O
flexible	B-KEY
manufacturing	I-KEY
system	I-KEY
-LRB-	O
FMS	B-KEY
-RRB-	O
design	O
problems	O
.	O
This	O
article	O
presents	O
stochastic	O
vehicle	B-KEY
travel	I-KEY
time	I-KEY
models	O
for	O
AGV-based	O
material	O
handling	O
systems	O
with	O
emphasis	O
on	O
the	O
empty	O
travel	O
times	O
of	O
vehicles	O
.	O
Various	O
vehicle	B-KEY
dispatching	I-KEY
rules	I-KEY
examined	O
here	O
include	O
the	O
nearest	B-KEY
vehicle	I-KEY
selection	I-KEY
rule	I-KEY
and	O
longest	B-KEY
idle	I-KEY
vehicle	I-KEY
selection	I-KEY
rule	I-KEY
.	O
A	O
simulation	O
experiment	O
is	O
used	O
to	O
evaluate	O
and	O
demonstrate	O
the	O
presented	O
models	O

Neural	B-KEY
networks	I-KEY
in	O
optimal	O
filtration	O
The	O
combined	O
use	O
and	O
mutual	O
influence	O
of	O
neural	B-KEY
networks	I-KEY
and	O
optimal	B-KEY
filtering	I-KEY
is	O
considered	O
;	O
the	O
neural-network	O
and	O
filtering	O
approaches	O
are	O
compared	O
by	O
solving	O
two	O
simple	O
optimal-filtering	O
problems	O
:	O
linear	B-KEY
filtering	I-KEY
and	O
the	O
filtering	O
of	O
a	O
binary	B-KEY
telegraph	I-KEY
signal	I-KEY
corresponding	O
to	O
observations	B-KEY
in	O
discrete	B-KEY
white	I-KEY
noise	I-KEY

An	O
analytical	O
model	O
for	O
a	O
composite	B-KEY
adaptive	I-KEY
rectangular	I-KEY
structure	I-KEY
using	O
the	O
Heaviside	B-KEY
function	I-KEY
The	O
objective	O
of	O
this	O
article	O
is	O
to	O
describe	O
a	O
mathematical	B-KEY
model	I-KEY
,	O
based	O
on	O
the	O
Heaviside	B-KEY
function	I-KEY
and	O
on	O
the	O
delta	B-KEY
-	I-KEY
Dirac	I-KEY
distribution	I-KEY
,	O
for	O
a	O
composite	B-KEY
adaptive	I-KEY
rectangular	I-KEY
structure	I-KEY
with	O
embedded	O
and/or	O
bonded	O
piezoelectric	B-KEY
actuators	I-KEY
and	O
sensors	O
.	O
In	O
the	O
adopted	O
structure	O
model	O
,	O
the	O
laminae	O
are	O
made	O
up	O
a	O
configuration	O
of	O
rectangular	O
nonpiezoelectric	O
and	O
piezoelectric	B-KEY
patches	I-KEY
.	O
The	O
laminae	O
do	O
not	O
all	O
have	O
the	O
same	O
area	O
nor	O
do	O
they	O
present	O
the	O
same	O
configuration	O
,	O
such	O
that	O
there	O
are	O
points	O
where	O
there	O
is	O
no	O
material	O
.	O
The	O
equations	B-KEY
of	I-KEY
motion	I-KEY
and	O
the	O
boundary	B-KEY
conditions	I-KEY
,	O
which	O
describe	O
the	O
electromechanical	B-KEY
coupling	I-KEY
,	O
are	O
based	O
on	O
the	O
Mindlin	B-KEY
displacement	I-KEY
field	I-KEY
,	O
on	O
the	O
linear	O
theory	O
of	O
piezoelectricity	O
,	O
and	O
on	O
the	O
Hamilton	B-KEY
principle	I-KEY

Ultrafast	B-KEY
compound	I-KEY
imaging	I-KEY
for	O
2-D	O
motion	O
vector	O
estimation	O
:	O
application	O
to	O
transient	B-KEY
elastography	I-KEY
This	O
paper	O
describes	O
a	O
new	O
technique	O
for	O
two-dimensional	O
-LRB-	O
2-D	O
-RRB-	O
imaging	O
of	O
the	O
motion	O
vector	O
at	O
a	O
very	O
high	B-KEY
frame	I-KEY
rate	I-KEY
with	O
ultrasound	B-KEY
.	O
Its	O
potential	O
is	O
experimentally	O
demonstrated	O
for	O
transient	B-KEY
elastography	I-KEY
.	O
But	O
,	O
beyond	O
this	O
application	O
,	O
it	O
also	O
could	O
be	O
promising	O
for	O
color	O
flow	O
and	O
reflectivity	B-KEY
imaging	I-KEY
.	O
To	O
date	O
,	O
only	O
axial	B-KEY
displacements	I-KEY
induced	O
in	O
human	B-KEY
tissues	I-KEY
by	O
low-frequency	O
vibrators	O
were	O
measured	O
during	O
transient	B-KEY
elastography	I-KEY
.	O
The	O
proposed	O
technique	O
allows	O
us	O
to	O
follow	O
both	O
axial	O
and	O
lateral	B-KEY
displacements	I-KEY
during	O
the	O
shear	B-KEY
wave	I-KEY
propagation	I-KEY
and	O
thus	O
should	O
improve	O
Young	O
's	O
modulus	O
image	O
reconstruction	O
.	O
The	O
process	O
is	O
a	O
combination	O
of	O
several	O
ideas	O
well-known	O
in	O
ultrasonic	B-KEY
imaging	I-KEY
:	O
ultra-fast	O
imaging	O
,	O
multisynthetic	B-KEY
aperture	I-KEY
beamforming	I-KEY
,	O
1-D	O
speckle	O
tracking	O
,	O
and	O
compound	O
imaging	O
.	O
Classical	O
beamforming	O
in	O
the	O
transmit	O
mode	O
is	O
replaced	O
here	O
by	O
a	O
single	B-KEY
plane	I-KEY
wave	I-KEY
insonification	I-KEY
increasing	O
the	O
frame	O
rate	O
by	O
at	O
least	O
a	O
factor	O
of	O
128	O
.	O
The	O
beamforming	O
is	O
achieved	O
only	O
in	O
the	O
receive	O
mode	O
on	O
two	O
independent	O
subapertures	O
.	O
Comparison	O
of	O
successive	O
frames	O
by	O
a	O
classical	O
1-D	O
speckle	O
tracking	O
algorithm	O
allows	O
estimation	O
of	O
displacements	O
along	O
two	O
different	O
directions	O
linked	O
to	O
the	O
subapertures	O
beams	O
.	O
The	O
variance	O
of	O
the	O
estimates	O
is	O
finally	O
improved	O
by	O
tilting	O
the	O
emitting	O
plane	O
wave	O
at	O
each	O
insonification	O
,	O
thus	O
allowing	O
reception	O
of	O
successive	O
decorrelated	B-KEY
speckle	I-KEY
patterns	I-KEY

Personal	B-KEY
cards	I-KEY
for	O
on-line	O
purchases	O
Buying	O
presents	O
over	O
the	O
Web	O
has	O
advantages	O
for	O
a	O
busy	O
person	O
:	O
lots	O
of	O
choices	O
,	O
24-hour	O
accessibility	O
,	O
quick	O
delivery	O
,	O
and	O
you	O
do	O
n't	O
even	O
have	O
to	O
wrap	O
the	O
gift	O
.	O
But	O
many	O
people	O
like	O
to	O
select	O
a	O
card	O
or	O
write	O
a	O
personal	O
note	O
to	O
go	O
with	O
their	O
presents	O
,	O
and	O
the	O
options	O
for	O
doing	O
that	O
have	O
been	O
limited	O
.	O
Two	O
companies	O
have	O
seen	O
this	O
limitation	O
as	O
an	O
opportunity	O
:	O
4YourSoul	O
.	O
com	O
and	O
CardintheBox.com	B-KEY

Analysis	O
and	O
operation	O
of	O
hybrid	B-KEY
active	I-KEY
filter	I-KEY
for	O
harmonic	O
elimination	O
This	O
paper	O
presents	O
a	O
hybrid	B-KEY
active	I-KEY
filter	I-KEY
topology	O
and	O
its	O
control	O
to	O
suppress	O
the	O
harmonic	O
currents	O
from	O
entering	O
the	O
power	O
source	O
.	O
The	O
adopted	O
hybrid	B-KEY
active	I-KEY
filter	I-KEY
consists	O
of	O
one	O
active	O
filter	O
and	O
one	O
passive	O
filter	O
connected	O
in	O
series	O
.	O
By	O
controlling	O
the	O
equivalent	B-KEY
output	I-KEY
voltage	I-KEY
of	O
active	B-KEY
filter	I-KEY
,	O
the	O
harmonic	B-KEY
currents	I-KEY
generated	O
by	O
the	O
nonlinear	B-KEY
load	I-KEY
are	O
blocked	O
and	O
flowed	O
into	O
the	O
passive	B-KEY
filter	I-KEY
.	O
The	O
power	O
rating	O
of	O
the	O
converter	O
is	O
reduced	O
compared	O
with	O
the	O
pure	O
active	B-KEY
filters	I-KEY
to	O
filter	O
the	O
harmonic	B-KEY
currents	I-KEY
.	O
The	O
harmonic	B-KEY
current	I-KEY
detecting	O
approach	O
and	O
DC-link	B-KEY
voltage	I-KEY
regulation	I-KEY
are	O
proposed	O
to	O
obtain	O
equivalent	O
voltage	O
of	O
active	B-KEY
filter	I-KEY
.	O
The	O
effectiveness	O
of	O
the	O
adopted	O
topology	O
and	O
control	O
scheme	O
has	O
been	O
verified	O
by	O
the	O
computer	B-KEY
simulation	I-KEY
and	O
experimental	O
results	O
in	O
a	O
scaled-down	B-KEY
laboratory	I-KEY
prototype	I-KEY

Dynamic	B-KEY
modification	I-KEY
of	O
object	B-KEY
Petri	I-KEY
nets	I-KEY
.	O
An	O
application	O
to	O
modelling	O
protocols	B-KEY
with	O
fork-join	B-KEY
structures	I-KEY
In	O
this	O
paper	O
we	O
discuss	O
possibilities	O
of	O
modelling	O
protocols	B-KEY
by	O
objects	O
in	O
object-based	O
high-level	O
Petri	O
nets	O
.	O
Some	O
advantages	O
of	O
dynamically	O
modifying	O
the	O
structure	O
of	O
token	B-KEY
objects	I-KEY
are	O
discussed	O
and	O
the	O
need	O
for	O
further	O
investigations	O
into	O
mathematically	B-KEY
rigorous	I-KEY
foundations	I-KEY
of	O
object	B-KEY
net	I-KEY
formalisms	I-KEY
incorporating	O
facilities	O
for	O
such	O
operations	O
on	O
its	O
token	O
nets	O
is	O
emphasised	O

Two	O
quantum	B-KEY
analogues	I-KEY
of	O
Fisher	O
information	O
from	O
a	O
large	B-KEY
deviation	I-KEY
viewpoint	I-KEY
of	O
quantum	B-KEY
estimation	I-KEY
We	O
discuss	O
two	O
quantum	B-KEY
analogues	I-KEY
of	O
the	O
Fisher	O
information	O
,	O
the	O
symmetric	B-KEY
logarithmic	I-KEY
derivative	I-KEY
Fisher	I-KEY
information	I-KEY
and	O
Kubo-Mori-Bogoljubov	B-KEY
Fisher	I-KEY
information	I-KEY
from	O
a	O
large	B-KEY
deviation	I-KEY
viewpoint	I-KEY
of	O
quantum	B-KEY
estimation	I-KEY
and	O
prove	O
that	O
the	O
former	O
gives	O
the	O
true	O
bound	O
and	O
the	O
latter	O
gives	O
the	O
bound	O
of	O
consistent	B-KEY
superefficient	I-KEY
estimators	I-KEY
.	O
As	O
another	O
comparison	O
,	O
it	O
is	O
shown	O
that	O
the	O
difference	O
between	O
them	O
is	O
characterized	O
by	O
the	O
change	O
of	O
the	O
order	O
of	O
limits	O

Towards	O
strong	B-KEY
stability	I-KEY
of	O
concurrent	B-KEY
repetitive	I-KEY
processes	I-KEY
sharing	O
resources	O
The	O
paper	O
presents	O
a	O
method	O
for	O
design	O
of	O
stability	O
conditions	O
of	O
concurrent	O
,	O
repetitive	O
processes	O
sharing	O
common	B-KEY
resources	I-KEY
.	O
Steady-state	B-KEY
behaviour	I-KEY
of	O
the	O
system	O
with	O
m	O
cyclic	B-KEY
processes	I-KEY
utilising	O
a	O
resource	O
with	O
the	O
mutual	B-KEY
exclusion	I-KEY
is	O
considered	O
.	O
Based	O
on	O
a	O
recurrent	B-KEY
equations	I-KEY
framework	I-KEY
necessary	B-KEY
and	I-KEY
sufficient	I-KEY
conditions	I-KEY
for	O
the	O
existence	O
of	O
maximal	B-KEY
performance	I-KEY
steady-state	I-KEY
are	O
presented	O
.	O
It	O
was	O
shown	O
that	O
if	O
the	O
conditions	O
hold	O
then	O
the	O
m-process	O
system	O
is	O
marginally	O
stable	O
,	O
i.e.	O
,	O
a	O
steady-state	O
of	O
the	O
system	O
depends	O
on	O
the	O
perturbations	O
.	O
The	O
problem	O
of	O
finding	O
the	O
relative	O
positions	O
of	O
the	O
processes	O
leading	O
to	O
waiting-free	O
-LRB-	O
maximal	O
efficiency	O
-RRB-	O
steady-states	O
of	O
the	O
system	O
is	O
formulated	O
as	O
a	O
constraint	B-KEY
logic	I-KEY
programming	I-KEY
problem	O
.	O
An	O
example	O
illustrating	O
the	O
solving	O
of	O
the	O
problem	O
for	O
a	O
3-process	B-KEY
system	I-KEY
using	O
object-oriented	O
,	O
constraint	B-KEY
logic	I-KEY
programming	I-KEY
language	O
Oz	O
is	O
presented	O
.	O
A	O
condition	O
sufficient	O
for	O
strong	B-KEY
stability	I-KEY
of	O
the	O
m-process	O
system	O
is	O
given	O
.	O
When	O
the	O
condition	O
holds	O
then	O
for	O
any	O
initial	O
phases	O
of	O
the	O
processes	O
a	O
waiting-free	B-KEY
steady-state	I-KEY
will	O
be	O
reached	O

Differential	B-KEY
algebraic	I-KEY
systems	I-KEY
anew	O
It	O
is	O
proposed	O
to	O
figure	O
out	O
the	O
leading	O
term	O
in	O
differential	B-KEY
algebraic	I-KEY
systems	I-KEY
more	O
precisely	O
.	O
Low	B-KEY
index	I-KEY
linear	I-KEY
systems	I-KEY
with	O
those	O
properly	O
stated	O
leading	O
terms	O
are	O
considered	O
in	O
detail	O
.	O
In	O
particular	O
,	O
it	O
is	O
asked	O
whether	O
a	O
numerical	B-KEY
integration	I-KEY
method	I-KEY
applied	O
to	O
the	O
original	O
system	O
reaches	O
the	O
inherent	B-KEY
regular	I-KEY
ODE	I-KEY
without	O
conservation	O
,	O
i.e.	O
,	O
whether	O
the	O
discretization	O
and	O
the	O
decoupling	O
commute	B-KEY
in	O
some	O
sense	O
.	O
In	O
general	O
one	O
can	O
not	O
expect	O
this	O
commutativity	B-KEY
so	O
that	O
additional	O
difficulties	O
like	O
strong	O
stepsize	B-KEY
restrictions	I-KEY
may	O
arise	O
.	O
Moreover	O
,	O
abstract	B-KEY
differential	I-KEY
algebraic	I-KEY
equations	I-KEY
in	O
infinite-dimensional	O
Hilbert	O
spaces	O
are	O
introduced	O
,	O
and	O
the	O
index	O
notion	O
is	O
generalized	O
to	O
those	O
equations	O
.	O
In	O
particular	O
,	O
partial	O
differential	O
algebraic	O
equations	O
are	O
considered	O
in	O
this	O
abstract	O
formulation	O

Computing	O
stationary	B-KEY
Nash	I-KEY
equilibria	I-KEY
of	O
undiscounted	B-KEY
single-controller	I-KEY
stochastic	I-KEY
games	I-KEY
Given	O
a	O
two-person	O
,	O
nonzero-sum	O
stochastic	O
game	O
where	O
the	O
second	O
player	O
controls	O
the	O
transitions	O
,	O
we	O
formulate	O
a	O
linear	B-KEY
complementarity	I-KEY
problem	I-KEY
LCP	O
-LRB-	O
q	O
,	O
M	O
-RRB-	O
whose	O
solution	O
gives	O
a	O
Nash	O
equilibrium	O
pair	O
of	O
stationary	B-KEY
strategies	I-KEY
under	O
the	O
limiting	B-KEY
average	I-KEY
payoff	I-KEY
criterion	I-KEY
.	O
The	O
matrix	O
M	O
constructed	O
is	O
of	O
the	O
copositive	O
class	O
so	O
that	O
Lemke	O
's	O
algorithm	O
will	O
process	O
it	O
.	O
We	O
will	O
also	O
do	O
the	O
same	O
for	O
a	O
special	O
class	O
of	O
N-person	B-KEY
stochastic	I-KEY
games	I-KEY
called	O
polymatrix	B-KEY
stochastic	I-KEY
games	I-KEY

From	O
continuous	B-KEY
recovery	I-KEY
to	O
discrete	B-KEY
filtering	I-KEY
in	O
numerical	B-KEY
approximations	I-KEY
of	O
conservation	B-KEY
laws	I-KEY
Modern	O
numerical	B-KEY
approximations	I-KEY
of	O
conservation	B-KEY
laws	I-KEY
rely	O
on	O
numerical	B-KEY
dissipation	I-KEY
as	O
a	O
means	O
of	O
stabilization	O
.	O
The	O
older	O
,	O
alternative	O
approach	O
is	O
the	O
use	O
of	O
central	B-KEY
differencing	I-KEY
with	O
a	O
dose	O
of	O
artificial	B-KEY
dissipation	I-KEY
.	O
In	O
this	O
paper	O
we	O
review	O
the	O
successful	O
class	O
of	O
weighted	O
essentially	O
non-oscillatory	O
finite	B-KEY
volume	I-KEY
schemes	I-KEY
which	O
comprise	O
sophisticated	O
methods	O
of	O
the	O
first	O
kind	O
.	O
New	O
developments	O
in	O
image	B-KEY
processing	I-KEY
have	O
made	O
new	O
devices	O
possible	O
which	O
can	O
serve	O
as	O
highly	O
nonlinear	O
artificial	B-KEY
dissipation	I-KEY
terms	O
.	O
We	O
view	O
artificial	B-KEY
dissipation	I-KEY
as	O
discrete	B-KEY
filter	I-KEY
operation	O
and	O
introduce	O
several	O
new	O
algorithms	O
inspired	O
by	O
image	O
processing	O

Infrared-image	B-KEY
classification	I-KEY
using	O
hidden	B-KEY
Markov	I-KEY
trees	I-KEY
An	O
image	O
of	O
a	O
three-dimensional	O
target	O
is	O
generally	O
characterized	O
by	O
the	O
visible	O
target	O
subcomponents	O
,	O
with	O
these	O
dictated	O
by	O
the	O
target-sensor	B-KEY
orientation	I-KEY
-LRB-	O
target	B-KEY
pose	I-KEY
-RRB-	O
.	O
An	O
image	O
often	O
changes	O
quickly	O
with	O
variable	O
pose	O
.	O
We	O
define	O
a	O
class	O
as	O
a	O
set	O
of	O
contiguous	O
target-sensor	B-KEY
orientations	I-KEY
over	O
which	O
the	O
associated	O
target	O
image	O
is	O
relatively	O
stationary	O
with	O
aspect	O
.	O
Each	O
target	O
is	O
in	O
general	O
characterized	O
by	O
multiple	O
classes	O
.	O
A	O
distinct	O
set	O
of	O
Wiener	B-KEY
filters	I-KEY
are	O
employed	O
for	O
each	O
class	O
of	O
images	O
,	O
to	O
identify	O
the	O
presence	O
of	O
target	O
subcomponents	O
.	O
A	O
Karhunen-Loeve	B-KEY
representation	I-KEY
is	O
used	O
to	O
minimize	B-KEY
the	O
number	O
of	O
filters	O
-LRB-	O
templates	O
-RRB-	O
associated	O
with	O
a	O
given	O
subcomponent	O
.	O
The	O
statistical	O
relationships	O
between	O
the	O
different	O
target	O
subcomponents	O
are	O
modeled	O
via	O
a	O
hidden	B-KEY
Markov	I-KEY
tree	I-KEY
-LRB-	O
HMT	B-KEY
-RRB-	O
.	O
The	O
HMT	B-KEY
classifier	O
is	O
discussed	O
and	O
example	O
results	O
are	O
presented	O
for	O
forward-looking-infrared	O
-LRB-	O
FLIR	O
-RRB-	O
imagery	O
of	O
several	O
vehicles	B-KEY

19in	B-KEY
monitors	O
-LSB-	O
CRT	O
survey	O
-RSB-	O
Upgrade	O
your	O
monitor	O
from	O
as	O
little	O
as	O
Pounds	O
135	O
.	O
With	O
displays	O
on	O
test	O
and	O
ranging	O
up	O
to	O
Pounds	O
400	O
,	O
whether	O
you	O
're	O
after	O
the	O
last	O
word	O
in	O
quality	O
or	O
simply	O
looking	O
for	O
again	O
,	O
this	O
Labs	O
holds	O
the	O
answer	O
.	O
Looks	O
at	O
ADI	B-KEY
MicroScan	I-KEY
M900	I-KEY
,	O
CTX	B-KEY
PR960F	I-KEY
,	O
Eizo	B-KEY
FlexScan	I-KEY
T766	I-KEY
,	O
Hansol	B-KEY
920D	I-KEY
,	O
Hansol920P	B-KEY
,	O
Hitachi	B-KEY
CM715ET	I-KEY
,	O
Hitachi	B-KEY
CM721FET	I-KEY
,	O
liyama	B-KEY
Vision	I-KEY
Master	I-KEY
Pro	I-KEY
454	I-KEY
,	O
LG	B-KEY
Flatron	I-KEY
915FT	I-KEY
Plus	I-KEY
,	I-KEY
Mitsubishi	O
Diamond	B-KEY
Pro	I-KEY
920	I-KEY
,	I-KEY
NEC	O
MultiSync	B-KEY
FE950	I-KEY
+	I-KEY
,	O
Philips	O
109S40	B-KEY
,	I-KEY
Samsung	O
SyncMaster	B-KEY
959NF	I-KEY
,	I-KEY
Sony	O
Multiscan	B-KEY
CPD-G420	I-KEY
,	I-KEY
and	O
ViewSonic	O
G90f	B-KEY

On	O
the	O
monotonicity	B-KEY
conservation	I-KEY
in	O
numerical	B-KEY
solutions	I-KEY
of	O
the	O
heat	B-KEY
equation	I-KEY
It	O
is	O
important	O
to	O
choose	O
such	O
numerical	O
methods	O
in	O
practice	O
that	O
mirror	O
the	O
characteristic	B-KEY
properties	I-KEY
of	O
the	O
described	O
process	O
beyond	O
the	O
stability	O
and	O
convergence	O
.	O
The	O
investigated	O
qualitative	B-KEY
property	I-KEY
in	O
this	O
paper	O
is	O
the	O
conservation	O
of	O
the	O
monotonicity	O
in	O
space	O
of	O
the	O
initial	O
heat	O
distribution	O
.	O
We	O
prove	O
some	O
statements	O
about	O
the	O
monotonicity	B-KEY
conservation	I-KEY
and	O
total	O
monotonicity	O
of	O
one-step	B-KEY
vector-iterations	I-KEY
.	O
Then	O
,	O
applying	O
these	O
results	O
,	O
we	O
consider	O
the	O
numerical	B-KEY
solutions	I-KEY
of	O
the	O
one-dimensional	O
heat	B-KEY
equation	I-KEY
.	O
Our	O
main	O
theorem	O
formulates	O
the	O
necessary	B-KEY
and	I-KEY
sufficient	I-KEY
condition	I-KEY
of	O
the	O
uniform	O
monotonicity	B-KEY
conservation	I-KEY
.	O
The	O
sharpness	O
of	O
the	O
conditions	O
is	O
demonstrated	O
by	O
numerical	O
examples	O

Women	B-KEY
in	O
computing	O
around	O
the	O
world	O
This	O
paper	O
describes	O
the	O
participation	O
of	O
women	B-KEY
in	O
computing	O
in	O
more	O
than	O
30	O
countries	O
,	O
by	O
focussing	O
on	O
participation	O
at	O
undergraduate	O
level	O
.	O
A	O
brief	O
discussion	O
covers	O
how	O
societal	O
and	O
cultural	B-KEY
factors	I-KEY
may	O
affect	O
women	B-KEY
's	O
participation	O
.	O
Statistics	B-KEY
from	O
many	O
different	O
sources	O
are	O
presented	O
for	O
comparison	O
.	O
Generally	O
,	O
participation	O
is	O
low	O
-	O
most	O
countries	O
fall	O
in	O
the	O
10-40	O
%	O
range	O
with	O
a	O
few	O
below	O
10	O
%	O
and	O
a	O
few	O
above	O
40	O
%	O

Stability	B-KEY
analysis	I-KEY
of	O
the	O
characteristic	B-KEY
polynomials	I-KEY
whose	O
coefficients	O
are	O
polynomials	O
of	O
interval	B-KEY
parameters	I-KEY
using	O
monotonicity	B-KEY
We	O
analyze	O
the	O
stability	O
of	O
the	O
characteristic	B-KEY
polynomials	I-KEY
whose	O
coefficients	O
are	O
polynomials	O
of	O
interval	B-KEY
parameters	I-KEY
via	O
monotonicity	B-KEY
methods	O
.	O
Our	O
stability	O
conditions	O
are	O
based	O
on	O
Frazer-Duncan	O
's	O
theorem	O
and	O
all	O
conditions	O
can	O
be	O
checked	O
using	O
only	O
endpoint	B-KEY
values	I-KEY
of	O
interval	B-KEY
parameters	I-KEY
.	O
These	O
stability	O
conditions	O
are	O
necessary	O
and	O
sufficient	O
under	O
the	O
monotonicity	B-KEY
assumptions	O
.	O
When	O
the	O
monotonicity	B-KEY
conditions	O
do	O
not	O
hold	O
on	O
the	O
whole	O
parameter	O
region	O
,	O
we	O
present	O
an	O
interval	B-KEY
division	I-KEY
method	I-KEY
and	O
a	O
transformation	B-KEY
algorithm	I-KEY
in	O
order	O
to	O
apply	O
the	O
monotonicity	B-KEY
conditions	O
.	O
Then	O
,	O
our	O
stability	B-KEY
analysis	I-KEY
methods	O
can	O
be	O
applied	O
to	O
all	O
characteristic	B-KEY
polynomials	I-KEY
whose	O
coefficients	O
are	O
polynomials	O
of	O
interval	B-KEY
parameters	I-KEY

Modelling	O
dependencies	O
in	O
paired	O
comparison	O
data	O
a	O
log-linear	B-KEY
approach	I-KEY
In	O
many	O
Bradley-Terry	B-KEY
models	I-KEY
a	O
more	O
or	O
less	O
explicit	O
assumption	O
is	O
that	O
all	O
decisions	O
of	O
the	O
judges	O
are	O
independent	O
.	O
An	O
assumption	O
which	O
might	O
be	O
questionable	O
at	O
least	O
for	O
the	O
decisions	O
of	O
a	O
given	O
judge	O
.	O
In	O
paired	O
comparison	O
studies	O
,	O
a	O
judge	O
chooses	O
among	O
objects	O
several	O
times	O
,	O
and	O
in	O
such	O
cases	O
,	O
judgements	O
made	O
by	O
the	O
same	O
judge	O
are	O
likely	O
to	O
be	O
dependent	O
.	O
A	O
log-linear	O
representation	O
for	O
the	O
Bradley-Terry	B-KEY
model	I-KEY
is	O
developed	O
,	O
which	O
takes	O
into	O
account	O
dependencies	O
between	O
judgements	O
.	O
The	O
modelling	O
of	O
the	O
dependencies	O
is	O
embedded	O
in	O
the	O
analysis	O
of	O
multiple	B-KEY
binomial	I-KEY
responses	I-KEY
,	O
which	O
has	O
the	O
advantage	O
of	O
interpretability	O
in	O
terms	O
of	O
conditional	B-KEY
odds	I-KEY
ratios	I-KEY
.	O
Furthermore	O
,	O
the	O
modelling	O
is	O
done	O
in	O
the	O
framework	O
of	O
generalized	B-KEY
linear	I-KEY
models	I-KEY
,	O
thus	O
parameter	B-KEY
estimation	I-KEY
and	O
the	O
assessment	O
of	O
goodness	B-KEY
of	I-KEY
fit	I-KEY
can	O
be	O
obtained	O
in	O
the	O
standard	O
way	O
by	O
using	O
e.g.	O
GLIM	B-KEY
or	O
another	O
standard	O
software	O

ClioWeb	B-KEY
,	O
ClioRequest	B-KEY
,	O
and	O
Clio	B-KEY
database	I-KEY
:	O
enhancing	O
patron	O
and	O
staff	B-KEY
satisfaction	I-KEY
Faced	O
with	O
increased	O
demand	O
from	O
students	B-KEY
and	O
faculty	B-KEY
for	O
a	O
speedier	O
and	O
more	O
user-friendly	B-KEY
method	I-KEY
of	O
obtaining	O
materials	O
from	O
other	O
institutions	O
,	O
the	O
interlibrary	O
loan	O
-LRB-	O
ILL	O
-RRB-	O
department	O
sought	O
to	O
implement	O
a	O
management	B-KEY
system	I-KEY
which	O
would	O
accomplish	O
the	O
task	O
.	O
Students	B-KEY
wanted	O
remote	B-KEY
interconnectivity	I-KEY
to	O
the	O
system	O
and	O
staff	O
wanted	O
increased	O
workflow	B-KEY
efficiency	I-KEY
,	O
reduced	O
paper	O
work	O
,	O
and	O
better	O
data	B-KEY
management	I-KEY
.	O
This	O
paper	O
focuses	O
on	O
Washington	B-KEY
College	I-KEY
's	O
experience	O
in	O
selecting	O
and	O
implementing	O
an	O
interlibrary	O
loan	O
system	O
,	O
which	O
would	O
enhance	O
student	B-KEY
satisfaction	O
as	O
well	O
as	O
that	O
of	O
the	O
library	O
staff	O

Definition	O
of	O
a	O
similarity	O
measure	O
between	O
cases	O
based	O
on	O
auto/cross-fuzzy	O
thesauri	O
A	O
similarity	O
measure	O
between	O
cases	O
is	O
needed	O
in	O
order	O
to	O
evaluate	O
the	O
degree	O
of	O
similarity	O
when	O
using	O
past	O
similar	O
cases	O
in	O
order	O
to	O
resolve	O
current	O
problems	O
.	O
In	O
similar	B-KEY
case	I-KEY
retrieval	I-KEY
,	O
multiple	O
indices	O
are	O
set	O
up	O
in	O
order	O
to	O
characterize	O
the	O
queries	O
and	O
individual	O
cases	O
,	O
then	O
terms	O
are	O
given	O
as	O
values	O
to	O
each	O
.	O
The	O
similarity	O
measure	O
between	O
cases	O
commonly	O
used	O
is	O
defined	O
using	O
the	O
rate	O
at	O
which	O
the	O
values	O
provided	O
from	O
the	O
corresponding	B-KEY
indices	I-KEY
match	O
.	O
In	O
practice	O
,	O
however	O
,	O
values	O
can	O
not	O
be	O
expected	O
to	O
be	O
mutually	O
exclusive	O
.	O
As	O
a	O
result	O
,	O
a	O
natural	O
expansion	O
of	O
this	O
approach	O
is	O
to	O
have	O
relationships	O
in	O
which	O
mutually	O
similar	O
meanings	O
are	O
reflected	O
in	O
the	O
similarity	O
measure	O
between	O
cases	O
.	O
In	O
this	O
paper	O
the	O
authors	O
consider	O
an	O
auto-fuzzy	B-KEY
thesaurus	I-KEY
which	O
gives	O
the	O
relationship	O
for	O
values	O
between	O
corresponding	B-KEY
indices	I-KEY
and	O
a	O
cross-fuzzy	B-KEY
thesaurus	I-KEY
which	O
gives	O
the	O
relationship	O
for	O
values	O
between	O
mutually	B-KEY
distinct	I-KEY
indices	I-KEY
,	O
then	O
defines	O
a	O
similarity	O
measure	O
between	O
cases	O
which	O
considers	O
the	O
relationship	O
of	O
index	O
values	O
based	O
on	O
these	O
thesauri	O
.	O
This	O
definition	O
satisfies	O
the	O
characteristics	O
required	O
for	O
the	O
operation	O
of	O
case-based	B-KEY
retrieval	I-KEY
even	O
when	O
one	O
value	O
is	O
not	O
necessarily	O
given	O
in	O
the	O
index	O
.	O
Finally	O
,	O
using	O
a	O
test	O
similar	B-KEY
case	I-KEY
retrieval	I-KEY
system	O
,	O
the	O
authors	O
perform	O
a	O
comparative	O
analysis	O
of	O
the	O
proposed	O
similarity	O
measure	O
between	O
cases	O
and	O
a	O
conventional	O
approach	O

Access	O
to	O
information	O
for	O
blind	O
and	O
visually	B-KEY
impaired	I-KEY
clients	I-KEY
This	O
article	O
guides	O
I&R	O
providers	O
in	O
establishing	O
effective	O
communication	B-KEY
techniques	I-KEY
for	O
working	O
with	O
visually	O
impaired	O
consumers	O
.	O
The	O
authors	O
discuss	O
common	O
causes	O
of	O
vision	O
impairment	O
and	O
the	O
functional	O
implications	O
of	O
each	O
and	O
offer	O
information	O
on	O
disability	B-KEY
etiquette	I-KEY
and	O
effective	B-KEY
voice	I-KEY
,	O
accessible	B-KEY
media	I-KEY
and	O
in-person	B-KEY
communication	I-KEY
.	O
There	O
is	O
an	O
overview	O
of	O
assistive	B-KEY
technologies	I-KEY
used	O
by	O
people	O
who	O
are	O
visually	O
impaired-to	O
facilitate	O
written	O
and	O
electronic	B-KEY
communications	I-KEY
as	O
well	O
as	O
low-tech	O
solutions	O
for	O
producing	O
large-print	O
and	O
Braille	B-KEY
materials	I-KEY
in-house	O
.	O
Providers	O
who	O
implement	O
these	O
communication	B-KEY
techniques	I-KEY
will	O
be	O
well	O
equipped	O
to	O
serve	O
visually-impaired	O
consumers	O
,	O
and	O
consumers	O
will	O
be	O
more	O
likely	O
to	O
avail	O
themselves	O
of	O
these	O
services	O
when	O
providers	O
make	O
them	O
easily	O
accessible	O

Stability	B-KEY
in	O
the	O
numerical	B-KEY
solution	I-KEY
of	O
the	O
heat	B-KEY
equation	I-KEY
with	O
nonlocal	B-KEY
boundary	I-KEY
conditions	I-KEY
This	O
paper	O
deals	O
with	O
numerical	O
methods	O
for	O
the	O
solution	O
of	O
the	O
heat	B-KEY
equation	I-KEY
with	O
integral	B-KEY
boundary	I-KEY
conditions	I-KEY
.	O
Finite	B-KEY
differences	I-KEY
are	O
used	O
for	O
the	O
discretization	O
in	O
space	O
.	O
The	O
matrices	B-KEY
specifying	O
the	O
resulting	O
semidiscrete	B-KEY
problem	I-KEY
are	O
proved	O
to	O
satisfy	O
a	O
sectorial	B-KEY
resolvent	I-KEY
condition	I-KEY
,	O
uniformly	O
with	O
respect	O
to	O
the	O
discretization	O
parameter	O
.	O
Using	O
this	O
resolvent	O
condition	O
,	O
unconditional	O
stability	B-KEY
is	O
proved	O
for	O
the	O
fully	B-KEY
discrete	I-KEY
numerical	I-KEY
process	I-KEY
generated	O
by	O
applying	O
A	O
-LRB-	O
theta	O
-RRB-	O
-	O
stable	O
one-step	B-KEY
methods	I-KEY
to	O
the	O
semidiscrete	B-KEY
problem	I-KEY
.	O
This	O
stability	B-KEY
result	O
is	O
established	O
in	O
the	O
maximum	B-KEY
norm	I-KEY
;	O
it	O
improves	O
some	O
previous	O
results	O
in	O
the	O
literature	O
in	O
that	O
it	O
is	O
not	O
subject	O
to	O
various	O
unnatural	O
restrictions	O
which	O
were	O
imposed	O
on	O
the	O
boundary	O
conditions	O
and	O
on	O
the	O
one-step	B-KEY
methods	I-KEY

Ethernet	B-KEY
networks	O
:	O
getting	O
down	O
to	O
business	O
While	O
it	O
seems	O
pretty	O
clear	O
that	O
Ethernet	B-KEY
has	O
won	O
the	O
battle	O
for	O
the	O
mindshare	O
as	O
the	O
network	O
of	O
choice	O
for	O
the	O
factory	B-KEY
floor	I-KEY
,	O
there	O
's	O
still	O
a	O
war	O
to	O
be	O
won	O
in	O
implementation	O
as	O
cutting-edge	B-KEY
manufacturers	I-KEY
begin	O
to	O
adopt	O
the	O
technology	O
on	O
a	O
widespread	O
basis	O

Computing	O
failure	O
probabilities	O
.	O
Applications	O
to	O
reliability	O
analysis	O
The	O
paper	O
presents	O
one	O
method	O
for	O
calculating	O
failure	O
probabilities	O
with	O
applications	O
to	O
reliability	O
analysis	O
.	O
The	O
method	O
is	O
based	O
on	O
transforming	O
the	O
initial	O
set	O
of	O
variables	O
to	O
a	O
n-dimensional	B-KEY
uniform	I-KEY
random	I-KEY
variable	I-KEY
in	O
the	O
unit	B-KEY
hypercube	I-KEY
,	O
together	O
with	O
the	O
limit	B-KEY
condition	I-KEY
set	O
and	O
calculating	O
the	O
associated	O
probability	O
using	O
a	O
recursive	B-KEY
method	I-KEY
based	O
on	O
the	O
Gauss-Legendre	B-KEY
quadrature	I-KEY
formulas	I-KEY
to	O
calculate	O
the	O
resulting	O
multiple	O
integrals	O
.	O
An	O
example	O
of	O
application	O
is	O
used	O
to	O
illustrate	O
the	O
proposed	O
method	O

Code	B-KEY
generator	I-KEY
for	O
the	O
HPF	B-KEY
Library	O
and	O
Fortran	O
95	O
transformational	O
functions	O
One	O
of	O
the	O
language	O
features	O
of	O
the	O
core	O
language	O
of	O
HPF	B-KEY
2.0	O
-LRB-	O
High	B-KEY
Performance	I-KEY
Fortran	I-KEY
-RRB-	O
is	O
the	O
HPF	B-KEY
Library	O
.	O
The	O
HPF	B-KEY
Library	O
consists	O
of	O
55	O
generic	O
functions	O
.	O
The	O
implementation	O
of	O
this	O
library	O
presents	O
the	O
challenge	O
that	O
all	O
data	B-KEY
types	I-KEY
,	O
data	O
kinds	O
,	O
array	O
ranks	O
and	O
input	O
distributions	O
need	O
to	O
be	O
supported	O
.	O
For	O
instance	O
,	O
more	O
than	O
2	O
billion	O
separate	O
functions	O
are	O
required	O
to	O
support	O
COPY-SCATTER	O
fully	O
.	O
The	O
efficient	O
support	O
of	O
these	O
billions	O
of	O
specific	O
functions	O
is	O
one	O
of	O
the	O
outstanding	O
problems	O
of	O
HPF	B-KEY
.	O
We	O
have	O
solved	O
this	O
problem	O
by	O
developing	O
a	O
library	B-KEY
generator	I-KEY
which	O
utilizes	O
the	O
mechanism	O
of	O
parameterized	B-KEY
templates	I-KEY
.	O
This	O
mechanism	O
allows	O
the	O
procedures	O
to	O
be	O
instantiated	O
at	O
compile	O
time	O
for	O
arguments	O
with	O
a	O
specific	O
type	O
,	O
kind	O
,	O
rank	O
and	O
distribution	O
over	O
a	O
specific	O
processor	O
array	O
.	O
We	O
describe	O
the	O
algorithms	O
used	O
in	O
the	O
different	O
library	B-KEY
functions	I-KEY
.	O
The	O
implementation	O
gives	O
the	O
ease	O
of	O
generating	O
a	O
large	O
number	O
of	O
library	O
routines	O
from	O
a	O
single	O
template	O
.	O
The	O
templates	O
can	O
be	O
extended	O
with	O
special	O
code	O
for	O
specific	O
combinations	O
of	O
the	O
input	O
arguments	O
.	O
We	O
describe	O
in	O
detail	O
the	O
implementation	O
and	O
performance	O
of	O
the	O
matrix	B-KEY
multiplication	I-KEY
template	O
for	O
the	O
Fujitsu	O
VPP5000	O
platform	O

A	O
hybrid	B-KEY
ML-EM	I-KEY
algorithm	I-KEY
for	O
calculation	O
of	O
maximum	B-KEY
likelihood	I-KEY
estimates	I-KEY
in	O
semiparametric	O
shared	O
frailty	O
models	O
This	O
paper	O
describes	O
a	O
generalised	O
hybrid	B-KEY
ML-EM	I-KEY
algorithm	I-KEY
for	O
the	O
calculation	O
of	O
maximum	B-KEY
likelihood	I-KEY
estimates	I-KEY
in	O
semiparametric	O
shared	O
frailty	O
models	O
,	O
the	O
Cox	B-KEY
proportional	I-KEY
hazard	I-KEY
models	I-KEY
with	O
hazard	B-KEY
functions	I-KEY
multiplied	O
by	O
a	O
-LRB-	O
parametric	O
-RRB-	O
frailty	O
random	O
variable	O
.	O
This	O
hybrid	O
method	O
is	O
much	O
faster	O
than	O
the	O
standard	O
EM	O
method	O
and	O
faster	O
than	O
the	O
standard	O
direct	O
maximum	O
likelihood	O
method	O
-LRB-	O
ML	O
,	O
Newton-Raphson	O
-RRB-	O
for	O
large	O
samples	O
.	O
We	O
have	O
previously	O
applied	O
this	O
method	O
to	O
semiparametric	O
shared	O
gamma	O
frailty	O
models	O
,	O
and	O
verified	O
by	O
simulations	B-KEY
the	O
asymptotic	O
and	O
small	O
sample	O
statistical	O
properties	O
of	O
the	O
frailty	B-KEY
variance	I-KEY
estimates	I-KEY
.	O
Let	O
theta	O
/	O
sub	O
0	O
/	O
be	O
the	O
true	O
value	O
of	O
the	O
frailty	O
variance	O
parameter	O
.	O
Then	O
the	O
asymptotic	B-KEY
distribution	I-KEY
is	O
normal	O
for	O
theta	O
/	O
sub	O
0	O
/	O
>	O
0	O
while	O
it	O
is	O
a	O
50-50	O
mixture	O
between	O
a	O
point	O
mass	O
at	O
zero	O
and	O
a	O
normal	B-KEY
random	I-KEY
variable	I-KEY
on	O
the	O
positive	O
axis	O
for	O
theta	O
/	O
sub	O
0	O
/	O
=	O
0	O
.	O
For	O
small	O
samples	O
,	O
simulations	B-KEY
suggest	O
that	O
the	O
frailty	B-KEY
variance	I-KEY
estimates	I-KEY
are	O
approximately	O
distributed	O
as	O
an	O
x	O
-	O
-LRB-	O
100-x	O
-RRB-	O
%	O
mixture	O
,	O
0	O
<	O
or	O
=	O
x	O
<	O
or	O
=	O
50	O
,	O
between	O
a	O
point	O
mass	O
at	O
zero	O
and	O
a	O
normal	B-KEY
random	I-KEY
variable	I-KEY
on	O
the	O
positive	O
axis	O
even	O
for	O
theta	O
/	O
sub	O
0	O
/	O
>	O
0	O
.	O
We	O
apply	O
this	O
method	O
and	O
verify	O
by	O
simulations	B-KEY
these	O
statistical	O
results	O
for	O
semiparametric	B-KEY
shared	I-KEY
log-normal	I-KEY
frailty	I-KEY
models	I-KEY
.	O
We	O
also	O
apply	O
the	O
semiparametric	O
shared	O
gamma	O
and	O
log-normal	O
frailty	O
models	O
to	O
Busselton	B-KEY
Health	I-KEY
Study	I-KEY
coronary	B-KEY
heart	I-KEY
disease	I-KEY
data	I-KEY

Center-crossing	B-KEY
recurrent	I-KEY
neural	I-KEY
networks	I-KEY
for	O
the	O
evolution	O
of	O
rhythmic	O
behavior	O
A	O
center-crossing	B-KEY
recurrent	I-KEY
neural	I-KEY
network	I-KEY
is	O
one	O
in	O
which	O
the	O
null	O
-LRB-	O
hyper	O
-RRB-	O
surfaces	O
of	O
each	O
neuron	O
intersect	O
at	O
their	O
exact	O
centers	O
of	O
symmetry	B-KEY
,	O
ensuring	O
that	O
each	O
neuron	O
's	O
activation	B-KEY
function	I-KEY
is	O
centered	O
over	O
the	O
range	O
of	O
net	O
inputs	O
that	O
it	O
receives	O
.	O
We	O
demonstrate	O
that	O
relative	O
to	O
a	O
random	B-KEY
initial	I-KEY
population	I-KEY
,	O
seeding	O
the	O
initial	O
population	O
of	O
an	O
evolutionary	B-KEY
search	I-KEY
with	O
center-crossing	O
networks	O
significantly	O
improves	O
both	O
the	O
frequency	O
and	O
the	O
speed	O
with	O
which	O
high-fitness	B-KEY
oscillatory	I-KEY
circuits	I-KEY
evolve	O
on	O
a	O
simple	O
walking	O
task	O
.	O
The	O
improvement	O
is	O
especially	O
striking	O
at	O
low	B-KEY
mutation	I-KEY
variances	I-KEY
.	O
Our	O
results	O
suggest	O
that	O
seeding	O
with	O
center-crossing	O
networks	O
may	O
often	O
be	O
beneficial	O
,	O
since	O
a	O
wider	O
range	O
of	O
dynamics	O
is	O
more	O
likely	O
to	O
be	O
easily	O
accessible	O
from	O
a	O
population	O
of	O
center-crossing	O
networks	O
than	O
from	O
a	O
population	O
of	O
random	B-KEY
networks	I-KEY

Towards	O
a	O
NMR	B-KEY
implementation	I-KEY
of	O
a	O
quantum	B-KEY
lattice	I-KEY
gas	I-KEY
algorithm	I-KEY
Recent	O
theoretical	O
results	O
suggest	O
that	O
an	O
array	O
of	O
quantum	B-KEY
information	I-KEY
processors	I-KEY
communicating	O
via	O
classical	O
channels	O
can	O
be	O
used	O
to	O
solve	O
fluid	B-KEY
dynamics	I-KEY
problems	I-KEY
.	O
Quantum	O
lattice-gas	O
algorithms	O
-LRB-	O
QLGA	O
-RRB-	O
running	O
on	O
such	O
architectures	O
have	O
been	O
shown	O
to	O
solve	O
the	O
diffusion	B-KEY
equation	I-KEY
and	O
the	O
nonlinear	B-KEY
Burgers	I-KEY
equations	I-KEY
.	O
In	O
this	O
report	O
,	O
we	O
describe	O
progress	O
towards	O
an	O
ensemble	O
nuclear	B-KEY
magnetic	I-KEY
resonance	I-KEY
-LRB-	O
NMR	O
-RRB-	O
implementation	O
of	O
a	O
QLGA	O
that	O
solves	O
the	O
diffusion	B-KEY
equation	I-KEY
.	O
The	O
methods	O
rely	O
on	O
NMR	O
techniques	O
to	O
encode	O
an	O
initial	O
mass	O
density	O
into	O
an	O
ensemble	O
of	O
two-qubit	O
quantum	B-KEY
information	I-KEY
processors	I-KEY
.	O
Using	O
standard	O
pulse	O
techniques	O
,	O
the	O
mass	O
density	O
can	O
then	O
manipulated	O
and	O
evolved	O
through	O
the	O
steps	O
of	O
the	O
algorithm	O
.	O
We	O
provide	O
the	O
experimental	O
results	O
of	O
our	O
first	O
attempt	O
to	O
realize	O
the	O
NMR	B-KEY
implementation	I-KEY
.	O
The	O
results	O
qualitatively	O
follow	O
the	O
ideal	O
simulation	O
,	O
but	O
the	O
observed	O
implementation	O
errors	O
highlight	O
the	O
need	O
for	O
improved	O
control	O

Edison	O
's	O
direct	O
current	O
influenced	O
``	O
Broadway	O
''	O
show	O
lighting	O
During	O
the	O
early	O
decades	O
of	O
the	O
20	O
th	O
century	O
,	O
midtown	O
Manhattan	B-KEY
in	O
New	B-KEY
York	I-KEY
City	I-KEY
developed	O
an	O
extensive	O
underground	O
direct	O
current	O
-LRB-	O
DC	O
-RRB-	O
power	O
distribution	O
system	O
.	O
This	O
was	O
a	O
result	O
of	O
the	O
original	O
introduction	O
of	O
direct	O
current	O
by	O
Thomas	O
Edison	O
's	O
pioneering	O
Pearl	O
Street	O
Station	O
in	O
1882	O
.	O
The	O
availability	O
of	O
DC	O
power	O
in	O
the	O
theater	B-KEY
district	I-KEY
,	O
led	O
to	O
the	O
perpetuation	O
of	O
an	O
archaic	O
form	O
of	O
stage	B-KEY
lighting	I-KEY
control	I-KEY
through	O
nearly	O
three-quarters	O
of	O
the	O
20	O
th	O
century	O
.	O
This	O
control	O
device	O
was	O
known	O
as	O
a	O
``	O
resistance	B-KEY
dimmer	I-KEY
.	O
''	O
It	O
was	O
essentially	O
a	O
series-connected	B-KEY
rheostat	I-KEY
,	O
but	O
it	O
was	O
wound	O
with	O
a	O
special	O
resistance	O
``	O
taper	O
''	O
so	O
as	O
to	O
provide	O
a	O
uniform	O
change	O
in	O
the	O
apparent	B-KEY
light	I-KEY
output	I-KEY
of	O
typical	O
incandescent	B-KEY
lamps	I-KEY
throughout	O
the	O
travel	O
of	O
its	O
manually	O
operated	O
arm	O
.	O
The	O
development	O
and	O
use	O
of	O
DC	B-KEY
powered	I-KEY
stage	I-KEY
lighting	I-KEY
is	O
discussed	O
in	O
this	O
article	O

Development	O
of	O
a	O
real-time	B-KEY
monitoring	I-KEY
system	I-KEY
This	O
paper	O
describes	O
a	O
pattern	O
recognition	O
-LRB-	O
PR	O
-RRB-	O
technique	O
,	O
which	O
uses	O
learning	B-KEY
vector	I-KEY
quantization	I-KEY
-LRB-	O
LVQ	B-KEY
-RRB-	O
.	O
This	O
method	O
is	O
adapted	O
for	O
practical	O
application	O
to	O
solve	O
problems	O
in	O
the	O
area	O
of	O
condition	B-KEY
monitoring	I-KEY
and	O
fault	B-KEY
diagnosis	I-KEY
where	O
a	O
number	O
of	O
fault	B-KEY
signatures	I-KEY
are	O
involved	O
.	O
In	O
these	O
situations	O
,	O
the	O
aim	O
is	O
health	B-KEY
monitoring	I-KEY
,	O
including	O
identification	O
of	O
deterioration	O
of	O
the	O
healthy	O
condition	O
and	O
identification	O
of	O
causes	O
of	O
the	O
failure	O
in	O
real-time	O
.	O
For	O
this	O
reason	O
a	O
fault	B-KEY
database	I-KEY
is	O
developed	O
which	O
contains	O
the	O
collected	O
information	O
about	O
various	O
states	O
of	O
operation	O
of	O
the	O
system	O
in	O
the	O
form	O
of	O
pattern	B-KEY
vectors	I-KEY
.	O
The	O
task	O
of	O
the	O
real-time	B-KEY
monitoring	I-KEY
system	I-KEY
is	O
to	O
correlate	O
patterns	O
of	O
unknown	O
faults	O
with	O
the	O
known	O
fault	B-KEY
signatures	I-KEY
in	O
the	O
fault	B-KEY
database	I-KEY
.	O
This	O
will	O
determine	O
cause	O
of	O
failure	O
and	O
degree	O
of	O
deterioration	O
of	O
the	O
system	O
under	O
test	O
.	O
The	O
problem	O
of	O
fault	B-KEY
diagnosis	I-KEY
may	O
involve	O
a	O
large	O
number	O
of	O
patterns	O
and	O
large	B-KEY
sampling	I-KEY
time	I-KEY
,	O
which	O
affects	O
the	O
learning	O
stage	O
of	O
neural	B-KEY
networks	I-KEY
.	O
The	O
study	O
here	O
also	O
aims	O
to	O
find	O
a	O
fast	B-KEY
learning	I-KEY
model	I-KEY
of	O
neural	B-KEY
networks	I-KEY
for	O
instances	O
when	O
a	O
high	O
number	O
of	O
patterns	O
and	O
numerous	O
processing	O
elements	O
are	O
involved	O
.	O
It	O
begins	O
searching	O
for	O
an	O
appropriate	O
solution	O
.	O
The	O
study	O
is	O
extended	O
to	O
the	O
enforcement	O
learning	O
models	O
and	O
considers	O
LVQ	B-KEY
as	O
a	O
network	O
emerged	O
from	O
the	O
competitive	B-KEY
learning	I-KEY
model	I-KEY
through	O
enforcement	B-KEY
training	I-KEY
.	O
Finally	O
,	O
tests	O
show	O
an	O
accuracy	O
of	O
92.3	O
per	O
cent	O
in	O
the	O
fault	B-KEY
diagnostic	I-KEY
capability	I-KEY
of	O
the	O
technique	O

The	O
Information	O
Age	O
interview	O
-	O
Capital	B-KEY
One	I-KEY
Credit	B-KEY
card	I-KEY
company	I-KEY
Capital	B-KEY
One	I-KEY
attributes	O
its	O
rapid	O
customer	B-KEY
growth	I-KEY
to	O
the	O
innovative	O
use	O
of	O
cutting-edge	B-KEY
technology	I-KEY
.	O
European	O
CIO	O
Catherine	O
Doran	O
talks	O
about	O
the	O
systems	O
that	O
have	O
fuelled	O
that	O
runaway	O
success	O

Knowledge	B-KEY
flow	I-KEY
management	I-KEY
for	O
distributed	B-KEY
team	I-KEY
software	I-KEY
development	I-KEY
Cognitive	B-KEY
cooperation	I-KEY
is	O
often	O
neglected	O
in	O
current	O
team	O
software	O
development	O
processes	O
.	O
This	O
issue	O
becomes	O
more	O
important	O
than	O
ever	O
when	O
team	O
members	O
are	O
globally	O
distributed	O
.	O
This	O
paper	O
presents	O
a	O
notion	O
of	O
knowledge	O
flow	O
and	O
the	O
related	O
management	O
mechanism	O
for	O
realizing	O
an	O
ordered	B-KEY
knowledge	I-KEY
sharing	I-KEY
and	O
cognitive	B-KEY
cooperation	I-KEY
in	O
a	O
geographically	O
distributed	B-KEY
team	I-KEY
software	I-KEY
development	I-KEY
process	O
.	O
The	O
knowledge	O
flow	O
can	O
carry	O
and	O
accumulate	O
knowledge	O
when	O
it	O
goes	O
through	O
from	O
one	O
team	O
member	O
to	O
another	O
.	O
The	O
coordination	O
between	O
the	O
knowledge	O
flow	O
process	O
and	O
the	O
workflow	B-KEY
process	I-KEY
of	O
a	O
development	O
team	O
provides	O
a	O
new	O
way	O
to	O
improve	O
traditional	O
team	O
software	O
development	O
processes	O
.	O
A	O
knowledge	B-KEY
grid	I-KEY
platform	I-KEY
has	O
been	O
implemented	O
to	O
support	O
the	O
knowledge	B-KEY
flow	I-KEY
management	I-KEY
across	O
the	O
Internet	B-KEY

The	O
year	O
of	O
the	O
racehorse	O
-LSB-	O
China	B-KEY
Telecom	O
-RSB-	O
Does	O
China	B-KEY
really	O
offer	O
the	O
telecoms	B-KEY
industry	I-KEY
a	O
route	O
out	O
of	O
the	O
telecoms	O
slump	O
?	O
According	O
to	O
the	O
Chinese	O
government	O
it	O
has	O
yet	O
to	O
receive	O
a	O
single	O
application	O
from	O
foreign	O
companies	O
looking	O
to	O
invest	O
in	O
the	O
country	O
's	O
domestic	O
telecoms	O
sector	O
since	O
the	O
country	O
joined	O
the	O
World	O
Trade	O
Organisation	O

Generalized	B-KEY
spatio-chromatic	I-KEY
diffusion	I-KEY
A	O
framework	O
for	O
diffusion	B-KEY
of	O
color	B-KEY
images	I-KEY
is	O
presented	O
.	O
The	O
method	O
is	O
based	O
on	O
the	O
theory	O
of	O
thermodynamics	B-KEY
of	O
irreversible	B-KEY
transformations	I-KEY
which	O
provides	O
a	O
suitable	O
basis	O
for	O
designing	O
correlations	O
between	O
the	O
different	O
color	B-KEY
channels	I-KEY
.	O
More	O
precisely	O
,	O
we	O
derive	O
an	O
equation	O
for	O
color	B-KEY
evolution	I-KEY
which	O
comprises	O
a	O
purely	O
spatial	O
diffusive	B-KEY
term	O
and	O
a	O
nonlinear	O
term	O
that	O
depends	O
on	O
the	O
interactions	O
among	O
color	O
channels	O
over	O
space	O
.	O
We	O
apply	O
the	O
proposed	O
equation	O
to	O
images	O
represented	O
in	O
several	O
color	O
spaces	O
,	O
such	O
as	O
RGB	B-KEY
,	O
CIELAB	B-KEY
,	O
Opponent	B-KEY
colors	I-KEY
,	O
and	O
IHS	B-KEY

Shortchanging	O
the	O
future	O
of	O
information	O
technology	O
:	O
the	O
untapped	B-KEY
resource	I-KEY
Building	O
on	O
ideas	O
from	O
a	O
virtual	B-KEY
workshop	I-KEY
and	O
additional	O
input	O
from	O
the	O
scientific	O
community	O
,	O
the	O
CISE	B-KEY
Directorate	I-KEY
at	O
the	O
National	B-KEY
Science	I-KEY
Foundation	I-KEY
established	O
the	O
Information	B-KEY
Technology	I-KEY
Workforce	I-KEY
Program	I-KEY
-LRB-	O
ITWF	B-KEY
-RRB-	O
in	O
March	O
2000	O
to	O
support	O
a	O
broad	O
set	O
of	O
scientific	B-KEY
research	I-KEY
studies	I-KEY
focused	O
on	O
the	O
under-representation	O
of	O
women	O
and	O
minorities	O
in	O
the	O
information	O
technology	O
workforce	O
.	O
In	O
this	O
paper	O
,	O
we	O
explore	O
various	O
approaches	O
that	O
the	O
funded	O
researchers	O
are	O
taking	O
to	O
address	O
the	O
problem	O
of	O
women	O
in	O
information	O
technology	O
.	O
We	O
begin	O
with	O
a	O
brief	O
history	B-KEY
of	O
the	O
ITWF	B-KEY
,	O
and	O
then	O
focus	O
on	O
some	O
of	O
the	O
research	O
projects	O
in	O
terms	O
of	O
their	O
goals	O
,	O
approaches	O
,	O
and	O
expected	O
outcomes	O

Simultaneous	O
iterative	O
reconstruction	O
of	O
emission	O
and	O
attenuation	O
images	O
in	O
positron	O
emission	O
tomography	O
from	O
emission	O
data	O
only	O
For	O
quantitative	O
image	B-KEY
reconstruction	I-KEY
in	O
positron	B-KEY
emission	I-KEY
tomography	I-KEY
attenuation	I-KEY
correction	I-KEY
is	O
mandatory	O
.	O
In	O
case	O
that	O
no	O
data	O
are	O
available	O
for	O
the	O
calculation	O
of	O
the	O
attenuation	B-KEY
correction	I-KEY
factors	I-KEY
one	O
can	O
try	O
to	O
determine	O
them	O
from	O
the	O
emission	O
data	O
alone	O
.	O
However	O
,	O
it	O
is	O
not	O
clear	O
if	O
the	O
information	O
content	O
is	O
sufficient	O
to	O
yield	O
an	O
adequate	O
attenuation	O
correction	O
together	O
with	O
a	O
satisfactory	O
activity	B-KEY
distribution	I-KEY
.	O
Therefore	O
,	O
we	O
determined	O
the	O
log	B-KEY
likelihood	I-KEY
distribution	I-KEY
for	O
a	O
thorax	B-KEY
phantom	I-KEY
depending	O
on	O
the	O
choice	O
of	O
attenuation	O
and	O
activity	B-KEY
pixel	I-KEY
values	I-KEY
to	O
measure	O
the	O
crosstalk	B-KEY
between	O
both	O
.	O
In	O
addition	O
an	O
iterative	B-KEY
image	I-KEY
reconstruction	I-KEY
-LRB-	O
one-dimensional	O
Newton-type	O
algorithm	O
with	O
a	O
maximum	O
likelihood	O
estimator	O
-RRB-	O
,	O
which	O
simultaneously	O
reconstructs	O
the	O
images	O
of	O
the	O
activity	O
distribution	O
and	O
the	O
attenuation	O
coefficients	O
is	O
used	O
to	O
demonstrate	O
the	O
problems	O
and	O
possibilities	O
of	O
such	O
a	O
reconstruction	O
.	O
As	O
result	O
we	O
show	O
that	O
for	O
a	O
change	O
of	O
the	O
log	O
likelihood	O
in	O
the	O
range	O
of	O
statistical	B-KEY
noise	I-KEY
,	O
the	O
associated	O
change	O
in	O
the	O
activity	O
value	O
of	O
a	O
structure	O
is	O
between	O
6	O
%	O
and	O
263	O
%	O
.	O
In	O
addition	O
,	O
we	O
show	O
that	O
it	O
is	O
not	O
possible	O
to	O
choose	O
the	O
best	O
maximum	O
on	O
the	O
basis	O
of	O
the	O
log	O
likelihood	O
when	O
a	O
regularization	O
is	O
used	O
,	O
because	O
the	O
coupling	O
between	O
different	O
structures	O
mediated	O
by	O
the	O
-LRB-	O
smoothing	B-KEY
-RRB-	O
regularization	O
prevents	O
an	O
adequate	O
solution	O
due	O
to	O
crosstalk	B-KEY
.	O
We	O
conclude	O
that	O
taking	O
into	O
account	O
the	O
attenuation	B-KEY
information	I-KEY
in	O
the	O
emission	O
data	O
improves	O
the	O
performance	O
of	O
image	B-KEY
reconstruction	I-KEY
with	O
respect	O
to	O
the	O
bias	O
of	O
the	O
activities	O
,	O
however	O
,	O
the	O
reconstruction	O
still	O
is	O
not	O
quantitative	O

A	O
conflict	O
between	O
language	O
and	O
atomistic	O
information	O
Fred	O
Dretske	O
and	O
Jerry	O
Fodor	O
are	O
responsible	O
for	O
popularizing	O
three	O
well-known	O
theses	O
in	O
contemporary	O
philosophy	B-KEY
of	I-KEY
mind	I-KEY
:	O
the	O
thesis	O
of	O
Information-Based	B-KEY
Semantics	I-KEY
-LRB-	O
IBS	B-KEY
-RRB-	O
,	O
the	O
thesis	O
of	O
Content	B-KEY
Atomism	I-KEY
-LRB-	O
Atomism	O
-RRB-	O
and	O
the	O
thesis	O
of	O
the	O
Language	B-KEY
of	I-KEY
Thought	I-KEY
-LRB-	O
LOT	B-KEY
-RRB-	O
.	O
LOT	B-KEY
concerns	O
the	O
semantically	O
relevant	O
structure	O
of	O
representations	O
involved	O
in	O
cognitive	B-KEY
states	I-KEY
such	O
as	O
beliefs	B-KEY
and	O
desires	B-KEY
.	O
It	O
maintains	O
that	O
all	O
such	O
representations	O
must	O
have	O
syntactic	O
structures	O
mirroring	O
the	O
structure	O
of	O
their	O
contents	O
.	O
IBS	B-KEY
is	O
a	O
thesis	O
about	O
the	O
nature	O
of	O
the	O
relations	O
that	O
connect	O
cognitive	O
representations	O
and	O
their	O
parts	O
to	O
their	O
contents	O
-LRB-	O
semantic	O
relations	O
-RRB-	O
.	O
It	O
holds	O
that	O
these	O
relations	O
supervene	O
solely	O
on	O
relations	O
of	O
the	O
kind	O
that	O
support	O
information	O
content	O
,	O
perhaps	O
with	O
some	O
help	O
from	O
logical	O
principles	O
of	O
combination	O
.	O
Atomism	O
is	O
a	O
thesis	O
about	O
the	O
nature	O
of	O
the	O
content	O
of	O
simple	O
symbols	O
.	O
It	O
holds	O
that	O
each	O
substantive	O
simple	O
symbol	O
possesses	O
its	O
content	O
independently	O
of	O
all	O
other	O
symbols	O
in	O
the	O
representational	O
system	O
.	O
I	O
argue	O
that	O
Dretske	O
's	O
and	O
Fodor	O
's	O
theories	O
are	O
false	O
and	O
that	O
their	O
falsehood	O
results	O
from	O
a	O
conflict	O
IBS	B-KEY
and	O
Atomism	O
,	O
on	O
the	O
one	O
hand	O
,	O
and	O
LOT	B-KEY
,	O
on	O
the	O
other	O

From	O
information	B-KEY
gateway	I-KEY
to	O
digital	B-KEY
library	I-KEY
management	I-KEY
system	I-KEY
:	O
a	O
case	O
analysis	O
This	O
paper	O
discusses	O
the	O
design	O
,	O
implementation	O
and	O
evolution	O
of	O
the	O
Cornell	B-KEY
University	I-KEY
Library	I-KEY
Gateway	I-KEY
using	O
the	O
case	O
analysis	O
method	O
.	O
It	O
diagnoses	O
the	O
Gateway	O
within	O
the	O
conceptual	O
framework	O
of	O
definitions	O
and	O
best	O
practices	O
associated	O
with	O
information	B-KEY
gateways	I-KEY
,	O
portals	B-KEY
,	O
and	O
emerging	O
digital	B-KEY
library	I-KEY
management	I-KEY
systems	I-KEY
,	O
in	O
particular	O
the	O
product	O
ENCompass	B-KEY

Use	O
of	O
neural	O
networks	O
in	O
the	O
analysis	O
of	O
particle	B-KEY
size	I-KEY
distribution	I-KEY
by	O
laser	B-KEY
diffraction	I-KEY
:	O
tests	O
with	O
different	O
particle	O
systems	O
The	O
application	O
of	O
forward	B-KEY
light	I-KEY
scattering	I-KEY
methods	O
for	O
estimating	O
the	O
particle	B-KEY
size	I-KEY
distribution	I-KEY
-LRB-	O
PSD	O
-RRB-	O
is	O
usually	O
limited	O
by	O
the	O
occurrence	O
of	O
multiple	B-KEY
scattering	I-KEY
,	O
which	O
affects	O
the	O
angular	B-KEY
distribution	I-KEY
of	I-KEY
light	I-KEY
in	O
highly	O
concentrated	O
suspensions	O
,	O
thus	O
resulting	O
in	O
false	O
calculations	O
by	O
the	O
conventionally	O
adopted	O
algorithms	O
.	O
In	O
this	O
paper	O
,	O
a	O
previously	O
proposed	O
neural	O
network-based	O
method	O
is	O
tested	O
with	O
different	O
particle	O
systems	O
,	O
in	O
order	O
to	O
evaluate	O
its	O
applicability	O
.	O
In	O
the	O
first	O
step	O
of	O
the	O
study	O
,	O
experiments	O
were	O
carried	O
out	O
with	O
solid-liquid	B-KEY
suspensions	I-KEY
having	O
different	O
characteristics	O
of	O
particle	O
shape	O
and	O
size	O
distribution	O
,	O
under	O
varying	O
solid	O
concentrations	O
.	O
The	O
experimental	O
results	O
,	O
consisting	O
of	O
the	O
angular	B-KEY
distribution	I-KEY
of	I-KEY
light	I-KEY
intensity	O
,	O
particle	O
shape	O
and	O
suspension	O
concentration	O
,	O
were	O
used	O
as	O
input	O
data	O
in	O
the	O
fitting	O
of	O
neural	B-KEY
network	I-KEY
models	I-KEY
-LRB-	O
NN	O
-RRB-	O
that	O
replaced	O
the	O
optical	O
model	O
to	O
provide	O
the	O
PSD	O
.	O
The	O
reference	O
values	O
of	O
particle	O
shape	O
and	O
PSD	O
for	O
the	O
NN	O
fitting	O
were	O
based	O
on	O
image	B-KEY
analysis	I-KEY
.	O
Comparisons	O
between	O
the	O
PSD	O
values	O
computed	O
by	O
the	O
NN	O
model	O
and	O
the	O
reference	O
values	O
indicate	O
that	O
the	O
method	O
can	O
be	O
used	O
in	O
monitoring	O
the	O
PSD	O
of	O
particles	O
with	O
different	O
shapes	O
in	O
highly	O
concentrated	O
suspensions	O
,	O
thus	O
extending	O
the	O
range	O
of	O
application	O
of	O
forward	O
laser	B-KEY
diffraction	I-KEY
to	O
a	O
number	O
of	O
systems	O
with	O
industrial	O
interest	O

Completeness	B-KEY
of	O
timed	B-KEY
mu	I-KEY
CRL	I-KEY
Previously	O
a	O
straightforward	O
extension	O
of	O
the	O
process	B-KEY
algebra	I-KEY
mu	O
CRL	O
was	O
proposed	O
to	O
explicitly	O
deal	O
with	O
time	O
.	O
The	O
process	B-KEY
algebra	I-KEY
mu	O
CRL	O
has	O
been	O
especially	O
designed	O
to	O
deal	O
with	O
data	O
in	O
a	O
process	B-KEY
algebraic	I-KEY
context	O
.	O
Using	O
the	O
features	O
for	O
data	O
,	O
only	O
a	O
minor	O
extension	O
of	O
the	O
language	O
was	O
needed	O
to	O
obtain	O
a	O
very	O
expressive	O
variant	O
of	O
time	O
.	O
Previously	O
it	O
contained	O
syntax	O
,	O
operational	B-KEY
semantics	I-KEY
and	O
axioms	O
characterising	O
timed	B-KEY
mu	I-KEY
CRL	I-KEY
.	O
It	O
did	O
not	O
contain	O
an	O
in	O
depth	O
analysis	O
of	O
theory	O
of	O
timed	B-KEY
mu	I-KEY
CRL	I-KEY
.	O
This	O
paper	O
fills	O
this	O
gap	O
,	O
by	O
providing	O
soundness	O
and	O
completeness	B-KEY
results	O
.	O
The	O
main	O
tool	O
to	O
establish	O
these	O
is	O
a	O
mapping	O
of	O
timed	O
to	O
untimed	O
mu	O
CRL	O
and	O
employing	O
the	O
completeness	B-KEY
results	O
obtained	O
for	O
untimed	O
mu	O
CRL	O

Tablet	B-KEY
PCs	I-KEY
on	O
the	O
way	O
-LSB-	O
publishing	B-KEY
markets	I-KEY
-RSB-	O
Previews	O
of	O
hardware	O
and	O
software	O
look	O
promising	O
for	O
publishing	B-KEY
markets	I-KEY

How	O
closely	O
can	O
a	O
personal	B-KEY
computer	I-KEY
clock	I-KEY
track	O
the	O
UTC	B-KEY
timescale	I-KEY
via	O
the	O
Internet	B-KEY
?	O
Nowadays	O
many	O
software	B-KEY
packages	I-KEY
allow	O
you	O
to	O
keep	O
the	O
clock	O
of	O
your	O
personal	O
computer	O
synchronized	O
to	O
time	B-KEY
servers	I-KEY
spread	O
over	O
the	O
internet	B-KEY
.	O
We	O
present	O
how	O
a	O
didactic	B-KEY
laboratory	I-KEY
can	O
evaluate	O
,	O
in	O
a	O
statistical	B-KEY
sense	I-KEY
,	O
the	O
minimum	O
synch	B-KEY
error	I-KEY
of	O
this	O
process	O
-LRB-	O
the	O
other	O
extreme	O
,	O
the	O
maximum	O
,	O
is	O
guaranteed	O
by	O
the	O
code	O
itself	O
-RRB-	O
.	O
The	O
measurement	O
set-up	O
utilizes	O
the	O
global	B-KEY
positioning	I-KEY
system	I-KEY
satellite	I-KEY
constellation	I-KEY
in	O
`	O
common	O
view	O
'	O
between	O
two	O
similar	O
timing	O
stations	O
:	O
one	O
acts	O
as	O
a	O
time	B-KEY
server	I-KEY
for	O
the	O
other	O
,	O
so	O
the	O
final	B-KEY
timing	I-KEY
difference	I-KEY
at	O
the	O
second	O
station	O
represents	O
the	O
total	O
synch	B-KEY
error	I-KEY
through	O
the	O
internet	B-KEY
.	O
Data	O
recorded	O
over	O
batches	O
of	O
10000	O
samples	O
show	O
a	O
typical	O
RMS	O
value	O
of	O
35	O
ms.	O
This	O
measurement	O
configuration	O
allows	O
students	O
to	O
obtain	O
a	O
much	O
better	O
understanding	O
of	O
the	O
synch	O
task	O
and	O
pushes	O
them	O
,	O
at	O
all	O
times	O
,	O
to	O
look	O
for	O
an	O
experimental	O
verification	O
of	O
data	O
results	O
,	O
even	O
when	O
they	O
come	O
from	O
the	O
most	O
sophisticated	O
`	O
black	B-KEY
boxes	I-KEY
'	O
now	O
readily	O
available	O
off	O
the	O
shelf	O

Application	O
of	O
Sugeno	B-KEY
fuzzy-logic	I-KEY
controller	I-KEY
to	O
the	O
stator	B-KEY
field-oriented	I-KEY
doubly-fed	I-KEY
asynchronous	I-KEY
motor	I-KEY
drive	I-KEY
This	O
study	O
deals	O
with	O
the	O
application	O
of	O
the	O
fuzzy-control	B-KEY
theory	O
to	O
wound-rotor	B-KEY
asynchronous	I-KEY
motor	I-KEY
with	O
both	O
its	O
stator	O
and	O
rotor	O
fed	O
by	O
two	O
PWM	B-KEY
voltage-source	I-KEY
inverters	I-KEY
,	O
in	O
which	O
the	O
system	O
operates	O
in	O
stator	B-KEY
field-oriented	I-KEY
control	I-KEY
.	O
Thus	O
,	O
after	O
determining	O
the	O
model	O
of	O
the	O
machine	O
,	O
we	O
present	O
two	O
types	O
of	O
fuzzy	O
controller	O
:	O
Mamdani	O
and	O
Sugeno	O
controllers	O
.	O
The	O
training	B-KEY
of	O
the	O
last	O
one	O
is	O
carried	O
out	O
starting	O
from	O
the	O
first	O
.	O
Simulation	O
study	O
is	O
conducted	O
to	O
show	O
the	O
effectiveness	O
of	O
the	O
proposed	O
method	O

A	O
new	O
architecture	O
for	O
implementing	O
pipelined	B-KEY
FIR	I-KEY
ADF	I-KEY
based	O
on	O
classification	O
of	O
coefficients	O
In	O
this	O
paper	O
,	O
we	O
propose	O
a	O
new	O
method	O
for	O
implementing	O
pipelined	O
finite-impulse	O
response	O
-LRB-	O
FIR	O
-RRB-	O
adaptive	B-KEY
digital	I-KEY
filter	I-KEY
-LRB-	O
ADF	O
-RRB-	O
,	O
with	O
an	O
aim	O
of	O
reducing	O
the	O
maximum	B-KEY
delay	I-KEY
of	O
the	O
filtering	O
portion	O
of	O
conventional	O
delayed	O
least	O
mean	O
square	O
-LRB-	O
DLMS	O
-RRB-	O
pipelined	O
ADF	O
.	O
We	O
achieve	O
a	O
filtering	O
section	O
with	O
a	O
maximum	B-KEY
delay	I-KEY
of	O
one	O
by	O
simplifying	O
a	O
pre-upsampled	O
and	O
a	O
post-downsampled	O
FIR	O
filter	O
using	O
the	O
concept	O
of	O
classification	O
of	O
coefficients	O
.	O
This	O
reduction	O
is	O
independent	O
of	O
the	O
order	O
of	O
the	O
filter	O
,	O
which	O
is	O
an	O
advantage	O
when	O
the	O
order	O
of	O
the	O
filter	O
is	O
very	O
large	O
,	O
and	O
as	O
a	O
result	O
the	O
method	O
can	O
also	O
be	O
applied	O
to	O
infinite	O
impulse	O
response	O
-LRB-	O
IIR	O
-RRB-	O
filters	O
.	O
Furthermore	O
,	O
when	O
the	O
proposed	O
method	O
is	O
compared	O
with	O
the	O
transpose	O
ADF	O
,	O
which	O
has	O
a	O
filtering	O
section	O
with	O
zero	O
delay	O
,	O
it	O
is	O
realized	O
that	O
it	O
significantly	O
reduces	O
the	O
maximum	B-KEY
delay	I-KEY
associated	O
with	O
updating	O
the	O
coefficients	O
of	O
FIR	O
ADF	O
.	O
The	O
effect	O
of	O
this	O
is	O
that	O
,	O
the	O
proposed	O
method	O
exhibits	O
a	O
higher	O
convergence	B-KEY
speed	I-KEY
in	O
comparison	O
to	O
the	O
transpose	O
FIR	O
ADF	O

Word	B-KEY
spotting	I-KEY
based	O
on	O
a	B-KEY
posterior	I-KEY
measure	I-KEY
of	O
keyword	B-KEY
confidence	I-KEY
In	O
this	O
paper	O
,	O
an	O
approach	O
of	O
keyword	B-KEY
confidence	I-KEY
estimation	O
is	O
developed	O
that	O
well	O
combines	O
acoustic	O
layer	O
scores	O
and	O
syllable-based	O
statistical	O
language	O
model	O
-LRB-	O
LM	O
-RRB-	O
scores	O
.	O
An	O
a	O
posteriori	O
-LRB-	O
AP	O
-RRB-	O
confidence	O
measure	O
and	O
its	O
forward-backward	B-KEY
calculating	I-KEY
algorithm	I-KEY
are	O
deduced	O
.	O
A	O
zero	O
false	O
alarm	O
-LRB-	O
ZFA	O
-RRB-	O
assumption	O
is	O
proposed	O
for	O
evaluating	O
relative	B-KEY
confidence	I-KEY
measures	I-KEY
by	O
word	B-KEY
spotting	I-KEY
task	O
.	O
In	O
a	O
word	B-KEY
spotting	I-KEY
experiment	O
with	O
a	O
vocabulary	O
of	O
240	O
keywords	O
,	O
the	O
keyword	O
accuracy	O
under	O
the	O
AP	O
measure	O
is	O
above	O
94	O
%	O
,	O
which	O
well	O
approaches	O
its	O
theoretical	O
upper	O
limit	O
.	O
In	O
addition	O
,	O
a	O
syllable	B-KEY
lattice	I-KEY
Hidden	I-KEY
Markov	I-KEY
Model	I-KEY
-LRB-	O
SLHMM	O
-RRB-	O
is	O
formulated	O
and	O
a	O
unified	O
view	O
of	O
confidence	B-KEY
estimation	I-KEY
,	O
word	B-KEY
spotting	I-KEY
,	O
optimal	B-KEY
path	I-KEY
search	I-KEY
,	O
and	O
N-best	B-KEY
syllable	I-KEY
re-scoring	I-KEY
is	O
presented	O
.	O
The	O
proposed	O
AP	O
measure	O
can	O
be	O
easily	O
applied	O
to	O
various	O
speech	B-KEY
recognition	I-KEY
systems	I-KEY
as	O
well	O

Genetic	O
algorithm-neural	O
network	O
estimation	O
of	O
Cobb	O
angle	O
from	O
torso	O
asymmetry	O
in	O
scoliosis	O
Scoliosis	O
severity	O
,	O
measured	O
by	O
the	O
Cobb	B-KEY
angle	I-KEY
,	O
was	O
estimated	O
by	O
artificial	B-KEY
neural	I-KEY
network	I-KEY
from	O
indices	O
of	O
torso	B-KEY
surface	I-KEY
asymmetry	I-KEY
using	O
a	O
genetic	B-KEY
algorithm	I-KEY
to	O
select	O
the	O
optimal	O
set	O
of	O
input	B-KEY
torso	I-KEY
indices	I-KEY
.	O
Estimates	O
of	O
the	O
Cobb	B-KEY
angle	I-KEY
were	O
accurate	O
within	O
5	O
degrees	O
in	O
two-thirds	O
,	O
and	O
within	O
10	O
degrees	O
in	O
six-sevenths	O
,	O
of	O
a	O
test	O
set	O
of	O
115	O
scans	O
of	O
48	O
scoliosis	B-KEY
patients	I-KEY
,	O
showing	O
promise	O
for	O
future	O
longitudinal	O
studies	O
to	O
detect	O
scoliosis	B-KEY
progression	I-KEY
without	O
use	O
of	O
X-rays	O

Bisimulation	B-KEY
minimization	I-KEY
and	O
symbolic	B-KEY
model	I-KEY
checking	I-KEY
State	B-KEY
space	I-KEY
minimization	I-KEY
techniques	I-KEY
are	O
crucial	O
for	O
combating	O
state	B-KEY
explosion	I-KEY
.	O
A	O
variety	O
of	O
explicit-state	B-KEY
verification	I-KEY
tools	I-KEY
use	O
bisimulation	B-KEY
minimization	I-KEY
to	O
check	O
equivalence	O
between	O
systems	O
,	O
to	O
minimize	O
components	O
before	O
composition	O
,	O
or	O
to	O
reduce	O
a	O
state	O
space	O
prior	O
to	O
model	O
checking	O
.	O
Experimental	B-KEY
results	I-KEY
on	O
bisimulation	B-KEY
minimization	I-KEY
in	O
symbolic	B-KEY
model	I-KEY
checking	I-KEY
contexts	O
,	O
however	O
,	O
are	O
mixed	O
.	O
We	O
explore	O
bisimulation	B-KEY
minimization	I-KEY
as	O
an	O
optimization	B-KEY
in	O
symbolic	B-KEY
model	I-KEY
checking	I-KEY
of	O
invariance	B-KEY
properties	I-KEY
.	O
We	O
consider	O
three	O
bisimulation	B-KEY
minimization	I-KEY
algorithms	O
.	O
From	O
each	O
,	O
we	O
produce	O
a	O
BDD-based	O
model	O
checker	O
for	O
invariant	O
properties	O
and	O
compare	O
this	O
model	O
checker	O
to	O
a	O
conventional	O
one	O
based	O
on	O
backwards	O
reachability	O
.	O
Our	O
comparisons	O
,	O
both	O
theoretical	O
and	O
experimental	O
,	O
suggest	O
that	O
bisimulation	B-KEY
minimization	I-KEY
is	O
not	O
viable	O
in	O
the	O
context	O
of	O
invariance	B-KEY
verification	I-KEY
,	O
because	O
performing	O
the	O
minimization	O
requires	O
as	O
many	O
,	O
if	O
not	O
more	O
,	O
computational	O
resources	O
as	O
model	O
checking	O
the	O
unminimized	O
system	O
through	O
backwards	B-KEY
reachability	I-KEY

IT	O
challenge	O
:	O
cross	B-KEY
selling	I-KEY
-LSB-	O
finance	O
-RSB-	O
Like	O
most	O
financial	B-KEY
institutions	I-KEY
,	O
FleetBoston	B-KEY
,	O
Fidelity	B-KEY
and	O
Berkshire	B-KEY
Group	I-KEY
of	O
Companies	O
are	O
being	O
charged	O
with	O
developing	O
a	O
strong	O
technology	O
platform	O
that	O
will	O
allow	O
them	O
to	O
cross	B-KEY
sell	I-KEY
their	O
products	O
and	O
services	O
.	O
They	O
discuss	O
their	O
solutions	O
,	O
advice	O
and	O
technology	O
choices	O

Bluetooth	B-KEY
bites	O
back	O
It	O
is	O
now	O
more	O
than	O
four	O
years	O
since	O
we	O
started	O
to	O
hear	O
about	O
Bluetooth	B-KEY
,	O
and	O
from	O
the	O
user	O
's	O
point	O
of	O
view	O
very	O
little	O
seems	O
to	O
have	O
happened	O
since	O
then	O
.	O
Paul	O
Haddlesey	O
looks	O
at	O
the	O
progress	O
,	O
and	O
the	O
role	O
Bluetooth	B-KEY
may	O
eventually	O
play	O
in	O
your	O
firm	O
's	O
communications	B-KEY
strategy	I-KEY

Eliminating	O
recency	O
with	O
self-review	B-KEY
:	O
the	O
case	O
of	O
auditors	O
'	O
`	O
going	O
concern	O
'	O
judgments	O
This	O
paper	O
examines	O
the	O
use	O
of	O
self-review	B-KEY
to	O
debias	O
recency	O
.	O
Recency	O
is	O
found	O
in	O
the	O
`	O
going	O
concern	O
'	O
judgments	O
of	O
staff	B-KEY
auditors	I-KEY
,	O
but	O
is	O
successfully	O
eliminated	O
by	O
the	O
auditor	O
's	O
use	O
of	O
a	O
simple	O
self-review	B-KEY
technique	O
that	O
would	O
be	O
extremely	O
easy	O
to	O
implement	O
in	O
audit	O
practice	O
.	O
Auditors	O
who	O
self-review	B-KEY
are	O
also	O
less	O
inclined	O
to	O
make	O
audit	B-KEY
report	I-KEY
choices	I-KEY
that	O
are	O
inconsistent	O
with	O
their	O
going	O
concern	O
judgments	O
.	O
These	O
results	O
are	O
important	O
because	O
the	O
judgments	O
of	O
staff	B-KEY
auditors	I-KEY
often	O
determine	O
the	O
type	O
and	O
extent	O
of	O
documentation	B-KEY
in	O
audit	B-KEY
workpapers	I-KEY
and	O
serve	O
as	O
preliminary	O
inputs	O
for	O
senior	B-KEY
auditors	I-KEY
'	O
judgments	O
and	O
choices	O
.	O
If	O
staff	B-KEY
auditors	I-KEY
'	O
judgments	O
are	O
affected	O
by	O
recency	O
,	O
the	O
impact	O
of	O
this	O
bias	O
may	O
be	O
impounded	O
in	O
the	O
ultimate	O
judgments	O
and	O
choices	O
of	O
senior	B-KEY
auditors	I-KEY
.	O
Since	O
biased	O
judgments	O
can	O
expose	O
auditors	O
to	O
significant	O
costs	O
involving	O
extended	B-KEY
audit	I-KEY
procedures	I-KEY
,	O
legal	B-KEY
liability	I-KEY
and	O
diminished	B-KEY
reputation	I-KEY
,	O
simple	O
debiasing	O
techniques	O
that	O
reduce	O
this	O
exposure	O
are	O
valuable	O
.	O
The	O
paper	O
also	O
explores	O
some	O
future	O
research	O
needs	O
and	O
other	O
important	O
issues	O
concerning	O
judgment	B-KEY
debiasing	I-KEY
in	O
applied	B-KEY
professional	I-KEY
settings	I-KEY

Keen	O
but	O
confused	O
-LSB-	O
workflow	B-KEY
&	O
content	B-KEY
management	I-KEY
-RSB-	O
IT	O
users	O
find	O
workflow	B-KEY
,	O
content	O
and	O
business	B-KEY
process	I-KEY
management	I-KEY
software	I-KEY
appealing	O
but	O
by	O
no	O
means	O
straightforward	O
to	O
implement	O
.	O
Pat	O
Sweet	O
reports	O
on	O
our	O
latest	O
research	B-KEY

Data	O
quality	O
-	O
unlocking	O
the	O
ROI	O
in	O
CRM	B-KEY
While	O
many	O
organisations	O
realise	O
their	O
most	O
valuable	O
asset	O
is	O
their	O
customers	O
,	O
many	O
more	O
fail	O
to	O
realise	O
the	O
importance	O
of	O
auditing	O
,	O
maintaining	O
and	O
updating	O
the	O
information	O
contained	O
in	O
their	O
customer	B-KEY
databases	I-KEY
.	O
Today	O
's	O
growing	O
awareness	O
in	O
the	O
importance	O
of	O
data	O
quality	O
in	O
relation	O
to	O
CRM	B-KEY
and	O
ROI	O
will	O
help	O
change	O
this	O
attitude	O
.	O
In	O
response	O
,	O
CRM	B-KEY
vendors	O
will	O
follow	O
suit	O
and	O
begin	O
to	O
differentiate	O
themselves	O
by	O
offering	O
data	O
quality	O
as	O
part	O
of	O
an	O
enterprise-wide	O
data	B-KEY
management	I-KEY
methodology	O

Packet	O
promises	O
past	O
&	O
present	O
-LSB-	O
IP	O
switching	O
-RSB-	O
With	O
the	O
death	O
of	O
the	O
competitive	O
carrier	O
market	O
and	O
the	O
significant	O
slashing	O
of	O
RBOC	O
capex	O
budgets	O
,	O
softswitch	B-KEY
vendors	I-KEY
have	O
been	O
forced	O
to	O
retrench	O
.	O
Now	O
instead	O
of	O
focusing	O
primarily	O
on	O
limited	O
Internet	O
off-load	O
applications	O
,	O
packet-based	O
softswitches	O
are	O
set	O
to	O
gel	O
around	O
real	O
user	O
needs	O
for	O
services	O
such	O
as	O
voice	B-KEY
over	I-KEY
IP	I-KEY
and	O
IP	B-KEY
Centrex	I-KEY

The	O
importance	O
of	O
continuity	B-KEY
:	O
a	O
reply	O
to	O
Chris	O
Eliasmith	O
In	O
his	O
reply	O
to	O
Eliasmith	O
-LRB-	O
see	O
ibid.	O
,	O
vol	O
.11	O
,	O
p.417-26	O
,	O
2001	O
-RRB-	O
Poznanski	O
considers	O
how	O
the	O
notion	O
of	O
continuity	B-KEY
of	O
dynamic	B-KEY
representations	I-KEY
serves	O
as	O
a	O
beacon	O
for	O
an	O
integrative	B-KEY
neuroscience	I-KEY
to	O
emerge	O
.	O
He	O
considers	O
how	O
the	O
importance	O
of	O
continuity	B-KEY
has	O
come	O
under	O
attack	O
from	O
Eliasmith	O
-LRB-	O
2001	O
-RRB-	O
who	O
claims	O
:	O
-LRB-	O
i	O
-RRB-	O
continuous	B-KEY
nature	O
of	O
neurons	B-KEY
is	O
not	O
relevant	O
to	O
the	O
information	O
they	O
process	O
,	O
and	O
-LRB-	O
ii	O
-RRB-	O
continuity	B-KEY
is	O
not	O
important	O
for	O
understanding	O
cognition	B-KEY
because	O
the	O
various	O
sources	O
of	O
noise	O
introduce	O
uncertainty	B-KEY
into	O
spike	B-KEY
arrival	I-KEY
times	I-KEY
,	O
so	O
encoding	O
and	O
decoding	O
spike	B-KEY
trains	I-KEY
must	O
be	O
discrete	O
at	O
some	O
level	O

Human	B-KEY
factors	I-KEY
research	O
on	O
data	O
modeling	O
:	O
a	O
review	O
of	O
prior	O
research	O
,	O
an	O
extended	O
framework	O
and	O
future	O
research	O
directions	O
This	O
study	O
reviews	O
and	O
synthesizes	O
human	B-KEY
factors	I-KEY
research	O
on	O
conceptual	B-KEY
data	I-KEY
modeling	I-KEY
.	O
In	O
addition	O
to	O
analyzing	O
the	O
variables	O
used	O
in	O
earlier	O
studies	O
and	O
summarizing	O
the	O
results	O
of	O
this	O
stream	O
of	O
research	O
,	O
we	O
propose	O
a	O
new	O
framework	O
to	O
help	O
with	O
future	B-KEY
efforts	I-KEY
in	O
this	O
area	O
.	O
The	O
study	O
finds	O
that	O
prior	O
research	O
has	O
focused	O
on	O
issues	O
that	O
are	O
relevant	O
when	O
conceptual	O
models	O
are	O
used	O
for	O
communication	O
between	O
systems	O
analysts	O
and	O
developers	O
-LRB-	O
Analyst	B-KEY
Developer	I-KEY
models	I-KEY
-RRB-	O
whereas	O
the	O
issues	O
important	O
for	O
models	O
that	O
are	O
used	O
to	O
facilitate	O
communication	O
between	O
analysts	O
and	O
users	O
-LRB-	O
User-Analyst	B-KEY
models	I-KEY
-RRB-	O
have	O
received	O
little	O
attention	O
and	O
,	O
hence	O
,	O
require	O
a	O
significantly	O
stronger	O
role	O
in	O
future	O
research	O
.	O
In	O
addition	O
,	O
we	O
emphasize	O
the	O
importance	O
of	O
building	O
a	O
strong	O
theoretical	O
foundation	O
and	O
using	O
it	O
to	O
guide	O
future	O
empirical	O
work	O
in	O
this	O
area	O

Fuzzy	O
business	O
-LSB-	O
Halden	B-KEY
Reactor	I-KEY
Project	I-KEY
-RSB-	O
The	O
Halden	B-KEY
Reactor	I-KEY
Project	I-KEY
has	O
developed	O
two	O
systems	O
to	O
investigate	O
how	O
signal	O
validation	O
and	O
thermal	B-KEY
performance	I-KEY
monitoring	I-KEY
techniques	O
can	O
be	O
improved	O
.	O
PEANO	B-KEY
is	O
an	O
online	O
calibration	B-KEY
monitoring	O
system	O
that	O
makes	O
use	O
of	O
artificial	B-KEY
intelligence	I-KEY
techniques	O
.	O
The	O
system	O
has	O
been	O
tested	O
in	O
cooperation	O
with	O
EPRI	O
and	O
Edan	O
Engineering	O
,	O
using	O
real	O
data	O
from	O
a	O
US	O
PWR	B-KEY
plant	O
.	O
These	O
tests	O
showed	O
that	O
PEANO	B-KEY
could	O
reliably	O
assess	O
the	O
performance	O
of	O
the	O
process	O
instrumentation	O
at	O
different	O
plant	O
conditions	O
.	O
Real	O
cases	O
of	O
zero	O
and	O
span	O
drifts	O
were	O
successfully	O
detected	O
by	O
the	O
system	O
.	O
TEMPO	B-KEY
is	O
a	O
system	O
for	O
thermal	B-KEY
performance	I-KEY
monitoring	I-KEY
and	O
optimisation	O
,	O
which	O
relies	O
on	O
plant-wide	O
first	O
principle	O
models	O
.	O
The	O
system	O
has	O
been	O
installed	O
on	O
a	O
Swedish	O
BWR	B-KEY
plant	O
.	O
Results	O
obtained	O
show	O
an	O
overall	O
rms	O
deviation	O
from	O
measured	O
values	O
of	O
a	O
few	O
tenths	O
of	O
a	O
percent	O
,	O
and	O
giving	O
goodness-of-fits	O
in	O
the	O
order	O
of	O
95	O
%	O
.	O
The	O
high	O
accuracy	O
demonstrated	O
is	O
a	O
good	O
basis	O
for	O
detecting	O
possible	O
faults	O
and	O
efficiency	O
losses	O
in	O
steam	B-KEY
turbine	I-KEY
cycles	I-KEY

Resolution	B-KEY
of	O
a	O
current-mode	O
algorithmic	O
analog-to-digital	B-KEY
converter	I-KEY
Errors	O
limiting	O
the	O
resolution	B-KEY
of	O
current-mode	O
algorithmic	O
analog-to-digital	B-KEY
converters	I-KEY
are	O
mainly	O
related	O
to	O
current	O
mirror	O
operation	O
.	O
While	O
systematic	O
errors	O
can	O
be	O
minimized	O
by	O
proper	O
circuit	B-KEY
techniques	I-KEY
,	O
random	O
sources	O
are	O
unavoidable	O
.	O
In	O
this	O
paper	O
a	O
statistical	B-KEY
analysis	I-KEY
of	O
the	O
resolution	B-KEY
of	O
a	O
typical	O
converter	O
is	O
carried	O
out	O
taking	O
into	O
account	O
process	O
tolerances	O
.	O
To	O
support	O
the	O
analysis	O
,	O
a	O
4-bit	O
ADC	O
,	O
realized	O
in	O
a	O
0.35	O
-	O
mu	O
m	O
CMOS	B-KEY
technology	I-KEY
,	O
was	O
exhaustively	O
simulated	O
.	O
Results	O
were	O
found	O
to	O
be	O
in	O
excellent	O
agreement	O
with	O
theoretical	O
derivations	O

Citizen	B-KEY
centric	I-KEY
identity	I-KEY
management	I-KEY
:	O
chip	O
tricks	O
?	O
Accelerating	O
and	O
harmonizing	O
the	O
diffusion	O
and	O
acceptance	O
of	O
electronic	B-KEY
services	I-KEY
in	O
Europe	O
in	O
a	O
secure	O
and	O
practical	O
way	O
has	O
become	O
a	O
priority	O
of	O
several	O
initiatives	O
in	O
the	O
past	O
few	O
years	O
and	O
a	O
critical	O
factor	O
for	O
citizen	O
and	O
business	O
information	O
society	O
services	O
.	O
As	O
identification	O
and	O
authentication	B-KEY
is	O
a	O
critical	O
element	O
in	O
accessing	O
public	O
services	O
the	O
combination	O
of	O
public	B-KEY
key	I-KEY
infrastructure	I-KEY
-LRB-	O
PKI	O
-RRB-	O
and	O
smart	B-KEY
cards	I-KEY
emerges	O
as	O
the	O
solution	O
of	O
choice	O
for	O
eGovernment	O
in	O
Europe	O
.	O
National	O
governments	B-KEY
and	O
private	O
initiatives	O
alike	O
vouch	O
their	O
support	O
for	O
this	O
powerful	O
combination	O
to	O
deliver	O
an	O
essential	O
layer	O
of	O
reliable	O
electronic	B-KEY
services	I-KEY
and	O
address	O
identity	O
requirements	O
in	O
a	O
broad	O
range	O
of	O
application	O
areas	O
.	O
A	O
recent	O
study	O
suggests	O
that	O
several	O
eGovernment	O
implementations	O
point	O
to	O
the	O
direction	O
of	O
electronic	O
citizen	O
identity	O
management	O
as	O
an	O
up	O
and	O
coming	O
challenge	O
.	O
The	O
paper	O
discusses	O
the	O
eGovernment	O
needs	O
for	O
user	B-KEY
identification	I-KEY
applicability	O
and	O
the	O
need	O
for	O
standardization	B-KEY

CRONE	B-KEY
control	I-KEY
:	O
principles	O
and	O
extension	O
to	O
time-variant	B-KEY
plants	I-KEY
with	O
asymptotically	B-KEY
constant	I-KEY
coefficients	I-KEY
The	O
principles	O
of	O
CRONE	B-KEY
control	I-KEY
,	O
a	O
frequency-domain	O
robust	B-KEY
control	I-KEY
design	O
methodology	O
based	O
on	O
fractional	O
differentiation	O
,	O
are	O
presented	O
.	O
Continuous	O
time-variant	B-KEY
plants	I-KEY
with	O
asymptotically	B-KEY
constant	I-KEY
coefficients	I-KEY
are	O
analysed	O
in	O
the	O
frequency	O
domain	O
,	O
through	O
their	O
representation	O
using	O
time-variant	B-KEY
frequency	I-KEY
responses	I-KEY
.	O
A	O
stability	B-KEY
theorem	I-KEY
for	O
feedback	B-KEY
systems	I-KEY
including	O
time-variant	B-KEY
plants	I-KEY
with	O
asymptotically	B-KEY
constant	I-KEY
coefficients	I-KEY
is	O
proposed	O
.	O
Finally	O
,	O
CRONE	B-KEY
control	I-KEY
is	O
extended	O
to	O
robust	B-KEY
control	I-KEY
of	O
these	O
plants	O

WAM!Net	B-KEY
:	O
private	O
pipes	O
for	O
electronic	B-KEY
media	I-KEY
``	O
We	O
are	O
the	O
digital	O
version	O
of	O
FedEx	O
.	O
We	O
offer	O
storage	O
and	O
intelligent	B-KEY
workflow	I-KEY
.	O
''	O
The	O
United	B-KEY
States	I-KEY
military	I-KEY
-	O
especially	O
during	O
war	O
time	O
-	O
is	O
pretty	O
careful	O
about	O
the	O
way	O
it	O
handles	O
its	O
workflow	O
and	O
communications	O
.	O
Before	O
a	O
company	O
is	O
awarded	O
a	O
government	O
contract	O
,	O
the	O
company	O
and	O
its	O
technology	O
are	O
screened	O
and	O
verified	O
.	O
If	O
the	O
technology	O
or	O
its	O
creators	O
are	O
n't	O
trustworthy	O
and	O
secure	O
,	O
chances	O
are	O
they	O
are	O
n't	O
getting	O
by	O
Uncle	O
Sam	O
.	O
Record	O
companies	O
and	O
publishing	O
houses	O
tend	O
to	O
feel	O
the	O
same	O
way	O
.	O
After	O
all	O
,	O
security	O
is	O
just	O
as	O
important	O
to	O
a	O
record	O
executive	O
as	O
it	O
is	O
to	O
a	O
Navy	O
commander	O
.	O
WAM!Net	B-KEY
,	O
a	O
Wide-Area	B-KEY
Media	I-KEY
network	I-KEY
-LRB-	O
hence	O
,	O
the	O
name	O
-RRB-	O
passes	O
muster	O
with	O
both	O
.	O
The	O
company	O
,	O
which	O
employs	O
about	O
320	O
employees	O
around	O
the	O
world	O
,	O
has	O
15000	O
customers	O
including	O
the	O
US	O
Navy	O
and	O
a	O
host	O
of	O
record	B-KEY
labels	I-KEY
,	O
publishing	B-KEY
companies	I-KEY
,	O
healthcare	B-KEY
providers	I-KEY
,	O
and	O
advertising	B-KEY
agencies	I-KEY
,	O
all	O
of	O
whom	O
use	O
its	O
network	O
as	O
a	O
way	O
to	O
transport	O
,	O
store	O
,	O
and	O
receive	O
data	O
.	O
``	O
We	O
are	O
the	O
digital	O
version	O
of	O
FedEx	O
.	O
We	O
offer	O
storage	O
and	O
intelligent	B-KEY
workflow	I-KEY
,	O
''	O
says	O
Murad	O
Velani	O
,	O
executive	O
vice	O
president	O
of	O
sales	O
and	O
marketing	O
for	O
WAM!Net	B-KEY
.	O
``	O
We	O
started	O
out	O
as	O
purely	O
transport	O
and	O
we	O
've	O
become	O
a	O
digital	B-KEY
platform	I-KEY
.	O
''	O

Positional	B-KEY
control	O
of	O
pneumatic	O
manipulators	O
for	O
construction	O
tasks	O
This	O
paper	O
describes	O
solutions	O
that	O
can	O
be	O
applied	O
to	O
pneumatic	B-KEY
manipulator	I-KEY
problems	O
in	O
positioning	B-KEY
,	O
both	O
for	O
angle	B-KEY
trajectories	I-KEY
and	O
for	O
long	O
linear	O
trajectories	O
,	O
used	O
in	O
construction	B-KEY
tasks	I-KEY
.	O
Optimal	O
positioning	B-KEY
of	O
a	O
pneumatic	B-KEY
manipulator	I-KEY
along	O
angle	B-KEY
trajectories	I-KEY
with	O
minimum	O
control	O
energy	O
consumption	O
is	O
given	O
.	O
The	O
implementation	O
of	O
the	O
control	O
system	O
is	O
presented	O
.	O
Control	O
algorithms	O
for	O
a	O
long	B-KEY
linear	I-KEY
trajectory	I-KEY
manipulator	I-KEY
based	O
on	O
two-phase	O
and	O
three-phase	B-KEY
motion	I-KEY
modes	I-KEY
of	O
the	O
end-effector	O
are	O
investigated	O
.	O
Conventional	O
and	O
fuzzy	B-KEY
logic	I-KEY
controls	I-KEY
of	O
a	O
pneumatic	B-KEY
manipulator	I-KEY
were	O
applied	O
and	O
experimental	O
testing	O
was	O
carried	O
out	O
.	O
The	O
obtained	O
results	O
allow	O
widening	O
the	O
application	O
range	O
of	O
pneumatic	B-KEY
manipulators	I-KEY
in	O
construction	O
,	O
particularly	O
in	O
gantry	B-KEY
type	I-KEY
machines	I-KEY

Completion	B-KEY
to	O
involution	B-KEY
and	O
semidiscretisations	B-KEY
We	O
discuss	O
the	O
relation	O
between	O
the	O
completion	B-KEY
to	O
involution	B-KEY
of	O
linear	B-KEY
over-determined	I-KEY
systems	I-KEY
of	O
partial	B-KEY
differential	I-KEY
equations	I-KEY
with	O
constant	B-KEY
coefficients	I-KEY
and	O
the	O
properties	O
of	O
differential	B-KEY
algebraic	I-KEY
equations	I-KEY
obtained	O
by	O
their	O
semidiscretisation	B-KEY
.	O
For	O
a	O
certain	O
class	O
of	O
``	O
weakly	O
over-determined	O
''	O
systems	O
,	O
we	O
show	O
that	O
the	O
differential	B-KEY
algebraic	I-KEY
equations	I-KEY
do	O
not	O
contain	O
hidden	O
constraints	O
,	O
if	O
and	O
only	O
if	O
the	O
original	O
partial	O
differential	O
system	O
is	O
involutive	B-KEY
.	O
We	O
also	O
demonstrate	O
how	O
the	O
formal	O
theory	O
can	O
be	O
used	O
to	O
obtain	O
an	O
existence	O
and	O
uniqueness	B-KEY
theorem	I-KEY
for	O
smooth	O
solutions	O
of	O
strongly	B-KEY
hyperbolic	I-KEY
systems	I-KEY
and	O
to	O
estimate	O
the	O
drift	O
off	O
the	O
constraints	O
,	O
if	O
an	O
underlying	O
equation	O
is	O
numerically	O
solved	O
.	O
Finally	O
,	O
we	O
show	O
for	O
general	O
linear	O
systems	O
how	O
the	O
index	B-KEY
of	O
differential	B-KEY
algebraic	I-KEY
equations	I-KEY
obtained	O
by	O
semidiscretisations	B-KEY
can	O
be	O
predicted	O
from	O
the	O
result	O
of	O
a	O
completion	B-KEY
of	O
the	O
partial	O
differential	O
system	O

Use	O
of	O
Bayesian	B-KEY
Belief	I-KEY
Networks	I-KEY
when	O
combining	O
disparate	O
sources	O
of	O
information	O
in	O
the	O
safety	B-KEY
assessment	I-KEY
of	O
software-based	B-KEY
systems	I-KEY
The	O
paper	O
discusses	O
how	O
disparate	O
sources	O
of	O
information	O
can	O
be	O
combined	O
in	O
the	O
safety	B-KEY
assessment	I-KEY
of	O
software-based	B-KEY
systems	I-KEY
.	O
The	O
emphasis	O
is	O
put	O
on	O
an	O
emerging	O
methodology	O
,	O
relevant	O
for	O
intelligent	B-KEY
product-support	I-KEY
systems	I-KEY
,	O
to	O
combine	O
information	O
about	O
disparate	O
evidences	O
systematically	O
based	O
on	O
Bayesian	B-KEY
Belief	I-KEY
Networks	I-KEY
.	O
The	O
objective	O
is	O
to	O
show	O
the	O
link	O
between	O
basic	O
information	O
and	O
the	O
confidence	O
one	O
can	O
have	O
in	O
a	O
system	O
.	O
How	O
one	O
combines	O
the	O
Bayesian	O
Belief	O
Net	O
-LRB-	O
BBN	O
-RRB-	O
method	O
with	O
a	O
software	B-KEY
safety	I-KEY
standard	I-KEY
-LRB-	O
RTCA/DO	O
-178	O
B	O
,	O
-RRB-	O
for	O
safety	B-KEY
assessment	I-KEY
of	O
software-based	B-KEY
systems	I-KEY
is	O
also	O
discussed	O
.	O
Finally	O
,	O
the	O
applicability	O
of	O
the	O
BBN	O
methodology	O
and	O
experiences	O
from	O
cooperative	O
research	O
work	O
together	O
with	O
Kongsberg	O
Defence	O
&	O
Aerospace	O
and	O
Det	O
Norske	O
Veritas	O
,	O
and	O
ongoing	O
research	O
with	O
VTT	O
Automation	O
are	O
presented	O

Methods	O
for	O
outlier	B-KEY
detection	I-KEY
in	O
prediction	O
If	O
a	O
prediction	B-KEY
sample	I-KEY
is	O
different	O
from	O
the	O
calibration	B-KEY
samples	I-KEY
,	O
it	O
can	O
be	O
considered	O
as	O
an	O
outlier	O
in	O
prediction	O
.	O
In	O
this	O
work	O
,	O
two	O
techniques	O
,	O
the	O
use	O
of	O
uncertainty	B-KEY
estimation	I-KEY
and	O
the	O
convex	B-KEY
hull	I-KEY
method	I-KEY
are	O
studied	O
to	O
detect	O
such	O
prediction	O
outliers	O
.	O
Classical	O
techniques	O
-LRB-	O
Mahalanobis	B-KEY
distance	I-KEY
and	O
X-residuals	B-KEY
-RRB-	O
,	O
potential	B-KEY
functions	I-KEY
and	O
robust	B-KEY
techniques	I-KEY
are	O
used	O
for	O
comparison	O
.	O
It	O
is	O
concluded	O
that	O
the	O
combination	O
of	O
the	O
convex	B-KEY
hull	I-KEY
method	I-KEY
and	O
uncertainty	B-KEY
estimation	I-KEY
offers	O
a	O
practical	O
way	O
for	O
detecting	O
outliers	O
in	O
prediction	O
.	O
By	O
adding	O
the	O
potential	B-KEY
function	I-KEY
method	O
,	O
inliers	B-KEY
can	O
also	O
be	O
detected	O

When	O
the	O
unexpected	O
happens	O
-LSB-	O
disaster	B-KEY
planning	I-KEY
in	O
banks	O
-RSB-	O
A	O
business	O
disruption	O
can	O
be	O
as	O
simple	O
as	O
a	O
power	O
failure	O
or	O
as	O
complex	O
as	O
a	O
terrorist	O
attack	O
.	O
Regardless	O
,	O
you	O
will	O
need	O
to	O
have	O
a	O
plan	B-KEY
to	O
minimize	O
interruptions	O
to	O
both	O
your	O
bank	B-KEY
and	O
your	O
customers	O
.	O
Marketers	O
have	O
a	O
role	O
in	O
this	O
readiness	O
process	O

Incorporating	O
multi-leaf	O
collimator	O
leaf	O
sequencing	O
into	O
iterative	O
IMRT	O
optimization	O
Intensity	B-KEY
modulated	I-KEY
radiation	I-KEY
therapy	I-KEY
-LRB-	O
IMRT	O
-RRB-	O
treatment	B-KEY
planning	I-KEY
typically	O
considers	O
beam	B-KEY
optimization	I-KEY
and	O
beam	B-KEY
delivery	I-KEY
as	O
separate	O
tasks	O
.	O
Following	O
optimization	O
,	O
a	O
multi-leaf	O
collimator	O
-LRB-	O
MLC	O
-RRB-	O
or	O
other	O
beam	B-KEY
delivery	I-KEY
device	O
is	O
used	O
to	O
generate	O
fluence	B-KEY
patterns	I-KEY
for	O
patient	O
treatment	O
delivery	O
.	O
Due	O
to	O
limitations	O
and	O
characteristics	O
of	O
the	O
MLC	O
,	O
the	O
deliverable	O
intensity	O
distributions	O
often	O
differ	O
from	O
those	O
produced	O
by	O
the	O
optimizer	O
,	O
leading	O
to	O
differences	O
between	O
the	O
delivered	O
and	O
the	O
optimized	O
doses	O
.	O
Objective	B-KEY
function	I-KEY
parameters	I-KEY
are	O
then	O
adjusted	O
empirically	O
,	O
and	O
the	O
plan	O
is	O
reoptimized	O
to	O
achieve	O
a	O
desired	O
deliverable	B-KEY
dose	I-KEY
distribution	I-KEY
.	O
The	O
resulting	O
plan	O
,	O
though	O
usually	O
acceptable	O
,	O
may	O
not	O
be	O
the	O
best	O
achievable	O
.	O
A	O
method	O
has	O
been	O
developed	O
to	O
incorporate	O
the	O
MLC	O
restrictions	O
into	O
the	O
optimization	O
process	O
.	O
Our	O
in-house	O
IMRT	O
system	O
has	O
been	O
modified	O
to	O
include	O
the	O
calculation	O
of	O
the	O
deliverable	O
intensity	O
into	O
the	O
optimizer	O
.	O
In	O
this	O
process	O
,	O
prior	O
to	O
dose	O
calculation	O
,	O
the	O
MLC	O
leaf	O
sequencer	O
is	O
used	O
to	O
convert	O
intensities	O
to	O
dynamic	O
MLC	O
sequences	O
,	O
from	O
which	O
the	O
deliverable	O
intensities	O
are	O
then	O
determined	O
.	O
All	O
other	O
optimization	O
steps	O
remain	O
the	O
same	O
.	O
To	O
evaluate	O
the	O
effectiveness	O
of	O
deliverable-based	O
optimization	O
,	O
17	O
patient	O
cases	O
have	O
been	O
studied	O
.	O
Compared	O
with	O
standard	O
optimization	O
plus	O
conversion	O
to	O
deliverable	O
beams	O
,	O
deliverable-based	O
optimization	O
results	O
show	O
improved	O
isodose	O
coverage	O
and	O
a	O
reduced	O
dose	O
to	O
critical	O
structures	O
.	O
Deliverable-based	O
optimization	O
results	O
are	O
close	O
to	O
the	O
original	O
nondeliverable	O
optimization	O
results	O
,	O
suggesting	O
that	O
IMRT	O
can	O
overcome	O
the	O
MLC	O
limitations	O
by	O
adjusting	O
individual	O
beamlets	O
.	O
The	O
use	O
of	O
deliverable-based	O
optimization	O
may	O
reduce	O
the	O
need	O
for	O
empirical	B-KEY
adjustment	I-KEY
of	O
objective	B-KEY
function	I-KEY
parameters	I-KEY
and	O
reoptimization	O
of	O
a	O
plan	O
to	O
achieve	O
desired	O
results	O

Valuing	O
corporate	O
debt	O
:	O
the	O
effect	O
of	O
cross-holdings	O
of	O
stock	O
and	O
debt	O
We	O
have	O
developed	O
a	O
simple	O
approach	O
to	O
valuing	O
risky	O
corporate	O
debt	O
when	O
corporations	O
own	O
securities	B-KEY
issued	O
by	O
other	O
corporations	O
.	O
We	O
assume	O
that	O
corporate	O
debt	O
can	O
be	O
valued	O
as	O
an	O
option	B-KEY
on	O
corporate	B-KEY
business	I-KEY
asset	I-KEY
value	I-KEY
,	O
and	O
derive	O
payoff	B-KEY
functions	I-KEY
when	O
there	O
exist	O
cross-holdings	O
of	O
stock	O
or	O
debt	O
between	O
two	O
firms	O
.	O
Next	O
we	O
show	O
that	O
payoff	B-KEY
functions	I-KEY
with	O
multiple	B-KEY
cross-holdings	I-KEY
can	O
be	O
solved	O
by	O
the	O
contraction	O
principle	O
.	O
The	O
payoff	B-KEY
functions	I-KEY
which	O
we	O
derive	O
provide	O
a	O
number	O
of	O
insights	O
about	O
the	O
risk	O
structure	O
of	O
company	O
cross-holdings	O
.	O
First	O
,	O
the	O
Modigliani-Miller	B-KEY
theorem	I-KEY
can	O
obtain	O
when	O
there	O
exist	O
cross-holdings	O
between	O
firms	O
.	O
Second	O
,	O
by	O
establishing	O
cross-shareholdings	B-KEY
each	O
of	O
stock	O
holders	O
distributes	O
a	O
part	O
of	O
its	O
payoff	O
values	O
to	O
the	O
bond	B-KEY
holder	I-KEY
of	O
the	O
other	O
's	O
firm	O
,	O
so	O
that	O
both	O
firms	O
can	O
decrease	O
credit	B-KEY
risks	I-KEY
by	O
cross-shareholdings	B-KEY
.	O
In	O
the	O
numerical	O
examples	O
,	O
we	O
show	O
that	O
the	O
correlation	B-KEY
in	O
firms	O
can	O
be	O
a	O
critical	O
condition	O
for	O
reducing	O
credit	B-KEY
risk	I-KEY
by	O
cross-holdings	O
of	O
stock	O
using	O
Monte	B-KEY
Carlo	I-KEY
simulation	I-KEY
.	O
Moreover	O
,	O
we	O
show	O
we	O
can	O
calculate	O
the	O
default	O
spread	O
easily	O
when	O
complicated	O
cross-holdings	O
exist	O
,	O
and	O
find	O
which	O
shares	O
are	O
beneficial	O
or	O
disadvantageous	O

Processing	O
of	O
complexly	B-KEY
shaped	I-KEY
multiply	I-KEY
connected	I-KEY
domains	I-KEY
in	O
finite	B-KEY
element	I-KEY
mesh	I-KEY
generation	I-KEY
Large	O
number	O
of	O
finite	B-KEY
element	I-KEY
models	I-KEY
in	O
modern	O
materials	O
science	O
and	O
engineering	O
is	O
defined	O
on	O
complexly	O
shaped	O
domains	O
,	O
quite	O
often	O
multiply	O
connected	O
.	O
Generation	O
of	O
quality	O
finite	O
element	O
meshes	O
on	O
such	O
domains	O
,	O
especially	O
in	O
cases	O
when	O
the	O
mesh	O
must	O
be	O
100	O
%	O
quadrilateral	O
,	O
is	O
highly	O
problematic	O
.	O
This	O
paper	O
describes	O
mathematical	O
fundamentals	O
and	O
practical	O
-	O
implementation	O
of	O
a	O
powerful	O
method	O
and	O
algorithm	O
allowing	O
transformation	O
of	O
multiply	O
connected	O
domains	O
of	O
arbitrary	B-KEY
geometrical	I-KEY
complexity	I-KEY
into	O
a	O
set	B-KEY
of	I-KEY
simple	I-KEY
domains	I-KEY
;	O
the	O
latter	O
can	O
then	O
be	O
processed	O
by	O
broadly	O
available	O
finite	B-KEY
element	I-KEY
mesh	I-KEY
generators	I-KEY
.	O
The	O
developed	O
method	O
was	O
applied	O
to	O
a	O
number	O
of	O
complex	O
geometries	O
,	O
including	O
those	O
arising	O
in	O
analysis	O
of	O
parasitic	B-KEY
inductances	I-KEY
and	O
capacitances	O
in	O
printed	B-KEY
circuit	I-KEY
boards	I-KEY
.	O
The	O
quality	O
of	O
practical	O
results	O
produced	O
by	O
the	O
method	O
and	O
its	O
programming	B-KEY
implementation	I-KEY
provide	O
evidence	O
that	O
the	O
algorithm	O
can	O
be	O
applied	O
to	O
other	O
finite	B-KEY
element	I-KEY
models	I-KEY
with	O
various	O
physical	O
backgrounds	O

International	O
swinging	O
:	O
making	O
Swing	O
components	O
locale-sensitive	O
Although	O
Java	B-KEY
and	O
its	O
GUI	B-KEY
library	I-KEY
Swing	O
provide	O
software	O
developers	O
with	O
a	O
highly	O
customizable	O
framework	O
for	O
creating	O
truly	O
``	O
international	O
''	O
applications	O
,	O
the	O
Swing	B-KEY
library	I-KEY
is	O
not	O
sensitive	O
to	O
locale	B-KEY
switches	I-KEY
:	O
it	O
can	O
not	O
automatically	O
change	O
an	O
application	O
's	O
appearance	O
to	O
conform	O
to	O
the	O
conventions	O
of	O
a	O
specific	O
locale	O
at	O
run	O
time	O
.	O
Several	O
types	O
of	O
applications	O
benefit	O
from	O
the	O
ability	O
to	O
easily	O
switch	O
the	O
language	O
at	O
run	O
time	O
.	O
Training	O
applications	O
and	O
other	O
programs	O
that	O
run	O
on	O
computers	O
in	O
public	O
spaces	O
-LRB-	O
such	O
as	O
libraries	O
,	O
airports	O
,	O
or	O
government	O
offices	O
-RRB-	O
may	O
need	O
to	O
support	O
multiple	O
languages	O
.	O
Other	O
applications	O
-LRB-	O
like	O
travel	B-KEY
dictionaries	I-KEY
or	O
translation	B-KEY
programs	I-KEY
-RRB-	O
are	O
inherently	O
multilingual	O
and	O
are	O
specifically	O
designed	O
to	O
support	O
users	O
of	O
dissimilar	O
tongues	O
.	O
Such	O
applications	O
would	O
greatly	O
benefit	O
if	O
the	O
user-interface	B-KEY
language	I-KEY
could	O
be	O
customized	O
at	O
run	O
time	O
.	O
The	O
article	O
shows	O
you	O
how	O
to	O
customize	O
Swing	O
to	O
support	O
locale	B-KEY
switching	I-KEY
at	O
run	O
time	O
.	O
The	O
author	O
has	O
created	O
a	O
new	O
look-and-feel	O
called	O
the	O
MLMetalLookandFeel	B-KEY
-LRB-	O
where	O
ML	O
stands	O
for	O
multilingual	O
-RRB-	O
.	O
This	O
new	O
look-and-feel	O
extends	O
the	O
standard	O
Metal	O
look-and-feel	O
but	O
is	O
locale-sensitive	O
at	O
run	O
time	O

Searching	O
a	O
scalable	B-KEY
approach	I-KEY
to	O
cerebellar	B-KEY
based	I-KEY
control	I-KEY
Decades	O
of	O
research	O
into	O
the	O
structure	O
and	O
function	O
of	O
the	O
cerebellum	O
have	O
led	O
to	O
a	O
clear	O
understanding	O
of	O
many	O
of	O
its	O
cells	O
,	O
as	O
well	O
as	O
how	O
learning	O
might	O
take	O
place	O
.	O
Furthermore	O
,	O
there	O
are	O
many	O
theories	O
on	O
what	O
signals	O
the	O
cerebellum	O
operates	O
on	O
,	O
and	O
how	O
it	O
works	O
in	O
concert	O
with	O
other	O
parts	O
of	O
the	O
nervous	B-KEY
system	I-KEY
.	O
Nevertheless	O
,	O
the	O
application	O
of	O
computational	B-KEY
cerebellar	I-KEY
models	I-KEY
to	O
the	O
control	O
of	O
robot	O
dynamics	O
remains	O
in	O
its	O
infant	O
state	O
.	O
To	O
date	O
,	O
few	O
applications	O
have	O
been	O
realized	O
.	O
The	O
currently	O
emerging	O
family	O
of	O
light-weight	B-KEY
robots	I-KEY
poses	O
a	O
new	O
challenge	O
to	O
robot	B-KEY
control	I-KEY
:	O
due	O
to	O
their	O
complex	O
dynamics	O
traditional	O
methods	O
,	O
depending	O
on	O
a	O
full	O
analysis	O
of	O
the	O
dynamics	O
of	O
the	O
system	O
,	O
are	O
no	O
longer	O
applicable	O
since	O
the	O
joints	O
influence	O
each	O
other	O
dynamics	O
during	O
movement	O
.	O
Can	O
artificial	O
cerebellar	O
models	O
compete	O
here	O
?	O

Multicriterion	O
optimization	O
of	O
composite	B-KEY
laminates	I-KEY
for	O
maximum	B-KEY
failure	I-KEY
margins	I-KEY
with	O
an	O
interactive	B-KEY
descent	I-KEY
algorithm	I-KEY
An	O
interactive	B-KEY
multicriterion	I-KEY
optimization	I-KEY
method	O
for	O
composite	B-KEY
laminates	I-KEY
subjected	O
to	O
multiple	B-KEY
loading	I-KEY
conditions	I-KEY
is	O
introduced	O
.	O
Laminate	O
margins	O
to	O
initial	O
failure	O
-LRB-	O
first	B-KEY
ply	I-KEY
failure	I-KEY
,	O
FPF	O
-RRB-	O
with	O
respect	O
of	O
the	O
applied	O
loading	O
conditions	O
are	O
treated	O
as	O
criteria	O
.	O
The	O
original	O
problem	O
is	O
reduced	O
to	O
a	O
,	O
bicriterion	B-KEY
problem	I-KEY
by	O
introducing	O
parameters	O
to	O
combine	O
criteria	O
in	O
a	O
linear	O
manner	O
.	O
The	O
problem	O
is	O
solved	O
by	O
using	O
an	O
interactive	B-KEY
descent	I-KEY
algorithm	I-KEY
.	O
Both	O
the	O
conditions	O
required	O
for	O
a	O
discrete	B-KEY
procedure	I-KEY
to	O
converge	B-KEY
towards	O
a	O
Pareto	B-KEY
optimum	I-KEY
and	O
numerical	O
examples	O
are	O
given	O

Sigma	O
-	O
admissible	O
families	O
over	O
linear	B-KEY
orders	I-KEY
Admissible	O
sets	O
of	O
the	O
form	O
HYP	O
-LRB-	O
M	O
-RRB-	O
,	O
where	O
M	O
is	O
a	O
recursively	B-KEY
saturated	I-KEY
system	I-KEY
,	O
are	O
treated	O
.	O
We	O
provide	O
descriptions	O
of	O
subsets	O
M	O
,	O
which	O
are	O
Sigma	O
/	O
sub	O
*	O
/	O
-	O
sets	O
in	O
HYP	O
-LRB-	O
M	O
-RRB-	O
,	O
and	O
of	O
families	O
of	O
subsets	O
M	O
,	O
which	O
form	O
Sigma	O
-	O
regular	O
families	O
in	O
HYP	O
-LRB-	O
M	O
-RRB-	O
,	O
in	O
terms	O
of	O
the	O
concept	O
of	O
being	O
fundamental	O
couched	O
in	O
the	O
article	O
.	O
Fundamental	B-KEY
subsets	I-KEY
and	O
families	O
are	O
characterized	O
for	O
models	O
of	O
dense	B-KEY
linear	I-KEY
orderings	I-KEY

Computing	O
2002	O
:	O
democracy	B-KEY
,	O
education	O
,	O
and	O
the	O
future	B-KEY
Computer	O
scientists	O
,	O
computer	O
engineers	O
,	O
information	O
technologists	O
,	O
and	O
their	O
collective	O
products	O
have	O
grown	O
and	O
changed	O
in	O
quantity	O
,	O
quality	O
,	O
and	O
nature	O
.	O
In	O
the	O
first	O
decade	O
of	O
this	O
new	O
century	O
,	O
it	O
should	O
become	O
apparent	O
to	O
everyone	O
that	O
the	O
computing	O
and	O
information	O
fields	O
,	O
broadly	O
defined	O
,	O
will	O
have	O
a	O
profound	O
impact	O
on	O
every	O
element	O
of	O
every	O
person	O
's	O
life	O
.	O
The	O
author	O
considers	O
how	O
women	B-KEY
and	O
girls	B-KEY
of	O
the	O
world	O
have	O
been	O
neither	O
educated	O
for	O
computing	O
nor	O
served	O
by	O
computing	O
.	O
Globally	O
,	O
women	B-KEY
's	O
participation	O
in	O
computer	O
science	O
grew	O
for	O
a	O
while	O
,	O
then	O
dropped	O
precipitously	O
.	O
Computing	O
,	O
science	O
,	O
engineering	O
,	O
and	O
society	B-KEY
will	O
suffer	O
if	O
this	O
decline	O
continues	O
,	O
because	O
women	B-KEY
have	O
different	O
perspectives	O
on	O
technology	O
,	O
what	O
it	O
is	O
important	O
for	O
,	O
how	O
it	O
should	O
be	O
built	O
,	O
which	O
projects	O
should	O
be	O
funded	O
,	O
and	O
so	O
on	O
.	O
To	O
create	O
a	O
positive	O
future	B-KEY
,	O
to	O
assure	O
that	O
women	B-KEY
equally	O
influence	O
the	O
future	B-KEY
,	O
computing	O
education	O
must	O
change	O

Automated	O
cerebrum	O
segmentation	O
from	O
three-dimensional	O
sagittal	O
brain	O
MR	O
images	O
We	O
present	O
a	O
fully	B-KEY
automated	I-KEY
cerebrum	I-KEY
segmentation	I-KEY
algorithm	I-KEY
for	O
full	O
three-dimensional	O
sagittal	O
brain	O
MR	O
images	O
.	O
First	O
,	O
cerebrum	O
segmentation	O
from	O
a	O
midsagittal	B-KEY
brain	I-KEY
MR	I-KEY
image	I-KEY
is	O
performed	O
utilizing	O
landmarks	B-KEY
,	O
anatomical	B-KEY
information	I-KEY
,	O
and	O
a	O
connectivity-based	B-KEY
threshold	I-KEY
segmentation	I-KEY
algorithm	I-KEY
as	O
previously	O
reported	O
.	O
Recognizing	O
that	O
the	O
cerebrum	O
in	O
laterally	B-KEY
adjacent	I-KEY
slices	I-KEY
tends	O
to	O
have	O
similar	O
size	O
and	O
shape	O
,	O
we	O
use	O
the	O
cerebrum	O
segmentation	O
result	O
from	O
the	O
midsagittal	B-KEY
brain	I-KEY
MR	I-KEY
image	I-KEY
as	O
a	O
mask	O
to	O
guide	O
cerebrum	O
segmentation	O
in	O
adjacent	O
lateral	O
slices	O
in	O
an	O
iterative	O
fashion	O
.	O
This	O
masking	B-KEY
operation	I-KEY
yields	O
a	O
masked	O
image	O
-LRB-	O
preliminary	O
cerebrum	O
segmentation	O
-RRB-	O
for	O
the	O
next	O
lateral	O
slice	O
,	O
which	O
may	O
truncate	O
brain	O
region	O
-LRB-	O
s	O
-RRB-	O
.	O
Truncated	O
regions	O
are	O
restored	O
by	O
first	O
finding	O
end	O
points	O
of	O
their	O
boundaries	O
,	O
by	O
comparing	O
the	O
mask	O
image	O
and	O
masked	B-KEY
image	I-KEY
boundaries	I-KEY
,	O
and	O
then	O
applying	O
a	O
connectivity-based	B-KEY
algorithm	I-KEY
.	O
The	O
resulting	O
final	O
extracted	O
cerebrum	O
image	O
for	O
this	O
slice	O
is	O
then	O
used	O
as	O
a	O
mask	O
for	O
the	O
next	O
lateral	O
slice	O
.	O
The	O
algorithm	O
yielded	O
satisfactory	O
fully	O
automated	O
cerebrum	O
segmentations	O
in	O
three-dimensional	O
sagittal	O
brain	O
MR	O
images	O
,	O
and	O
had	O
performance	O
superior	O
to	O
conventional	O
edge	O
detection	O
algorithms	O
for	O
segmentation	O
of	O
cerebrum	O
from	O
3D	O
sagittal	O
brain	O
MR	O
images	O

Arranging	O
solid	B-KEY
balls	I-KEY
to	O
represent	O
a	O
graph	O
By	O
solid	B-KEY
balls	I-KEY
,	O
we	O
mean	O
a	O
set	O
of	O
balls	O
in	O
R/sup	O
3	O
/	O
no	O
two	O
of	O
which	O
can	O
penetrate	O
each	O
other	O
.	O
Every	O
finite	B-KEY
graph	I-KEY
G	O
can	O
be	O
represented	O
by	O
arranging	O
solid	B-KEY
balls	I-KEY
in	O
the	O
following	O
way	O
:	O
Put	O
red	O
balls	O
in	O
R/sup	O
3	O
/	O
,	O
one	O
for	O
each	O
vertex	O
of	O
G	O
,	O
and	O
connect	O
two	O
red	O
balls	O
by	O
a	O
chain	O
when	O
they	O
correspond	O
to	O
a	O
pair	O
of	O
adjacent	B-KEY
vertices	I-KEY
of	O
G	O
,	O
where	O
a	O
chain	O
means	O
a	O
finite	B-KEY
sequence	I-KEY
of	O
blue	O
solid	B-KEY
balls	I-KEY
in	O
which	O
each	O
consecutive	O
balls	O
are	O
tangent	O
.	O
-LRB-	O
We	O
may	O
omit	O
the	O
chain	O
if	O
the	O
two	O
red	O
balls	O
are	O
already	O
tangent	O
.	O
-RRB-	O
The	O
ball	O
number	O
b	O
-LRB-	O
G	O
-RRB-	O
of	O
G	O
is	O
the	O
minimum	O
number	O
of	O
balls	O
-LRB-	O
red	O
and	O
blue	O
-RRB-	O
necessary	O
to	O
represent	O
G	O
.	O
If	O
we	O
put	O
the	O
balls	O
and	O
chains	O
on	O
a	O
table	O
so	O
that	O
all	O
balls	O
sit	O
on	O
the	O
table	O
,	O
then	O
the	O
minimum	O
number	O
of	O
balls	O
for	O
G	O
is	O
denoted	O
by	O
bT	O
-LRB-	O
G	O
-RRB-	O
.	O
Among	O
other	O
things	O
,	O
we	O
prove	O
that	O
b	O
-LRB-	O
K/sub	O
6	O
/	O
-RRB-	O
=	O
8	O
,	O
b	O
-LRB-	O
K/sub	O
7	O
/	O
-RRB-	O
=	O
13	O
and	O
b/sub	O
T	O
/	O
-LRB-	O
K/sub	O
5	O
/	O
-RRB-	O
=	O
8	O
,	O
b/sub	O
T	O
/	O
-LRB-	O
K/sub	O
6	O
/	O
-RRB-	O
=	O
14	O
.	O
We	O
also	O
prove	O
that	O
c/sub	O
1/n/sup	O
3	O
/	O
<	O
b	O
-LRB-	O
K/sub	O
n	O
/	O
-RRB-	O
<	O
c/sub	O
2/n/sup	O
3	O
/	O
log	O
n	O
,	O
c/sub	O
3/n/sup	O
4	O
/	O
/	O
log	O
n	O
<	O
b/sub	O
T	O
/	O
-LRB-	O
K/sub	O
n	O
/	O
-RRB-	O
<	O
c/sub	O
4/n/sup	O
4	O
/	O

An	O
intelligent	B-KEY
tutoring	I-KEY
system	I-KEY
for	O
a	O
power	O
plant	O
simulator	O
In	O
this	O
paper	O
,	O
an	O
intelligent	B-KEY
tutoring	I-KEY
system	I-KEY
-LRB-	O
ITS	O
-RRB-	O
is	O
proposed	O
for	O
a	O
power	O
plant	O
simulator	O
.	O
With	O
a	O
well	O
designed	O
ITS	O
,	O
the	O
need	O
for	O
an	O
instructor	O
is	O
minimized	O
and	O
the	O
operator	O
may	O
readily	O
and	O
efficiently	O
take	O
,	O
in	O
real-time	O
,	O
the	O
control	O
of	O
simulator	O
with	O
appropriate	O
messages	O
he	O
-LRB-	O
she	O
-RRB-	O
gets	O
from	O
the	O
tutoring	O
system	O
.	O
Using	O
SIMULINK	B-KEY
and	O
based	O
on	O
object	B-KEY
oriented	I-KEY
programming	I-KEY
-LRB-	O
OOP	O
-RRB-	O
and	O
C	B-KEY
programming	I-KEY
language	I-KEY
,	O
a	O
fossil-fuelled	B-KEY
power	I-KEY
plant	I-KEY
simulator	I-KEY
with	O
an	O
ITS	O
is	O
proposed	O
.	O
Promising	O
results	O
are	O
demonstrated	O
for	O
a	O
typical	O
power	O
plant	O

A	O
comparison	O
of	O
different	O
decision	B-KEY
algorithms	I-KEY
used	O
in	O
volumetric	B-KEY
storm	I-KEY
cells	I-KEY
classification	I-KEY
Decision	B-KEY
algorithms	I-KEY
useful	O
in	O
classifying	O
meteorological	B-KEY
volumetric	I-KEY
radar	I-KEY
data	I-KEY
are	O
discussed	O
.	O
Such	O
data	O
come	O
from	O
the	O
radar	B-KEY
decision	I-KEY
support	I-KEY
system	I-KEY
-LRB-	O
RDSS	O
-RRB-	O
database	O
of	O
Environment	O
Canada	O
and	O
concern	O
summer	B-KEY
storms	I-KEY
created	O
in	O
this	O
country	O
.	O
Some	O
research	O
groups	O
used	O
the	O
data	O
completed	O
by	O
RDSS	O
for	O
verifying	O
the	O
utility	O
of	O
chosen	O
methods	O
in	O
volumetric	B-KEY
storm	I-KEY
cells	I-KEY
classification	I-KEY
.	O
The	O
paper	O
consists	O
of	O
a	O
review	O
of	O
experiments	O
that	O
were	O
made	O
on	O
the	O
data	O
from	O
RDSS	O
database	O
of	O
Environment	O
Canada	O
and	O
presents	O
the	O
quality	O
of	O
particular	O
classifiers	O
.	O
The	O
classification	B-KEY
accuracy	I-KEY
coefficient	O
is	O
used	O
to	O
express	O
the	O
quality	O
.	O
For	O
five	O
research	O
groups	O
that	O
led	O
their	O
experiments	O
in	O
a	O
similar	O
way	O
it	O
was	O
possible	O
to	O
compare	O
received	O
outputs	O
.	O
Experiments	O
showed	O
that	O
the	O
support	B-KEY
vector	I-KEY
machine	I-KEY
-LRB-	O
SVM	O
-RRB-	O
method	O
and	O
rough	B-KEY
set	I-KEY
algorithms	I-KEY
which	O
use	O
object	B-KEY
oriented	I-KEY
reducts	I-KEY
for	O
rule	O
generation	O
to	O
classify	O
volumetric	O
storm	O
data	O
perform	O
better	O
than	O
other	O
classifiers	O

Cross-entropy	O
and	O
rare	O
events	O
for	O
maximal	O
cut	O
and	O
partition	B-KEY
problems	I-KEY
We	O
show	O
how	O
to	O
solve	O
the	O
maximal	O
cut	O
and	O
partition	B-KEY
problems	I-KEY
using	O
a	O
randomized	B-KEY
algorithm	I-KEY
based	O
on	O
the	O
cross-entropy	O
method	O
.	O
For	O
the	O
maximal	B-KEY
cut	I-KEY
problem	I-KEY
,	O
the	O
proposed	O
algorithm	O
employs	O
an	O
auxiliary	O
Bernoulli	B-KEY
distribution	I-KEY
,	O
which	O
transforms	O
the	O
original	O
deterministic	B-KEY
network	I-KEY
into	O
an	O
associated	O
stochastic	O
one	O
,	O
called	O
the	O
associated	B-KEY
stochastic	I-KEY
network	I-KEY
-LRB-	O
ASN	O
-RRB-	O
.	O
Each	O
iteration	O
of	O
the	O
randomized	B-KEY
algorithm	I-KEY
for	O
the	O
ASN	O
involves	O
the	O
following	O
two	O
phases	O
:	O
-LRB-	O
1	O
-RRB-	O
generation	O
of	O
random	B-KEY
cuts	I-KEY
using	O
a	O
multidimensional	O
Ber	O
-LRB-	O
p	O
-RRB-	O
distribution	O
and	O
calculation	O
of	O
the	O
associated	O
cut	O
lengths	O
-LRB-	O
objective	O
functions	O
-RRB-	O
and	O
some	O
related	O
quantities	O
,	O
such	O
as	O
rare-event	O
probabilities	B-KEY
;	O
-LRB-	O
2	O
-RRB-	O
updating	O
the	O
parameter	O
vector	O
p	O
on	O
the	O
basis	O
of	O
the	O
data	O
collected	O
in	O
the	O
first	O
phase	O
.	O
We	O
show	O
that	O
the	O
Ber	O
-LRB-	O
p	O
-RRB-	O
distribution	O
converges	O
in	O
distribution	O
to	O
a	O
degenerated	O
one	O
,	O
Ber	O
-LRB-	O
p/sub	O
d	O
/	O
*	O
-RRB-	O
,	O
p/sub	O
d	O
/	O
*	O
=	O
-LRB-	O
p/sub	O
d	O
/	O
,	O
/	O
sub	O
1	O
/	O
,	O
p/sub	O
d	O
,	O
n	O
/	O
-RRB-	O
in	O
the	O
sense	O
that	O
some	O
elements	O
of	O
p/sub	O
d	O
/	O
*	O
,	O
will	O
be	O
unities	O
and	O
the	O
rest	O
zeros	O
.	O
The	O
unity	O
elements	O
of	O
p/sub	O
d	O
/	O
*	O
uniquely	O
define	O
a	O
cut	O
which	O
will	O
be	O
taken	O
as	O
the	O
estimate	O
of	O
the	O
maximal	O
cut	O
.	O
A	O
similar	O
approach	O
is	O
used	O
for	O
the	O
partition	B-KEY
problem	I-KEY
.	O
Supporting	O
numerical	B-KEY
results	I-KEY
are	O
given	O
as	O
well	O
.	O
Our	O
numerical	O
studies	O
suggest	O
that	O
for	O
the	O
maximal	O
cut	O
and	O
partition	B-KEY
problems	I-KEY
the	O
proposed	O
algorithm	O
typically	O
has	O
polynomial	B-KEY
complexity	I-KEY
in	O
the	O
size	O
of	O
the	O
network	O

Geometric	B-KEY
source	I-KEY
separation	I-KEY
:	O
merging	O
convolutive	O
source	O
separation	O
with	O
geometric	B-KEY
beamforming	I-KEY
Convolutive	B-KEY
blind	I-KEY
source	I-KEY
separation	I-KEY
and	O
adaptive	B-KEY
beamforming	I-KEY
have	O
a	O
similar	O
goal-extracting	O
a	O
source	O
of	O
interest	O
-LRB-	O
or	O
multiple	O
sources	O
-RRB-	O
while	O
reducing	O
undesired	O
interferences	O
.	O
A	O
benefit	O
of	O
source	O
separation	O
is	O
that	O
it	O
overcomes	O
the	O
conventional	O
cross-talk	B-KEY
or	O
leakage	B-KEY
problem	I-KEY
of	O
adaptive	B-KEY
beamforming	I-KEY
.	O
Beamforming	O
on	O
the	O
other	O
hand	O
exploits	O
geometric	O
information	O
which	O
is	O
often	O
readily	O
available	O
but	O
not	O
utilized	O
in	O
blind	B-KEY
algorithms	I-KEY
.	O
We	O
propose	O
to	O
join	O
these	O
benefits	O
by	O
combining	O
cross-power	B-KEY
minimization	I-KEY
of	O
second-order	B-KEY
source	I-KEY
separation	I-KEY
with	O
geometric	B-KEY
linear	I-KEY
constraints	I-KEY
used	O
in	O
adaptive	B-KEY
beamforming	I-KEY
.	O
We	O
find	O
that	O
the	O
geometric	O
constraints	O
resolve	O
some	O
of	O
the	O
ambiguities	O
inherent	O
in	O
the	O
independence	O
criterion	O
such	O
as	O
frequency	B-KEY
permutations	I-KEY
and	O
degrees	B-KEY
of	I-KEY
freedom	I-KEY
provided	O
by	O
additional	O
sensors	B-KEY
.	O
We	O
demonstrate	O
the	O
new	O
method	O
in	O
performance	O
comparisons	O
for	O
actual	O
room	B-KEY
recordings	I-KEY
of	O
two	O
and	O
three	O
simultaneous	O
acoustic	B-KEY
sources	I-KEY

Pool	O
halls	O
,	O
chips	O
,	O
and	O
war	O
games	O
:	O
women	B-KEY
in	O
the	O
culture	B-KEY
of	I-KEY
computing	I-KEY
Computers	O
are	O
becoming	O
ubiquitous	O
in	O
our	O
society	O
and	O
they	O
offer	O
superb	O
opportunities	O
for	O
people	O
in	O
jobs	O
and	O
everyday	O
life	O
.	O
But	O
there	O
is	O
a	O
noticeable	O
sex	B-KEY
difference	I-KEY
in	O
use	O
of	O
computers	O
among	O
children	B-KEY
.	O
This	O
article	O
asks	O
why	O
computers	O
are	O
more	O
attractive	O
to	O
boys	O
than	O
to	O
girls	O
and	O
offers	O
a	O
cultural	O
framework	O
for	O
explaining	O
the	O
apparent	O
sex	B-KEY
differences	I-KEY
.	O
Although	O
the	O
data	O
are	O
fragmentary	O
,	O
the	O
world	O
of	O
computing	O
seems	O
to	O
be	O
more	O
consistent	O
with	O
male	B-KEY
adolescent	I-KEY
culture	I-KEY
than	O
with	O
feminine	O
values	O
and	O
goals	O
.	O
Furthermore	O
,	O
both	O
arcade	O
and	O
educational	B-KEY
software	I-KEY
is	O
designed	O
with	O
boys	O
in	O
mind	O
.	O
These	O
observations	O
lead	O
us	O
to	O
speculate	O
that	O
computing	O
is	O
neither	O
inherently	O
difficult	O
nor	O
uninteresting	O
to	O
girls	O
,	O
but	O
rather	O
that	O
computer	B-KEY
games	I-KEY
and	O
other	O
software	O
might	O
have	O
to	O
be	O
designed	O
differently	O
for	O
girls	O
.	O
Programs	O
to	O
help	O
teachers	B-KEY
instill	O
computer	O
efficacy	O
in	O
all	O
children	B-KEY
also	O
need	O
to	O
be	O
developed	O

Hybrid	O
fuzzy	B-KEY
modeling	I-KEY
of	O
chemical	B-KEY
processes	I-KEY
Fuzzy	B-KEY
models	I-KEY
have	O
been	O
proved	O
to	O
have	O
the	O
ability	O
of	O
modeling	O
all	O
plants	O
without	O
any	O
priori	O
information	O
.	O
However	O
,	O
the	O
performance	O
of	O
conventional	O
fuzzy	B-KEY
models	I-KEY
can	O
be	O
very	O
poor	O
in	O
the	O
case	O
of	O
insufficient	O
training	O
data	O
due	O
to	O
their	O
poor	O
extrapolation	O
capacity	O
.	O
In	O
order	O
to	O
overcome	O
this	O
problem	O
,	O
a	O
hybrid	O
grey-box	O
fuzzy	B-KEY
modeling	I-KEY
approach	O
is	O
proposed	O
in	O
this	O
paper	O
to	O
combine	O
expert	O
experience	O
,	O
local	O
linear	O
models	O
and	O
historical	O
data	O
into	O
a	O
uniform	O
framework	O
.	O
It	O
consists	O
of	O
two	O
layers	O
.	O
The	O
expert	B-KEY
fuzzy	I-KEY
model	I-KEY
constructed	O
from	O
linguistic	O
information	O
,	O
the	O
local	O
linear	O
model	O
and	O
the	O
T-S	O
type	O
fuzzy	O
model	O
constructed	O
from	O
data	O
are	O
all	O
put	O
in	O
the	O
first	O
layer	O
.	O
Layer	O
2	O
is	O
a	O
fuzzy	B-KEY
decision	I-KEY
module	I-KEY
that	O
is	O
used	O
to	O
decide	O
which	O
model	O
in	O
the	O
first	O
layer	O
should	O
be	O
employed	O
to	O
make	O
the	O
final	O
prediction	O
.	O
The	O
output	O
of	O
the	O
second	O
layer	O
is	O
the	O
output	O
of	O
the	O
hybrid	O
fuzzy	B-KEY
model	I-KEY
.	O
With	O
the	O
help	O
of	O
the	O
linguistic	O
information	O
,	O
the	O
poor	O
extrapolation	O
capacity	O
problem	O
caused	O
by	O
sparse	O
training	O
data	O
for	O
conventional	O
fuzzy	B-KEY
models	I-KEY
can	O
be	O
overcome	O
.	O
Simulation	O
result	O
for	O
pH	O
neutralization	O
process	O
demonstrates	O
its	O
modeling	O
ability	O
over	O
the	O
linear	O
models	O
,	O
the	O
expert	B-KEY
fuzzy	I-KEY
model	I-KEY
and	O
the	O
conventional	O
fuzzy	O
model	O

New	O
investors	O
get	O
steal	O
of	O
a	O
deal	O
-LSB-	O
Global	B-KEY
Crossing	I-KEY
-RSB-	O
Hutchison	B-KEY
Telecommunications	I-KEY
and	O
Singapore	B-KEY
Technologies	I-KEY
take	O
control	O
of	O
Global	B-KEY
Crossing	I-KEY
for	O
a	O
lot	O
less	O
money	O
than	O
they	O
originally	O
offered	O
.	O
The	O
deal	O
leaves	O
the	O
bankrupt	B-KEY
carrier	O
intact	O
,	O
but	O
does	O
n't	O
put	O
it	O
in	O
the	O
clear	O
just	O
yet	O

Process	B-KEY
planning	I-KEY
for	O
reliable	B-KEY
high-speed	I-KEY
machining	I-KEY
of	O
moulds	B-KEY
A	O
method	O
of	O
generating	O
NC	B-KEY
programs	I-KEY
for	O
the	O
high-speed	B-KEY
milling	I-KEY
of	O
moulds	B-KEY
is	O
investigated	O
.	O
Forging	B-KEY
dies	I-KEY
and	O
injection	B-KEY
moulds	I-KEY
,	O
whether	O
plastic	O
or	O
aluminium	O
,	O
have	O
a	O
complex	O
surface	O
geometry	O
.	O
In	O
addition	O
they	O
are	O
made	O
of	O
steels	O
of	O
hardness	O
as	O
much	O
as	O
30	O
or	O
even	O
50	O
HRC	O
.	O
Since	O
1995	O
,	O
high-speed	O
machining	O
has	O
been	O
much	O
adopted	O
by	O
the	O
die-making	O
industry	O
,	O
which	O
with	O
this	O
technology	O
can	O
reduce	O
its	O
use	O
of	O
Sinking	O
Electrodischarge	O
Machining	O
-LRB-	O
SEDM	O
-RRB-	O
.	O
EDM	O
,	O
in	O
general	O
,	O
calls	O
for	O
longer	O
machining	O
times	O
.	O
The	O
use	O
of	O
high-speed	O
machining	O
makes	O
it	O
necessary	O
to	O
redefine	O
the	O
preliminary	O
stages	O
of	O
the	O
process	O
.	O
In	O
addition	O
,	O
it	O
affects	O
the	O
methodology	O
employed	O
in	O
the	O
generation	O
of	O
NC	B-KEY
programs	I-KEY
,	O
which	O
requires	O
the	O
use	O
of	O
high-level	O
CAM	O
software	O
.	O
The	O
aim	O
is	O
to	O
generate	O
error-free	B-KEY
programs	I-KEY
that	O
make	O
use	O
of	O
optimum	B-KEY
cutting	I-KEY
strategies	I-KEY
in	O
the	O
interest	O
of	O
productivity	O
and	O
surface	O
quality	O
.	O
The	O
final	O
result	O
is	O
a	O
more	O
reliable	O
manufacturing	O
process	O
.	O
There	O
are	O
two	O
risks	O
in	O
the	O
use	O
of	O
high-speed	B-KEY
milling	I-KEY
on	O
hardened	B-KEY
steels	I-KEY
.	O
One	O
of	O
these	O
is	O
tool	B-KEY
breakage	I-KEY
,	O
which	O
may	O
be	O
very	O
costly	O
and	O
may	O
furthermore	O
entail	O
marks	O
on	O
the	O
workpiece	O
.	O
The	O
other	O
is	O
collisions	O
between	O
the	O
tool	O
and	O
the	O
workpiece	O
or	O
fixtures	O
,	O
the	O
result	O
of	O
which	O
may	O
be	O
damage	O
to	O
the	O
ceramic	B-KEY
bearings	I-KEY
in	O
the	O
spindles	O
.	O
in	O
order	O
to	O
minimize	O
these	O
risks	O
it	O
is	O
necessary	O
that	O
new	O
control	O
and	O
optimization	O
steps	O
be	O
included	O
in	O
the	O
CAM	B-KEY
methodology	I-KEY
.	O
There	O
are	O
three	O
things	O
that	O
the	O
firm	O
adopting	O
high-speed	O
methods	O
should	O
do	O
.	O
It	O
should	O
redefine	O
its	O
process	O
engineering	O
,	O
it	O
should	O
systematize	O
access	O
by	O
its	O
CAM	O
programmers	O
to	O
high-speed	O
knowhow	O
,	O
and	O
it	O
should	O
take	O
up	O
the	O
use	O
of	O
process	B-KEY
simulation	I-KEY
tools	I-KEY
.	O
In	O
the	O
latter	O
case	O
,	O
it	O
will	O
be	O
very	O
advantageous	O
to	O
use	O
tools	O
for	O
the	O
estimation	O
of	O
cutting	O
forces	O
.	O
The	O
new	O
work	O
methods	O
proposed	O
in	O
this	O
article	O
have	O
made	O
it	O
possible	O
to	O
introduce	O
high	O
speed	O
milling	O
-LRB-	O
HSM	O
-RRB-	O
into	O
the	O
die	O
industry	O
.	O
Examples	O
are	O
given	O
of	O
how	O
the	O
technique	O
has	O
been	O
applied	O
with	O
CAM	B-KEY
programming	I-KEY
re-engineered	I-KEY
as	O
here	O
proposed	O
,	O
with	O
an	O
explanation	O
of	O
the	O
novel	O
features	O
and	O
the	O
results	O

Vector	B-KEY
algebra	I-KEY
proofs	I-KEY
for	O
geometry	O
theorems	O
Vector	B-KEY
mathematics	I-KEY
can	O
generate	O
simple	O
and	O
powerful	O
proofs	B-KEY
of	O
theorems	O
in	O
plane	B-KEY
geometry	I-KEY
.	O
These	O
proofs	B-KEY
can	O
also	O
be	O
used	O
to	O
generalize	O
plane	B-KEY
geometry	I-KEY
theorems	O
to	O
higher	O
dimensions	O
.	O
We	O
present	O
three	O
vector	O
proofs	B-KEY
that	O
show	O
the	O
power	O
of	O
this	O
technique	O
.	O
1	O
.	O
For	O
any	O
quadrilateral	B-KEY
,	O
the	O
sum	O
of	O
the	O
squares	O
of	O
the	O
diagonals	O
is	O
less	O
than	O
or	O
equal	O
to	O
the	O
sum	O
of	O
the	O
squares	O
of	O
the	O
sides	O
.	O
2	O
.	O
The	O
area	O
of	O
a	O
quadrilateral	B-KEY
is	O
half	O
the	O
product	O
of	O
the	O
diagonals	O
multiplied	O
by	O
the	O
sine	O
of	O
an	O
included	O
angle	O
.	O
3	O
.	O
One	O
quarter	O
of	O
all	O
triangles	O
are	O
acute	O
-LRB-	O
Based	O
upon	O
the	O
options	O
detailed	O
below	O
,	O
with	O
respect	O
to	O
the	O
relative	O
lengths	O
of	O
the	O
sides	O
-RRB-	O
.	O
This	O
paper	O
presents	O
a	O
set	O
of	O
examples	O
of	O
vector	B-KEY
mathematics	I-KEY
applied	O
to	O
geometry	O
problems	O
.	O
Some	O
of	O
the	O
most	O
beautiful	O
and	O
sophisticated	O
proofs	B-KEY
in	O
mathematics	O
involve	O
using	O
multiple	B-KEY
representations	I-KEY
of	O
the	O
same	O
data	O
.	O
By	O
leveraging	O
the	O
advantages	O
of	O
each	O
representation	O
one	O
finds	O
new	O
and	O
useful	O
mathematical	O
facts	O

Will	O
new	O
Palms	O
win	O
laurels	O
.	O
?	O
PalmSource	B-KEY
's	O
latest	O
operating	B-KEY
system	I-KEY
for	O
mobile	B-KEY
devices	I-KEY
harnesses	O
the	O
ARM	B-KEY
architecture	I-KEY
to	O
support	O
more	O
powerful	O
business	O
software	O
,	O
but	O
there	O
are	O
concerns	O
over	O
compatibility	B-KEY
with	O
older	O
applications	O

Reconstruction	O
of	O
MR	O
images	O
from	O
data	O
acquired	O
on	O
an	O
arbitrary	B-KEY
k-space	I-KEY
trajectory	I-KEY
using	O
the	O
same-image	B-KEY
weight	I-KEY
A	O
sampling	B-KEY
density	I-KEY
compensation	I-KEY
function	O
denoted	O
``	O
same-image	O
-LRB-	O
SI	O
-RRB-	O
weight	O
''	O
is	O
proposed	O
to	O
reconstruct	O
MR	O
images	O
from	O
the	O
data	O
acquired	O
on	O
an	O
arbitrary	B-KEY
k-space	I-KEY
trajectory	I-KEY
.	O
An	O
equation	O
for	O
the	O
SI	O
weight	O
is	O
established	O
on	O
the	O
SI	O
criterion	O
and	O
an	O
iterative	O
scheme	O
is	O
developed	O
to	O
find	O
the	O
weight	O
.	O
The	O
SI	O
weight	O
is	O
then	O
used	O
to	O
reconstruct	O
images	O
from	O
the	O
data	O
calculated	O
on	O
a	O
random	B-KEY
trajectory	I-KEY
in	O
a	O
numerical	B-KEY
phantom	I-KEY
case	O
and	O
from	O
the	O
data	O
acquired	O
on	O
interleaved	O
spirals	O
in	O
an	O
in	O
vivo	O
experiment	O
,	O
respectively	O
.	O
In	O
addition	O
,	O
Pipe	O
and	O
Menon	O
's	O
weight	O
-LRB-	O
MRM	O
1999	O
;	O
41:179	O
-186	O
-RRB-	O
is	O
also	O
used	O
in	O
the	O
reconstructions	O
to	O
make	O
a	O
comparison	O
.	O
The	O
images	O
obtained	O
with	O
the	O
SI	O
weight	O
were	O
found	O
to	O
be	O
slightly	O
more	O
accurate	O
than	O
those	O
obtained	O
with	O
Pipe	O
's	O
weight	O

Run-time	O
data-flow	O
analysis	O
Parallelizing	B-KEY
compilers	I-KEY
have	O
made	O
great	O
progress	O
in	O
recent	O
years	O
.	O
However	O
,	O
there	O
still	O
remains	O
a	O
gap	O
between	O
the	O
current	O
ability	O
of	O
parallelizing	B-KEY
compilers	I-KEY
and	O
their	O
final	O
goals	O
.	O
In	O
order	O
to	O
achieve	O
the	O
maximum	O
parallelism	O
,	O
run-time	O
techniques	O
were	O
used	O
in	O
parallelizing	B-KEY
compilers	I-KEY
during	O
last	O
few	O
years	O
.	O
First	O
,	O
this	O
paper	O
presents	O
a	O
basic	O
run-time	B-KEY
privatization	I-KEY
method	I-KEY
.	O
The	O
definition	O
of	O
run-time	B-KEY
dead	I-KEY
code	I-KEY
is	O
given	O
and	O
its	O
side	O
effect	O
is	O
discussed	O
.	O
To	O
eliminate	O
the	O
imprecision	O
caused	O
by	O
the	O
run-time	B-KEY
dead	I-KEY
code	I-KEY
,	O
backward	B-KEY
data-flow	I-KEY
information	I-KEY
must	O
be	O
used	O
.	O
Proteus	B-KEY
Test	I-KEY
,	O
which	O
can	O
use	O
backward	O
information	O
in	O
run-time	O
,	O
is	O
then	O
presented	O
to	O
exploit	O
more	O
dynamic	B-KEY
parallelism	I-KEY
.	O
Also	O
,	O
a	O
variation	O
of	O
Proteus	B-KEY
Test	I-KEY
,	O
the	O
Advanced	O
Proteus	B-KEY
Test	I-KEY
,	O
is	O
offered	O
to	O
achieve	O
partial	O
parallelism	O
.	O
Proteus	B-KEY
Test	I-KEY
was	O
implemented	O
on	O
the	O
parallelizing	B-KEY
compiler	I-KEY
AFT	O
.	O
In	O
the	O
end	O
of	O
this	O
paper	O
the	O
program	O
fpppp.f	O
of	O
Spec95fp	O
Benchmark	O
is	O
taken	O
as	O
an	O
example	O
,	O
to	O
show	O
the	O
effectiveness	O
of	O
Proteus	B-KEY
Test	I-KEY

A	O
portable	O
Auto	B-KEY
Attendant	I-KEY
System	I-KEY
with	O
sophisticated	O
dialog	O
structure	O
An	O
attendant	B-KEY
system	I-KEY
connects	O
the	O
caller	O
to	O
the	O
party	O
he/she	O
wants	O
to	O
talk	O
to	O
.	O
Traditional	O
systems	O
require	O
the	O
caller	O
to	O
know	O
the	O
full	O
name	O
of	O
the	O
party	O
.	O
If	O
the	O
caller	O
forgets	O
the	O
name	O
,	O
the	O
system	O
fails	O
to	O
provide	O
service	O
for	O
the	O
caller	O
.	O
In	O
this	O
paper	O
we	O
propose	O
a	O
portable	O
Auto	B-KEY
Attendant	I-KEY
System	I-KEY
-LRB-	O
AAS	O
-RRB-	O
with	O
sophisticated	O
dialog	O
structure	O
that	O
gives	O
a	O
caller	O
more	O
flexibility	O
while	O
calling	O
.	O
The	O
caller	O
may	O
interact	O
with	O
the	O
system	O
to	O
request	O
a	O
phone	O
number	O
by	O
providing	O
just	O
a	O
work	O
area	O
,	O
specialty	O
,	O
surname	O
,	O
or	O
title	O
,	O
etc.	O
.	O
If	O
the	O
party	O
is	O
absent	O
,	O
the	O
system	O
may	O
provide	O
extra	O
information	O
such	O
as	O
where	O
he	O
went	O
,	O
when	O
he	O
will	O
be	O
back	O
,	O
and	O
what	O
he	O
is	O
doing	O
.	O
The	O
system	O
is	O
built	O
modularly	O
,	O
with	O
components	O
such	O
as	O
speech	B-KEY
recognizer	I-KEY
,	O
language	O
model	O
,	O
dialog	B-KEY
manager	I-KEY
and	O
text-to-speech	O
that	O
can	O
be	O
replaced	O
if	O
necessary	O
.	O
By	O
simply	O
changing	O
the	O
personnel	O
record	O
database	O
,	O
the	O
system	O
can	O
easily	O
be	O
ported	O
to	O
other	O
companies	O
.	O
The	O
sophisticated	O
dialog	B-KEY
manager	I-KEY
applies	O
many	O
strategies	O
to	O
allow	O
natural	O
interaction	O
between	O
user	O
and	O
system	O
.	O
Functions	O
such	O
as	O
fuzzy	B-KEY
request	I-KEY
,	O
user	O
repairing	O
,	O
and	O
extra	O
information	O
query	O
,	O
which	O
are	O
not	O
provided	O
by	O
other	O
systems	O
,	O
are	O
integrated	O
into	O
our	O
system	O
.	O
Experimental	O
results	O
and	O
comparisons	O
to	O
other	O
systems	O
show	O
that	O
our	O
approach	O
provides	O
a	O
more	O
user	O
friendly	O
and	O
natural	O
interaction	O
for	O
auto	B-KEY
attendant	I-KEY
system	I-KEY

Bistability	B-KEY
of	O
harmonically	B-KEY
forced	I-KEY
relaxation	I-KEY
oscillations	I-KEY
Relaxation	O
oscillations	O
appear	O
in	O
processes	O
which	O
involve	O
transitions	O
between	O
two	O
states	O
characterized	O
by	O
fast	O
and	O
slow	O
time	O
scales	O
.	O
When	O
a	O
relaxation	O
oscillator	O
is	O
coupled	O
to	O
an	O
external	B-KEY
periodic	I-KEY
force	I-KEY
its	O
entrainment	B-KEY
by	O
the	O
force	O
results	O
in	O
a	O
response	O
which	O
can	O
include	O
multiple	O
periodicities	O
and	O
bistability	B-KEY
.	O
The	O
prototype	O
of	O
these	O
behaviors	O
is	O
the	O
harmonically	O
driven	O
van	B-KEY
der	I-KEY
Pol	I-KEY
equation	I-KEY
which	O
displays	O
regions	O
in	O
the	O
parameter	O
space	O
of	O
the	O
driving	O
force	O
amplitude	O
where	O
stable	O
orbits	O
of	O
periods	O
2n	O
+	O
or-1	O
coexist	O
,	O
flanked	O
by	O
regions	O
of	O
periods	O
2n	O
+1	O
and	O
2n-1	O
.	O
The	O
parameter	O
regions	O
of	O
such	O
bistable	B-KEY
orbits	O
are	O
derived	O
analytically	O
for	O
the	O
closely	O
related	O
harmonically	B-KEY
driven	I-KEY
Stoker-Haag	I-KEY
piecewise	I-KEY
discontinuous	I-KEY
equation	I-KEY
.	O
The	O
results	O
are	O
valid	O
over	O
most	O
of	O
the	O
control	B-KEY
parameter	I-KEY
space	I-KEY
of	O
the	O
system	O
.	O
Also	O
considered	O
are	O
the	O
reasons	O
for	O
the	O
more	O
complicated	O
dynamics	O
featuring	O
regions	O
of	O
high	O
multiple	O
periodicity	O
which	O
appear	O
like	O
noise	O
between	O
ordered	O
periodic	O
regions	O
.	O
Since	O
this	O
system	O
mimics	O
in	O
detail	O
the	O
less	O
analytically	O
tractable	O
forced	O
van	B-KEY
der	I-KEY
Pol	I-KEY
equation	I-KEY
,	O
the	O
results	O
suggest	O
extensions	O
to	O
situations	O
where	O
forced	O
relaxation	O
oscillations	O
are	O
a	O
component	O
of	O
the	O
operating	O
mechanisms	O

Approximation	O
theory	O
of	O
fuzzy	B-KEY
systems	I-KEY
based	O
upon	O
genuine	O
many-valued	B-KEY
implications	I-KEY
-	O
SISO	B-KEY
cases	O
It	O
is	O
proved	O
that	O
the	O
single	O
input	O
and	O
single	O
output	O
-LRB-	O
SISO	B-KEY
-RRB-	O
fuzzy	B-KEY
systems	I-KEY
based	O
upon	O
genuine	O
many-valued	B-KEY
implications	I-KEY
are	O
universal	B-KEY
approximators	I-KEY
.	O
It	O
is	O
shown	O
theoretically	O
that	O
fuzzy	O
control	O
systems	O
based	O
upon	O
genuine	O
many-valued	B-KEY
implications	I-KEY
are	O
equivalent	O
to	O
those	O
based	O
upon	O
t-norm	O
implications	O
,	O
the	O
general	O
approach	O
to	O
construct	O
fuzzy	B-KEY
systems	I-KEY
is	O
given	O
.	O
It	O
is	O
also	O
shown	O
that	O
defuzzifier	O
based	O
upon	O
center	O
of	O
areas	O
is	O
not	O
appropriate	O
to	O
the	O
fuzzy	B-KEY
systems	I-KEY
based	O
upon	O
genuine	O
many-valued	B-KEY
implications	I-KEY

On	O
bivariate	B-KEY
dependence	I-KEY
and	O
the	O
convex	B-KEY
order	I-KEY
We	O
investigate	O
the	O
interplay	O
between	O
variability	O
-LRB-	O
in	O
the	O
sense	O
of	O
the	O
convex	B-KEY
order	I-KEY
-RRB-	O
and	O
dependence	O
in	O
a	O
bivariate	O
framework	O
,	O
extending	O
some	O
previous	O
results	O
in	O
this	O
area	O
.	O
We	O
exploit	O
the	O
fact	O
that	O
discrete	B-KEY
uniform	I-KEY
distributions	I-KEY
are	O
dense	O
in	O
the	O
space	O
of	O
probability	B-KEY
measures	I-KEY
in	O
the	O
topology	B-KEY
of	O
weak	B-KEY
convergence	I-KEY
to	O
prove	O
our	O
central	O
result	O
.	O
We	O
also	O
obtain	O
a	O
partial	O
result	O
in	O
the	O
general	O
multivariate	O
case	O
.	O
Our	O
findings	O
can	O
be	O
interpreted	O
in	O
terms	O
of	O
the	O
impact	O
of	O
component	B-KEY
variability	I-KEY
on	O
the	O
mean	B-KEY
life	I-KEY
of	O
correlated	O
serial	O
and	O
parallel	B-KEY
systems	I-KEY

Using	O
technology	O
to	O
facilitate	O
the	O
design	O
and	O
delivery	O
of	O
warnings	O
This	O
paper	O
describes	O
several	O
ways	O
in	O
which	O
new	O
technologies	O
can	O
assist	O
in	O
the	O
design	O
and	O
delivery	O
of	O
warnings	O
.	O
There	O
are	O
four	O
discussion	O
points	O
:	O
-LRB-	O
1	O
-RRB-	O
current	O
product	B-KEY
information	I-KEY
can	O
be	O
delivered	O
via	O
the	O
Internet	B-KEY
;	O
-LRB-	O
2	O
-RRB-	O
computer	B-KEY
software	I-KEY
and	O
hardware	O
are	O
available	O
to	O
assist	O
in	O
the	O
design	O
,	O
construction	O
,	O
and	O
production	O
of	O
visual	O
and	O
auditory	B-KEY
warnings	I-KEY
;	O
-LRB-	O
3	O
-RRB-	O
various	O
detection	O
devices	O
can	O
be	O
used	O
to	O
recognize	O
instances	O
in	O
which	O
warnings	O
might	O
be	O
delivered	O
;	O
and	O
-LRB-	O
4	O
-RRB-	O
a	O
warning	B-KEY
presentation	I-KEY
can	O
be	O
modified	O
to	O
fit	O
conditions	O
and	O
persons	O
.	O
Implications	O
,	O
example	O
applications	O
and	O
future	O
prospects	O
of	O
these	O
points	O
are	O
described	O

A	O
question	O
of	O
perspective	O
:	O
assigning	O
Library	O
of	O
Congress	O
subject	O
headings	O
to	O
classical	B-KEY
literature	I-KEY
and	O
ancient	B-KEY
history	I-KEY
This	O
article	O
explains	O
the	O
concept	O
of	O
world	B-KEY
view	I-KEY
and	O
shows	O
how	O
the	O
world	B-KEY
view	I-KEY
of	O
cataloguers	O
influences	O
the	O
development	O
and	O
assignment	O
of	O
subject	O
headings	O
to	O
works	O
about	O
other	O
cultures	B-KEY
and	O
civilizations	B-KEY
,	O
using	O
works	O
from	O
classical	B-KEY
literature	I-KEY
and	O
ancient	B-KEY
history	I-KEY
as	O
examples	O
.	O
Cataloguers	O
are	O
encouraged	O
to	O
evaluate	O
the	O
headings	O
they	O
assign	O
to	O
works	O
in	O
classical	B-KEY
literature	I-KEY
and	O
ancient	B-KEY
history	I-KEY
in	O
terms	O
of	O
the	O
world	B-KEY
views	I-KEY
of	O
Ancient	B-KEY
Greece	I-KEY
and	O
Rome	O
so	O
that	O
headings	O
reflect	O
the	O
contents	O
of	O
the	O
works	O
they	O
describe	O
and	O
give	O
fuller	O
expression	O
to	O
the	O
diversity	O
of	O
thoughts	O
and	O
themes	O
that	O
characterize	O
these	O
ancient	O
civilizations	B-KEY

Efficient	O
parallel	B-KEY
programming	I-KEY
on	O
scalable	B-KEY
shared	I-KEY
memory	I-KEY
systems	O
with	O
High	B-KEY
Performance	I-KEY
Fortran	I-KEY
OpenMP	O
offers	O
a	O
high-level	O
interface	O
for	O
parallel	B-KEY
programming	I-KEY
on	O
scalable	B-KEY
shared	I-KEY
memory	I-KEY
-LRB-	O
SMP	O
-RRB-	O
architectures	O
.	O
It	O
provides	O
the	O
user	O
with	O
simple	O
work-sharing	O
directives	O
while	O
it	O
relies	O
on	O
the	O
compiler	O
to	O
generate	O
parallel	B-KEY
programs	I-KEY
based	O
on	O
thread	O
parallelism	O
.	O
However	O
,	O
the	O
lack	O
of	O
language	O
features	O
for	O
exploiting	O
data	O
locality	O
often	O
results	O
in	O
poor	O
performance	O
since	O
the	O
non-uniform	O
memory	O
access	O
times	O
on	O
scalable	O
SMP	O
machines	O
can	O
not	O
be	O
neglected	O
.	O
High	B-KEY
Performance	I-KEY
Fortran	I-KEY
-LRB-	O
HPF	O
-RRB-	O
,	O
the	O
de-facto	O
standard	O
for	O
data	O
parallel	B-KEY
programming	I-KEY
,	O
offers	O
a	O
rich	O
set	O
of	O
data	O
distribution	O
directives	O
in	O
order	O
to	O
exploit	O
data	O
locality	O
,	O
but	O
it	O
has	O
been	O
mainly	O
targeted	O
towards	O
distributed	O
memory	O
machines	O
.	O
In	O
this	O
paper	O
we	O
describe	O
an	O
optimized	O
execution	O
model	O
for	O
HPF	O
programs	O
on	O
SMP	O
machines	O
that	O
avails	O
itself	O
with	O
mechanisms	O
provided	O
by	O
OpenMP	O
for	O
work	O
sharing	O
and	O
thread	O
parallelism	O
,	O
while	O
exploiting	O
data	O
locality	O
based	O
on	O
user-specified	O
distribution	O
directives	O
.	O
Data	O
locality	O
does	O
not	O
only	O
ensure	O
that	O
most	O
memory	O
accesses	O
are	O
close	O
to	O
the	O
executing	O
threads	O
and	O
are	O
therefore	O
faster	O
,	O
but	O
it	O
also	O
minimizes	O
synchronization	O
overheads	O
,	O
especially	O
in	O
the	O
case	O
of	O
unstructured	O
reductions	O
.	O
The	O
proposed	O
shared	O
memory	O
execution	O
model	O
for	O
HPF	O
relies	O
on	O
a	O
small	O
set	O
of	O
language	O
extensions	O
,	O
which	O
resemble	O
the	O
OpenMP	O
work-sharing	O
features	O
.	O
These	O
extensions	O
,	O
together	O
with	O
an	O
optimized	O
shared	O
memory	O
parallelization	O
and	O
execution	O
model	O
,	O
have	O
been	O
implemented	O
in	O
the	O
ADAPTOR	O
HPF	O
compilation	O
system	O
and	O
experimental	O
results	O
verify	O
the	O
efficiency	O
of	O
the	O
chosen	O
approach	O

Well	O
behaved	O
women	B-KEY
rarely	O
make	O
history	B-KEY
!	O
The	O
author	O
considers	O
women	B-KEY
in	O
the	O
history	B-KEY
of	O
computer	B-KEY
science	I-KEY
.	O
Prior	O
to	O
the	O
ENIAC	B-KEY
,	O
women	B-KEY
were	O
extremely	O
important	O
to	O
the	O
computing	O
business	B-KEY
as	O
``	O
computers	O
''	O
.	O
Just	O
as	O
women	B-KEY
had	O
taken	O
over	O
the	O
tasks	O
as	O
secretaries	O
in	O
the	O
late	O
1800s	O
with	O
the	O
advent	O
of	O
the	O
typewriter	O
,	O
and	O
in	O
the	O
early	O
1900s	O
staffing	O
telephone	O
exchanges	O
,	O
so	O
computing	O
relied	O
on	O
women	B-KEY
as	O
the	O
``	O
workhorses	O
''	O
of	O
the	O
business	B-KEY

Calculation	O
of	O
the	O
probability	O
of	O
survival	O
of	O
an	O
insurance	B-KEY
company	I-KEY
with	O
allowance	O
for	O
the	O
rate	O
of	O
return	O
for	O
a	O
Poisson	O
stream	O
of	O
premiums	O
The	O
probability	O
of	O
survival	O
of	O
an	O
insurance	B-KEY
company	I-KEY
with	O
the	O
working	O
capital	O
is	O
calculated	O
for	O
a	O
Poisson	O
stream	O
of	O
premiums	O

Editorial	O
system	O
vendors	O
focus	O
on	O
Adobe	O
and	O
the	O
future	O
Looking	O
over	O
the	O
newspaper-system	B-KEY
market	I-KEY
,	O
we	O
note	O
that	O
the	O
Mac	O
is	O
getting	O
new	O
respect	O
.	O
Adobe	B-KEY
InDesign	I-KEY
has	O
established	O
itself	O
as	O
a	O
solid	O
alternative	O
to	O
Quark	O
XPress	O
for	O
pagination	B-KEY
.	O
Positioning	O
themselves	O
for	O
the	O
long	O
run	O
,	O
developers	O
are	O
gradually	O
shifting	O
to	O
new	O
software	O
architectures	O

Combining	O
PC	O
control	O
and	O
HMI	O
Integrating	O
PC-based	O
control	O
with	O
human	B-KEY
machine	I-KEY
interface	I-KEY
-LRB-	O
HMI	O
-RRB-	O
technology	O
can	O
benefit	O
a	O
plant	O
floor	O
system	O
.	O
However	O
,	O
before	O
one	O
decides	O
on	O
PC-based	O
control	O
,	O
there	O
are	O
many	O
things	O
one	O
should	O
consider	O
,	O
especially	O
when	O
using	O
a	O
soft	O
programmable	B-KEY
logic	I-KEY
controller	I-KEY
-LRB-	O
PLC	O
-RRB-	O
to	O
command	O
the	O
input/output	O
.	O
There	O
are	O
three	O
strategies	O
to	O
integrate	O
a	O
PC-based	B-KEY
control	I-KEY
system	I-KEY
with	O
an	O
HMI	O
:	O
treat	O
the	O
PC	O
running	O
the	O
control	O
application	O
as	O
if	O
it	O
were	O
a	O
PLC	O
,	O
integrate	O
the	O
system	O
using	O
standard	O
PC	B-KEY
interfaces	I-KEY
;	O
or	O
using	O
application	B-KEY
programming	I-KEY
interfaces	I-KEY

Real-time	O
transmission	O
of	O
pediatric	O
echocardiograms	O
using	O
a	O
single	B-KEY
ISDN	I-KEY
line	I-KEY
We	O
tested	O
the	O
adequacy	O
of	O
a	O
videoconferencing	B-KEY
system	I-KEY
using	O
a	O
single	O
integrated	O
systems	O
digital	O
network	O
-LRB-	O
ISDN	O
-RRB-	O
line	O
-LRB-	O
128	O
kilobits	O
per	O
second	O
-RRB-	O
for	O
the	O
remote	B-KEY
diagnosis	I-KEY
of	O
children	B-KEY
with	O
suspected	B-KEY
congenital	I-KEY
heart	I-KEY
disease	I-KEY
-LRB-	O
CHD	O
-RRB-	O
.	O
Real-time	B-KEY
echocardiogram	I-KEY
interpretation	I-KEY
was	O
compared	O
to	O
subsequent	O
videotape	B-KEY
review	I-KEY
in	O
401	O
studies	O
with	O
concordance	O
in	O
383	O
-LRB-	O
95.5	O
%	O
-RRB-	O
studies	O
.	O
A	O
new	O
diagnosis	O
of	O
CHD	O
was	O
made	O
in	O
98	O
studies	O
.	O
Immediate	B-KEY
patient	I-KEY
transfer	I-KEY
was	O
arranged	O
based	O
upon	O
a	O
real-time	O
diagnosis	O
in	O
five	O
studies	O
.	O
In	O
300	O
studies	O
,	O
a	O
normal	O
diagnosis	O
obviated	O
further	O
evaluation	O
.	O
A	O
single	B-KEY
ISDN	I-KEY
line	I-KEY
is	O
adequate	O
for	O
transmission	O
of	O
pediatric	O
echocardiograms	O
and	O
it	O
allows	O
for	O
remote	O
management	O
of	O
patients	O
with	O
CHD	O

An	O
intelligent	B-KEY
information	I-KEY
gathering	I-KEY
method	I-KEY
for	O
dynamic	B-KEY
information	I-KEY
mediators	I-KEY
The	O
Internet	B-KEY
is	O
spreading	O
into	O
our	O
society	O
rapidly	O
and	O
is	O
becoming	O
one	O
of	O
the	O
information	B-KEY
infrastructures	I-KEY
that	O
are	O
indispensable	O
for	O
our	O
daily	O
life	O
.	O
In	O
particular	O
,	O
the	O
WWW	B-KEY
is	O
widely	O
used	O
for	O
various	O
purposes	O
such	O
as	O
sharing	O
personal	O
information	O
,	O
academic	B-KEY
research	I-KEY
,	O
business	B-KEY
work	I-KEY
,	O
and	O
electronic	B-KEY
commerce	I-KEY
,	O
and	O
the	O
amount	O
of	O
available	O
information	O
is	O
increasing	O
rapidly	O
.	O
We	O
usually	O
utilize	O
information	O
sources	O
on	O
the	O
Internet	B-KEY
as	O
individual	O
stand-alone	O
sources	O
,	O
but	O
if	O
we	O
can	O
integrate	O
them	O
,	O
we	O
can	O
add	O
more	O
value	O
to	O
each	O
of	O
them	O
.	O
Hence	O
,	O
information	O
mediators	O
,	O
which	O
integrate	O
information	O
distributed	O
on	O
the	O
Internet	B-KEY
,	O
are	O
drawing	O
attention	O
.	O
In	O
this	O
paper	O
,	O
under	O
the	O
assumption	O
that	O
the	O
information	O
sources	O
to	O
be	O
integrated	O
are	O
updated	O
frequently	O
and	O
asynchronously	O
,	O
we	O
propose	O
an	O
information	O
gathering	O
method	O
that	O
constructs	O
an	O
answer	O
to	O
a	O
query	O
from	O
a	O
user	O
,	O
accessing	O
information	O
sources	O
to	O
be	O
integrated	O
properly	O
within	O
an	O
allowable	O
time	O
period	O
.	O
The	O
proposed	O
method	O
considers	O
the	O
reliability	O
of	O
data	O
in	O
the	O
cache	O
and	O
the	O
quality	O
of	O
answer	O
in	O
order	O
to	O
efficiently	O
access	O
information	O
sources	O
and	O
to	O
provide	O
appropriate	O
answers	O
to	O
the	O
user	O
.	O
As	O
evaluation	O
,	O
we	O
show	O
the	O
effectiveness	O
of	O
the	O
proposed	O
method	O
by	O
using	O
an	O
artificial	B-KEY
information	I-KEY
integration	I-KEY
problem	I-KEY
,	O
in	O
which	O
some	O
parameters	O
can	O
be	O
modified	O
,	O
and	O
a	O
real-world	B-KEY
flight	I-KEY
information	I-KEY
service	I-KEY
compared	O
with	O
a	O
conventional	O
FIFO	O
information	O
gathering	O
method	O

PGE	O
helps	O
customers	O
reduce	O
energy	O
costs	O
A	O
new	O
service	O
from	O
Portland	B-KEY
General	I-KEY
Electric	I-KEY
-LRB-	O
PGE	O
,	O
Portland	O
,	O
Oregon	B-KEY
,	O
US	O
-RRB-	O
is	O
saving	O
customers	O
tens	O
of	O
thousands	O
of	O
dollars	O
in	O
energy	O
costs	O
.	O
PGE	O
created	O
E-Manager	B-KEY
to	O
allow	O
facility	O
managers	O
to	O
analyze	O
their	O
energy	O
consumption	O
online	O
at	O
15-minute	O
intervals	O
.	O
Customers	O
can	O
go	O
to	O
the	O
Web	O
for	O
complete	O
data	O
,	O
powerful	O
analysis	O
tools	O
and	O
charts	O
,	O
helping	O
them	O
detect	O
abnormal	O
energy	O
use	O
and	O
focus	O
on	O
costly	O
problem	O
areas	O

Extracting	O
linguistic	O
DNA	O
:	O
NStein	O
goes	O
to	O
work	O
for	O
UPI	B-KEY
It	O
's	O
a	O
tantalizing	O
problem	O
for	O
categorization	O
.	O
United	B-KEY
Press	I-KEY
International	I-KEY
-LRB-	O
UPI	B-KEY
-RRB-	O
has	O
more	O
than	O
700	O
correspondents	O
creating	O
thousands	O
of	O
stories	O
every	O
week	O
,	O
running	O
the	O
gamut	O
from	O
business	O
news	O
to	O
sports	O
to	O
entertainment	O
to	O
global	O
coverage	O
of	O
America	O
's	O
war	O
on	O
terrorism	O
.	O
And	O
while	O
UPI	B-KEY
and	O
others	O
news	O
services	O
have	O
mechanisms	O
for	O
adding	O
keywords	O
and	O
categorizing	O
their	O
content	O
,	O
UPI	B-KEY
recognized	O
a	O
need	O
to	O
add	O
more	O
automation	O
to	O
the	O
process	O
.	O
With	O
the	O
recent	O
growth	O
and	O
improvement	O
in	O
tools	O
for	O
Computer-Aided	B-KEY
Indexing	I-KEY
-LRB-	O
CAI	O
-RRB-	O
,	O
UPI	B-KEY
undertook	O
a	O
process	O
of	O
looking	O
at	O
its	O
needs	O
and	O
evaluating	O
the	O
many	O
CAI	O
tools	O
out	O
there	O
.	O
In	O
the	O
end	O
,	O
they	O
chose	O
technology	O
from	O
Montreal-based	O
NStein	B-KEY
Technologies	I-KEY
.	O
``	O
Our	O
main	O
objective	O
was	O
to	O
acquire	O
the	O
best	O
CAI	O
tool	O
to	O
help	O
improve	O
our	O
customers	O
'	O
access	O
and	O
interaction	O
with	O
our	O
content	O
,	O
''	O
says	O
Steve	O
Sweet	O
,	O
CIO	O
at	O
UPI	B-KEY
.	O
``	O
We	O
examined	O
a	O
number	O
of	O
solutions	O
,	O
and	O
NStein	O
's	O
NServer	O
suite	O
clearly	O
came	O
out	O
on	O
top	O
.	O
The	O
combination	O
of	O
speed	O
,	O
scalability	O
,	O
accuracy	O
,	O
and	O
flexibility	O
was	O
what	O
really	O
sold	O
us	O
.	O
''	O

Dense	B-KEY
coding	I-KEY
in	O
entangled	B-KEY
states	I-KEY
We	O
consider	O
the	O
dense	B-KEY
coding	I-KEY
of	O
entangled	O
qubits	O
shared	O
between	O
two	O
parties	O
,	O
Alice	B-KEY
and	O
Bob	B-KEY
.	O
The	O
efficiency	O
of	O
classical	O
information	O
gain	O
through	O
quantum	O
entangled	O
qubits	O
is	O
also	O
considered	O
for	O
the	O
case	O
of	O
pairwise	B-KEY
entangled	I-KEY
qubits	I-KEY
and	O
maximally	B-KEY
entangled	I-KEY
qubits	I-KEY
.	O
We	O
conclude	O
that	O
using	O
the	O
pairwise	B-KEY
entangled	I-KEY
qubits	I-KEY
can	O
be	O
more	O
efficient	O
when	O
two	O
parties	O
communicate	O
whereas	O
using	O
the	O
maximally	B-KEY
entangled	I-KEY
qubits	I-KEY
can	O
be	O
more	O
efficient	O
when	O
the	O
N	O
parties	O
communicate	O

The	O
fully	O
entangled	B-KEY
fraction	O
as	O
an	O
inclusive	O
measure	O
of	O
entanglement	O
applications	O
Characterizing	O
entanglement	B-KEY
in	O
all	O
but	O
the	O
simplest	O
case	O
of	O
a	O
two	B-KEY
qubit	I-KEY
pure	I-KEY
state	I-KEY
is	O
a	O
hard	O
problem	O
,	O
even	O
understanding	O
the	O
relevant	O
experimental	O
quantities	O
that	O
are	O
related	O
to	O
entanglement	B-KEY
is	O
difficult	O
.	O
It	O
may	O
not	O
be	O
necessary	O
,	O
however	O
,	O
to	O
quantify	O
the	O
entanglement	B-KEY
of	O
a	O
state	O
in	O
order	O
to	O
quantify	O
the	O
quantum	B-KEY
information	I-KEY
processing	I-KEY
significance	O
of	O
a	O
state	O
.	O
It	O
is	O
known	O
that	O
the	O
fully	O
entangled	B-KEY
fraction	O
has	O
a	O
direct	O
relationship	O
to	O
the	O
fidelity	O
of	O
teleportation	O
maximized	O
under	O
the	O
actions	O
of	O
local	O
unitary	O
operations	O
.	O
In	O
the	O
case	O
of	O
two	O
qubits	O
we	O
point	O
out	O
that	O
the	O
fully	O
entangled	B-KEY
fraction	O
can	O
also	O
be	O
related	O
to	O
the	O
fidelities	O
,	O
maximized	O
under	O
the	O
actions	O
of	O
local	O
unitary	O
operations	O
,	O
of	O
other	O
important	O
quantum	O
information	O
tasks	O
such	O
as	O
dense	O
coding	O
,	O
entanglement	O
swapping	O
and	O
quantum	O
cryptography	O
in	O
such	O
a	O
way	O
as	O
to	O
provide	O
an	O
inclusive	O
measure	O
of	O
these	O
entanglement	O
applications	O
.	O
For	O
two	O
qubit	O
systems	O
the	O
fully	O
entangled	B-KEY
fraction	O
has	O
a	O
simple	O
known	O
closed-form	O
expression	O
and	O
we	O
establish	O
lower	O
and	O
upper	O
bounds	O
of	O
this	O
quantity	O
with	O
the	O
concurrence	O
.	O
This	O
approach	O
is	O
readily	O
extendable	O
to	O
more	O
complicated	O
systems	O

Efficient	B-KEY
tracking	I-KEY
of	O
the	O
cross-correlation	B-KEY
coefficient	I-KEY
In	O
many	O
-LRB-	O
audio	O
-RRB-	O
processing	O
algorithms	O
,	O
involving	O
manipulation	O
of	O
discrete-time	B-KEY
signals	I-KEY
,	O
the	O
performance	O
can	O
vary	O
strongly	O
over	O
the	O
repertoire	O
that	O
is	O
used	O
.	O
This	O
may	O
be	O
the	O
case	O
when	O
the	O
signals	O
from	O
the	O
various	O
channels	O
are	O
allowed	O
to	O
be	O
strongly	O
positively	O
or	O
negatively	O
correlated	O
.	O
We	O
propose	O
and	O
analyze	O
a	O
general	O
formula	O
for	O
tracking	O
the	O
-LRB-	O
time-dependent	O
-RRB-	O
correlation	O
between	O
two	O
signals	O
.	O
Some	O
special	O
cases	O
of	O
this	O
formula	O
lead	O
to	O
classical	O
results	O
known	O
from	O
the	O
literature	O
,	O
others	O
are	O
new	O
.	O
This	O
formula	O
is	O
recursive	O
in	O
nature	O
,	O
and	O
uses	O
only	O
the	O
instantaneous	O
values	O
of	O
the	O
two	O
signals	O
,	O
in	O
a	O
low-cost	O
and	O
low-complexity	O
manner	O
;	O
in	O
particular	O
,	O
there	O
is	O
no	O
need	O
to	O
take	O
square	O
roots	O
or	O
to	O
carry	O
out	O
divisions	O
.	O
Furthermore	O
,	O
this	O
formula	O
can	O
be	O
modified	O
with	O
respect	O
to	O
the	O
occurrence	O
of	O
the	O
two	O
signals	O
so	O
as	O
to	O
further	O
decrease	O
the	O
complexity	O
,	O
and	O
increase	O
ease	O
of	O
implementation	O
.	O
The	O
latter	O
modification	O
comes	O
at	O
the	O
expense	O
that	O
not	O
the	O
actual	O
correlation	O
is	O
tracked	O
,	O
but	O
,	O
rather	O
,	O
a	O
somewhat	O
deformed	O
version	O
of	O
it	O
.	O
To	O
overcome	O
this	O
problem	O
,	O
we	O
propose	O
,	O
for	O
a	O
number	O
of	O
instances	O
of	O
the	O
tracking	O
formula	O
,	O
a	O
simple	O
warping	B-KEY
operation	I-KEY
on	O
the	O
deformed	B-KEY
correlation	I-KEY
.	O
Now	O
we	O
obtain	O
,	O
at	O
least	O
for	O
sinusoidal	B-KEY
signals	I-KEY
,	O
the	O
correct	O
value	O
of	O
the	O
correlation	O
coefficient	O
.	O
Special	O
attention	O
is	O
paid	O
to	O
the	O
convergence	B-KEY
behavior	I-KEY
of	O
the	O
algorithm	O
for	O
stationary	B-KEY
signals	I-KEY
and	O
the	O
dynamic	B-KEY
behavior	I-KEY
if	O
there	O
is	O
a	O
transition	O
to	O
another	O
stationary	B-KEY
state	I-KEY
;	O
the	O
latter	O
is	O
considered	O
to	O
be	O
important	O
to	O
study	O
the	O
tracking	O
abilities	O
to	O
nonstationary	B-KEY
signals	I-KEY
.	O
We	O
illustrate	O
tracking	B-KEY
algorithm	I-KEY
by	O
using	O
it	O
for	O
stereo	B-KEY
music	I-KEY
fragments	I-KEY
,	O
obtained	O
from	O
a	O
number	O
of	O
digital	B-KEY
audio	I-KEY
recordings	I-KEY

A	O
brief	O
history	O
of	O
electronic	B-KEY
reserves	I-KEY
Electronic	B-KEY
reserves	I-KEY
has	O
existed	O
as	O
a	O
library	B-KEY
service	I-KEY
for	O
barely	O
ten	O
years	O
,	O
yet	O
its	O
history	O
,	O
however	O
brief	O
,	O
is	O
important	O
as	O
an	O
indicator	O
of	O
the	O
direction	O
being	O
taken	O
by	O
the	O
profession	O
of	O
Librarianship	B-KEY
as	O
a	O
whole	O
.	O
Recent	O
improvements	O
in	O
technology	O
and	O
a	O
desire	O
to	O
provide	O
better	O
service	O
to	O
students	B-KEY
and	O
faculty	B-KEY
have	O
resulted	O
in	O
the	O
implementation	O
of	O
e-reserves	O
by	O
ever	O
greater	O
numbers	O
of	O
academic	B-KEY
libraries	I-KEY
.	O
Yet	O
a	O
great	O
deal	O
of	O
confusion	O
still	O
surrounds	O
the	O
issue	O
of	O
copyright	B-KEY
compliance	I-KEY
.	O
Negotiation	B-KEY
,	O
litigation	B-KEY
,	O
and	O
legislation	B-KEY
in	O
particular	O
have	O
framed	O
the	O
debate	O
over	O
the	O
application	O
of	O
fair	O
use	O
to	O
an	O
e-reserves	B-KEY
environment	I-KEY
,	O
and	O
the	O
question	O
of	O
whether	O
or	O
not	O
permission	B-KEY
fees	I-KEY
should	O
be	O
paid	O
to	O
rights	O
holders	O
,	O
but	O
as	O
of	O
yet	O
no	O
definitive	O
answers	O
or	O
standards	O
have	O
emerged	O

Access	B-KEY
privilege	I-KEY
management	I-KEY
in	O
protection	B-KEY
systems	I-KEY
We	O
consider	O
the	O
problem	O
of	O
managing	O
access	O
privileges	O
on	O
protected	B-KEY
objects	I-KEY
.	O
We	O
associate	O
one	O
or	O
more	O
locks	B-KEY
with	O
each	O
object	O
,	O
one	O
lock	B-KEY
for	O
each	O
access	O
right	O
defined	O
by	O
the	O
object	O
type	O
.	O
Possession	O
of	O
an	O
access	O
right	O
on	O
a	O
given	O
object	O
is	O
certified	O
by	O
possession	O
of	O
a	O
key	O
for	O
this	O
object	O
,	O
if	O
this	O
key	O
matches	O
one	O
of	O
the	O
object	O
locks	B-KEY
.	O
We	O
introduce	O
a	O
number	O
of	O
variants	O
to	O
this	O
basic	O
key-lock	O
technique	O
.	O
Polymorphic	B-KEY
access	I-KEY
rights	I-KEY
make	O
it	O
possible	O
to	O
decrease	O
the	O
number	O
of	O
keys	O
required	O
to	O
certify	O
possession	O
of	O
complex	O
access	O
privileges	O
that	O
are	O
defined	O
in	O
terms	O
of	O
several	O
access	O
rights	O
.	O
Multiple	O
locks	B-KEY
on	O
the	O
same	O
access	O
right	O
allow	O
us	O
to	O
exercise	O
forms	O
of	O
selective	B-KEY
revocation	I-KEY
of	O
access	O
privileges	O
.	O
A	O
lock	B-KEY
conversion	O
function	O
can	O
be	O
used	O
to	O
reduce	O
the	O
number	O
of	O
locks	O
associated	O
with	O
any	O
given	O
object	O
to	O
a	O
single	O
lock	O
.	O
The	O
extent	O
of	O
the	O
results	O
obtained	O
is	O
evaluated	O
in	O
relation	O
to	O
alternative	O
methodologies	O
for	O
access	B-KEY
privilege	I-KEY
management	I-KEY

Relationship	O
between	O
strong	B-KEY
monotonicity	I-KEY
property	I-KEY
,	O
P/sub	O
2	O
/	O
-	O
property	O
,	O
and	O
the	O
GUS-property	B-KEY
in	O
semidefinite	B-KEY
linear	I-KEY
complementarity	I-KEY
problems	I-KEY
In	O
a	O
recent	O
paper	O
on	O
semidefinite	B-KEY
linear	I-KEY
complementarity	I-KEY
problems	I-KEY
,	O
Gowda	O
and	O
Song	O
-LRB-	O
2000	O
-RRB-	O
introduced	O
and	O
studied	O
the	O
P-property	O
,	O
P/sub	O
2	O
/	O
-	O
property	O
,	O
GUS-property	B-KEY
,	O
and	O
strong	B-KEY
monotonicity	I-KEY
property	I-KEY
for	O
linear	B-KEY
transformation	I-KEY
L	O
:	O
S/sup	O
n	O
/	O
to	O
S/sup	O
n	O
/	O
,	O
where	O
S/sup	O
n	O
/	O
is	O
the	O
space	O
of	O
all	O
symmetric	O
and	O
real	O
n	O
*	O
n	O
matrices	O
.	O
In	O
an	O
attempt	O
to	O
characterize	O
the	O
P/sub	O
2	O
/	O
-	O
property	O
,	O
they	O
raised	O
the	O
following	O
two	O
questions	O
:	O
-LRB-	O
i	O
-RRB-	O
Does	O
the	O
strong	O
monotonicity	O
imply	O
the	O
P/sub	O
2	O
/	O
-	O
property	O
?	O
-LRB-	O
ii	O
-RRB-	O
Does	O
the	O
GUS-property	B-KEY
imply	O
the	O
P/sub	O
2	O
/	O
-	O
property	O
?	O
In	O
this	O
paper	O
,	O
we	O
show	O
that	O
the	O
strong	B-KEY
monotonicity	I-KEY
property	I-KEY
implies	O
the	O
P/sub	O
2	O
/	O
-	O
property	O
for	O
any	O
linear	B-KEY
transformation	I-KEY
and	O
describe	O
an	O
equivalence	O
between	O
these	O
two	O
properties	O
for	O
Lyapunov	O
and	O
other	O
transformations	O
.	O
We	O
show	O
by	O
means	O
of	O
an	O
example	O
that	O
the	O
GUS-property	B-KEY
need	O
not	O
imply	O
the	O
P/sub	O
2	O
/	O
-	O
property	O
,	O
even	O
for	O
Lyapunov	B-KEY
transformations	I-KEY

Marketing	B-KEY
in	O
CSIR	B-KEY
libraries	I-KEY
and	O
information	B-KEY
centres	I-KEY
:	O
a	O
study	O
on	O
promotional	O
efforts	O
This	O
paper	O
examines	O
the	O
attitudes	O
of	O
librarians	O
towards	O
the	O
promotional	O
aspects	O
in	O
several	O
CSIR	B-KEY
libraries	I-KEY
and	O
information	B-KEY
centres	I-KEY
of	O
India	B-KEY
.	O
The	O
issues	O
related	O
to	O
promotional	B-KEY
activities	I-KEY
of	O
these	O
libraries	O
have	O
been	O
evaluated	O
to	O
determine	O
the	O
extent	O
to	O
which	O
they	O
are	O
being	O
practised	O
.	O
Librarians	O
hold	O
positive	O
attitudes	O
about	O
promotional	O
aspects	O
of	O
libraries	O
and	O
often	O
practise	O
them	O
without	O
knowing	O
they	O
are	O
practising	O
marketing	B-KEY
concepts	O
.	O
Suggestions	O
and	O
strategies	O
for	O
improving	O
the	O
promotional	B-KEY
activities	I-KEY
in	O
libraries	O
and	O
information	O
services	O
are	O
put	O
forth	O
so	O
as	O
to	O
meet	O
the	O
information	B-KEY
needs	I-KEY
and	O
demands	O
of	O
clientele	O

Analysis	O
of	O
the	O
surface	B-KEY
roughness	I-KEY
and	O
dimensional	B-KEY
accuracy	I-KEY
capability	I-KEY
of	O
fused	B-KEY
deposition	I-KEY
modelling	I-KEY
processes	I-KEY
Building	O
up	O
materials	O
in	O
layers	O
poses	O
significant	O
challenges	O
from	O
the	O
viewpoint	O
of	O
material	O
science	O
,	O
heat	O
transfer	O
and	O
applied	O
mechanics	O
.	O
However	O
,	O
numerous	O
aspects	O
of	O
the	O
use	O
of	O
these	O
technologies	O
have	O
yet	O
to	O
be	O
studied	O
.	O
One	O
of	O
these	O
aspects	O
is	O
the	O
characterization	O
of	O
the	O
surface	B-KEY
roughness	I-KEY
and	O
dimensional	B-KEY
precision	I-KEY
obtainable	O
in	O
layered	B-KEY
manufacturing	I-KEY
processes	I-KEY
.	O
In	O
this	O
paper	O
,	O
a	O
study	O
of	O
roughness	O
parameters	O
obtained	O
through	O
the	O
use	O
of	O
these	O
manufacturing	O
processes	O
was	O
made	O
.	O
Prototype	B-KEY
parts	I-KEY
were	O
manufactured	O
using	O
FDM	O
techniques	O
and	O
an	O
experimental	O
analysis	O
of	O
the	O
resulting	O
roughness	B-KEY
average	I-KEY
-LRB-	O
R/sub	O
a	O
/	O
-RRB-	O
and	O
rms	B-KEY
roughness	I-KEY
-LRB-	O
R/sub	O
q	O
/	O
-RRB-	O
obtained	O
through	O
the	O
use	O
of	O
these	O
manufacturing	O
processes	O
was	O
carried	O
out	O
.	O
Dimensional	O
parameters	O
were	O
also	O
studied	O
in	O
order	O
to	O
determine	O
the	O
capability	O
of	O
the	O
Fused	B-KEY
Deposition	I-KEY
Modelling	I-KEY
process	I-KEY
for	O
manufacturing	O
parts	O

Schedulability	O
analysis	O
of	O
real-time	O
traffic	O
in	O
WorldFIP	B-KEY
networks	I-KEY
:	O
an	O
integrated	O
approach	O
The	O
WorldFIP	O
protocol	O
is	O
one	O
of	O
the	O
profiles	O
that	O
constitute	O
the	O
European	O
fieldbus	O
standard	O
EN-50170	O
.	O
It	O
is	O
particularly	O
well	O
suited	O
to	O
be	O
used	O
in	O
distributed	B-KEY
computer-controlled	I-KEY
systems	I-KEY
where	O
a	O
set	O
of	O
process	O
variables	O
must	O
be	O
shared	O
among	O
network	O
devices	O
.	O
To	O
cope	O
with	O
the	O
real-time	O
requirements	O
of	O
such	O
systems	O
,	O
the	O
protocol	O
provides	O
communication	B-KEY
services	I-KEY
based	O
on	O
the	O
exchange	O
of	O
periodic	O
and	O
aperiodic	O
identified	O
variables	O
.	O
The	O
periodic	O
exchanges	O
have	O
the	O
highest	O
priority	O
and	O
are	O
executed	O
at	O
run	O
time	O
according	O
to	O
a	O
cyclic	O
schedule	O
.	O
Therefore	O
,	O
the	O
respective	O
schedulability	O
can	O
be	O
determined	O
at	O
pre-run-time	O
when	O
building	O
the	O
schedule	O
table	O
.	O
Concerning	O
the	O
aperiodic	B-KEY
exchanges	I-KEY
,	O
the	O
situation	O
is	O
different	O
since	O
their	O
priority	O
is	O
lower	O
and	O
they	O
are	O
bandied	O
according	O
to	O
a	O
first-come-first-served	B-KEY
policy	I-KEY
.	O
In	O
this	O
paper	O
,	O
a	O
response-time-based	O
schedulability	O
analysis	O
for	O
the	O
real-time	O
traffic	O
is	O
presented	O
.	O
Such	O
analysis	O
considers	O
both	O
types	O
of	O
traffic	O
in	O
an	O
integrated	O
way	O
,	O
according	O
to	O
their	O
priorities	O
.	O
Furthermore	O
,	O
a	O
fixed-priorities-based	O
policy	O
is	O
also	O
used	O
to	O
schedule	O
the	O
periodic	O
traffic	O
.	O
The	O
proposed	O
analysis	O
represents	O
an	O
improvement	O
relative	O
to	O
previous	O
work	O
and	O
it	O
can	O
be	O
evaluated	O
online	O
as	O
part	O
of	O
a	O
traffic	B-KEY
online	I-KEY
admission	I-KEY
control	I-KEY
.	O
This	O
feature	O
is	O
of	O
particular	O
importance	O
when	O
a	O
planning	O
scheduler	O
is	O
used	O
,	O
instead	O
of	O
the	O
typical	O
offline	O
static	O
scheduler	O
,	O
to	O
allow	O
online	O
changes	O
to	O
the	O
set	O
of	O
periodic	B-KEY
process	I-KEY
variables	I-KEY

Control	O
of	O
a	O
heavy-duty	O
robotic	B-KEY
excavator	I-KEY
using	O
time	O
delay	O
control	O
with	O
integral	B-KEY
sliding	I-KEY
surface	I-KEY
The	O
control	O
of	O
a	O
robotic	B-KEY
excavator	I-KEY
is	O
difficult	O
from	O
the	O
standpoint	O
of	O
the	O
following	O
problems	O
:	O
parameter	O
variations	O
in	O
mechanical	O
structures	O
,	O
various	O
nonlinearities	O
in	O
hydraulic	O
actuators	O
and	O
disturbance	O
due	O
to	O
the	O
contact	O
with	O
the	O
ground	O
.	O
In	O
addition	O
,	O
the	O
more	O
the	O
size	O
of	O
robotic	B-KEY
excavators	I-KEY
increases	O
,	O
the	O
more	O
the	O
length	O
and	O
mass	O
of	O
the	O
excavator	O
links	O
;	O
the	O
more	O
the	O
parameters	O
of	O
a	O
heavy-duty	O
excavator	O
vary	O
.	O
A	O
time-delay	B-KEY
control	I-KEY
with	O
switching	O
action	O
-LRB-	O
TDCSA	O
-RRB-	O
using	O
an	O
integral	B-KEY
sliding	I-KEY
surface	I-KEY
is	O
proposed	O
in	O
this	O
paper	O
for	O
the	O
control	O
of	O
a	O
21-ton	O
robotic	B-KEY
excavator	I-KEY
.	O
Through	O
analysis	O
and	O
experiments	O
,	O
we	O
show	O
that	O
using	O
an	O
integral	B-KEY
sliding	I-KEY
surface	I-KEY
for	O
the	O
switching	O
action	O
of	O
TDCSA	O
is	O
better	O
than	O
using	O
a	O
PD-type	O
sliding	O
surface	O
.	O
The	O
proposed	O
controller	O
is	O
applied	O
to	O
straight-line	O
motions	O
of	O
a	O
21-ton	O
robotic	B-KEY
excavator	I-KEY
with	O
a	O
speed	O
level	O
at	O
which	O
skillful	O
operators	O
work	O
.	O
Experiments	O
,	O
which	O
were	O
designed	O
for	O
surfaces	O
with	O
various	O
inclinations	O
and	O
over	O
broad	O
ranges	O
of	O
joint	O
motions	O
,	O
show	O
that	O
the	O
proposed	O
controller	O
exhibits	O
good	O
performance	O

Use	O
of	O
SPOT	B-KEY
images	I-KEY
as	O
a	O
tool	O
for	O
coastal	B-KEY
zone	I-KEY
management	I-KEY
and	O
monitoring	O
of	O
environmental	O
impacts	O
in	O
the	O
coastal	O
zone	O
Modern	O
techniques	O
such	O
as	O
remote	B-KEY
sensing	I-KEY
have	O
been	O
one	O
of	O
the	O
main	O
factors	O
leading	O
toward	O
the	O
achievement	O
of	O
serious	O
plans	O
regarding	O
coastal	O
management	O
.	O
A	O
multitemporal	B-KEY
analysis	I-KEY
of	O
land	B-KEY
use	I-KEY
in	O
certain	O
areas	O
of	O
the	O
Colombian	B-KEY
Caribbean	I-KEY
Coast	I-KEY
is	O
described	O
.	O
It	O
mainly	O
focuses	O
on	O
environmental	O
impacts	O
caused	O
by	O
anthropogenic	B-KEY
activities	I-KEY
,	O
such	O
as	O
deforestation	O
of	O
mangroves	O
due	O
to	O
shrimp	B-KEY
farming	I-KEY
.	O
Selection	O
of	O
sensitive	O
areas	O
,	O
percentage	O
of	O
destroyed	O
mangroves	O
,	O
possible	O
endangered	B-KEY
areas	I-KEY
,	O
etc.	O
,	O
are	O
some	O
of	O
the	O
results	O
of	O
this	O
analysis	O
.	O
Recommendations	O
for	O
a	O
coastal	O
management	O
plan	O
in	O
the	O
area	O
have	O
also	O
resulted	O
from	O
this	O
analysis	O
.	O
Some	O
other	O
consequences	O
of	O
the	O
deforestation	O
of	O
mangroves	O
in	O
the	O
coastal	O
zone	O
and	O
the	O
construction	O
of	O
shrimp	B-KEY
ponds	I-KEY
are	O
also	O
analyzed	O
,	O
such	O
as	O
the	O
increase	O
of	O
erosion	B-KEY
problems	I-KEY
in	O
these	O
areas	O
and	O
water	B-KEY
pollution	I-KEY
,	O
among	O
others	O
.	O
The	O
increase	O
of	O
erosion	O
in	O
these	O
areas	O
has	O
also	O
changed	O
part	O
of	O
their	O
morphology	O
,	O
which	O
has	O
been	O
studied	O
by	O
the	O
analysis	O
of	O
SPOT	B-KEY
images	I-KEY
in	O
previous	O
years	O
.	O
A	O
serious	O
concern	O
exists	O
about	O
the	O
future	O
of	O
these	O
areas	O
.	O
For	O
this	O
reason	O
new	O
techniques	O
like	O
satellite	B-KEY
images	I-KEY
-LRB-	O
SPOT	O
-RRB-	O
have	O
been	O
applied	O
with	O
good	O
results	O
,	O
leading	O
to	O
more	O
effective	O
control	O
and	O
coastal	O
management	O
in	O
the	O
area	O
.	O
The	O
use	O
of	O
SPOT	B-KEY
images	I-KEY
to	O
study	O
changes	O
of	O
the	O
land	B-KEY
use	I-KEY
of	O
the	O
area	O
is	O
a	O
useful	O
technique	O
to	O
determine	O
patterns	O
of	O
human	B-KEY
activities	I-KEY
and	O
suggest	O
solutions	O
for	O
severe	O
problems	O
in	O
these	O
areas	O

Evaluating	O
the	O
complexity	O
of	O
index	O
sets	O
for	O
families	O
of	O
general	B-KEY
recursive	I-KEY
functions	I-KEY
in	O
the	O
arithmetic	B-KEY
hierarchy	I-KEY
The	O
complexity	O
of	O
index	O
sets	O
of	O
families	O
of	O
general	B-KEY
recursive	I-KEY
functions	I-KEY
is	O
evaluated	O
in	O
the	O
Kleene-Mostowski	O
arithmetic	B-KEY
hierarchy	I-KEY

Support	B-KEY
communities	I-KEY
for	O
women	B-KEY
in	O
computing	B-KEY
This	O
article	O
highlights	O
the	O
many	O
activities	O
provided	O
by	O
the	O
support	B-KEY
communities	I-KEY
available	O
for	O
women	B-KEY
in	O
computing	B-KEY
.	O
Thousands	O
of	O
women	B-KEY
actively	O
participate	O
in	O
these	O
programs	O
and	O
they	O
receive	O
many	O
benefits	O
including	O
networking	B-KEY
and	O
professional	B-KEY
support	I-KEY
.	O
In	O
addition	O
,	O
the	O
organizations	O
and	O
associations	O
help	O
promote	O
the	O
accomplishments	O
of	O
women	B-KEY
computer	B-KEY
scientists	O
and	O
disseminate	O
valuable	O
information	O
.	O
This	O
article	O
surveys	O
some	O
of	O
these	O
organizations	O
and	O
concludes	O
with	O
a	O
list	O
of	O
suggestions	O
for	O
how	O
faculty	B-KEY
members	I-KEY
can	O
incorporate	O
the	O
benefits	O
of	O
these	O
organizations	O
in	O
their	O
own	O
institutions	O

Networking	O
without	O
wires	O
Several	O
types	O
of	O
devices	O
use	O
radio	O
transmitters	O
to	O
send	O
data	O
over	O
thin	O
air	O
.	O
Are	O
WLANs	O
,	O
wireless	B-KEY
local	I-KEY
area	I-KEY
networks	I-KEY
,	O
the	O
end	O
to	O
all	O
cables	O
?	O
Will	O
Dalrymple	O
weighs	O
up	O
the	O
costs	B-KEY
and	O
benefits	B-KEY

Detection	O
of	O
flaws	O
in	O
composites	B-KEY
from	O
scattered	B-KEY
elastic-wave	I-KEY
field	I-KEY
using	O
an	O
improved	B-KEY
mu	I-KEY
GA	I-KEY
and	O
a	O
local	B-KEY
optimizer	I-KEY
An	O
effective	O
technique	O
for	O
flaw	B-KEY
detection	I-KEY
of	O
composites	B-KEY
is	O
proposed	O
.	O
In	O
this	O
technique	O
,	O
the	O
detection	O
problem	O
is	O
formulated	O
as	O
an	O
optimization	B-KEY
problem	I-KEY
minimizing	O
the	O
difference	O
between	O
the	O
measured	O
and	O
calculated	O
surface	B-KEY
displacement	I-KEY
response	I-KEY
derived	O
from	O
scattered	B-KEY
elastic-wave	I-KEY
fields	I-KEY
.	O
A	O
combined	O
optimization	O
technique	O
using	O
an	O
improved	B-KEY
mu	I-KEY
GA	I-KEY
and	O
a	O
local	B-KEY
optimizer	I-KEY
is	O
developed	O
to	O
solve	O
the	O
optimization	B-KEY
problem	I-KEY
so	O
as	O
to	O
obtain	O
the	O
flaw	O
parameters	O
defining	O
flaw	B-KEY
configurations	I-KEY
.	O
Guidelines	O
for	O
implementing	O
the	O
detection	O
technique	O
,	O
including	O
formulation	O
of	O
the	O
objective	B-KEY
function	I-KEY
of	O
the	O
optimization	B-KEY
problem	I-KEY
using	O
different	O
error	B-KEY
norms	I-KEY
,	O
improvement	O
of	O
mu	O
GA	O
convergence	B-KEY
performance	O
,	O
switching	O
from	O
mu	O
GA	O
to	O
local	B-KEY
optimizer	I-KEY
in	O
the	O
optimization	O
process	O
,	O
and	O
suppression	O
of	O
the	O
effect	O
of	O
noise	O
on	O
detection	O
results	O
,	O
are	O
addressed	O
in	O
detail	O
.	O
Numerical	O
examples	O
are	O
presented	O
to	O
demonstrate	O
the	O
effectiveness	O
and	O
efficiency	O
of	O
the	O
proposed	O
detection	O
technique	O

Comparison	O
of	O
automated	B-KEY
digital	I-KEY
elevation	I-KEY
model	I-KEY
extraction	I-KEY
results	O
using	O
along-track	O
ASTER	O
and	O
across-track	B-KEY
SPOT	I-KEY
stereo	I-KEY
images	I-KEY
A	O
digital	O
elevation	O
model	O
-LRB-	O
DEM	O
-RRB-	O
can	O
be	O
extracted	O
automatically	O
from	O
stereo	B-KEY
satellite	I-KEY
images	I-KEY
.	O
During	O
the	O
past	O
decade	O
,	O
the	O
most	O
common	O
satellite	O
data	O
used	O
to	O
extract	O
DEM	O
was	O
the	O
across-track	O
SPOT	O
.	O
Recently	O
,	O
the	O
addition	O
of	O
along-track	B-KEY
ASTER	I-KEY
data	I-KEY
,	O
which	O
can	O
be	O
downloaded	O
freely	O
,	O
provides	O
another	O
attractive	O
alternative	O
to	O
extract	O
DEM	O
data	O
.	O
This	O
work	O
compares	O
the	O
automated	O
DEM	O
extraction	O
results	O
using	O
an	O
ASTER	B-KEY
stereo	I-KEY
pair	I-KEY
and	O
a	O
SPOT	O
stereo	O
pair	O
over	O
an	O
area	O
of	O
hilly	O
mountains	O
in	O
Drum	O
Mountain	O
,	O
Utah	O
,	O
when	O
compared	O
to	O
a	O
USGS	O
7.5-min	O
DEM	O
standard	O
product	O
.	O
The	O
result	O
shows	O
that	O
SPOT	O
produces	O
better	O
DEM	O
results	O
in	O
terms	O
of	O
accuracy	O
and	O
details	O
,	O
if	O
the	O
radiometric	B-KEY
variations	I-KEY
between	O
the	O
images	O
,	O
taken	O
on	O
subsequent	O
satellite	O
revolutions	O
,	O
are	O
small	O
.	O
Otherwise	O
,	O
the	O
ASTER	B-KEY
stereo	I-KEY
pair	I-KEY
is	O
a	O
better	O
choice	O
because	O
of	O
simultaneous	B-KEY
along-track	I-KEY
acquisition	I-KEY
during	O
a	O
single	O
pass	O
.	O
Compared	O
to	O
the	O
USGS	O
7.5-min	O
DEM	O
,	O
the	O
ASTER	O
and	O
the	O
SPOT	O
extracted	O
DEMs	O
have	O
a	O
standard	O
deviation	O
of	O
11.6	O
and	O
4.6	O
m	O
,	O
respectively	O

High-performance	O
servo	B-KEY
systems	I-KEY
based	O
on	O
multirate	B-KEY
sampling	I-KEY
control	I-KEY
In	O
this	O
paper	O
,	O
novel	O
multirate	O
two-degree-of-freedom	O
controllers	O
are	O
proposed	O
for	O
digital	B-KEY
control	I-KEY
systems	I-KEY
,	O
in	O
which	O
the	O
sampling	O
period	O
of	O
plant	O
output	O
is	O
restricted	O
to	O
be	O
relatively	O
longer	O
than	O
the	O
control	O
period	O
of	O
plant	O
input	O
.	O
The	O
proposed	O
feedforward	B-KEY
controller	O
assures	O
perfect	O
tracking	B-KEY
at	O
M	O
inter-sampling	O
points	O
.	O
On	O
the	O
other	O
hand	O
,	O
the	O
proposed	O
feedback	B-KEY
controller	O
assures	O
perfect	O
disturbance	B-KEY
rejection	I-KEY
at	O
M	O
inter-sample	O
points	O
in	O
the	O
steady	O
state	O
.	O
Illustrative	O
examples	O
of	O
position	B-KEY
control	I-KEY
for	O
hard	B-KEY
disk	I-KEY
drive	I-KEY
are	O
presented	O
,	O
and	O
advantages	O
of	O
these	O
approaches	O
are	O
demonstrated	O

An	O
ACM-W	B-KEY
literature	I-KEY
review	I-KEY
on	O
women	O
in	O
computing	O
The	O
pipeline	B-KEY
shrinkage	I-KEY
problem	I-KEY
for	O
women	O
in	O
computer	O
science	O
is	O
a	O
well-known	O
and	O
documented	O
phenomenon	O
where	O
the	O
ratio	O
of	O
women	O
to	O
men	O
involved	O
in	O
computing	O
shrinks	O
dramatically	O
from	O
early	O
student	O
years	O
to	O
working	O
years	O
.	O
During	O
the	O
last	O
decade	O
,	O
considerable	O
research	O
ensued	O
to	O
understand	O
the	O
reasons	O
behind	O
the	O
existence	O
of	O
the	O
shrinking	O
pipeline	O
and	O
in	O
some	O
cases	O
to	O
take	O
action	O
to	O
increase	O
the	O
numbers	O
of	O
women	O
in	O
computing	O
.	O
Through	O
the	O
work	O
of	O
a	O
National	O
Science	O
Foundation	O
funded	O
project	O
,	O
ACM	O
's	O
Committee	O
on	O
Women	O
in	O
Computing	O
-LRB-	O
ACM-W	O
-RRB-	O
has	O
taken	O
a	O
first	O
step	O
towards	O
pulling	O
this	O
research	O
together	O
.	O
A	O
large	O
number	O
of	O
articles	O
was	O
gathered	O
and	O
processed	O
on	O
the	O
topic	O
of	O
women	O
in	O
computing	O
and	O
the	O
shrinking	O
pipeline	O
.	O
The	O
committee	O
created	O
a	O
publicly	O
available	O
online	O
database	O
to	O
organize	O
the	O
references	O
of	O
this	O
body	O
of	O
work	O
by	O
topic	O
,	O
author	O
,	O
and	O
reference	O
information	O
.	O
The	O
database	O
,	O
constantly	O
being	O
updated	O
,	O
is	O
accessible	O
through	O
ACM-W	O
's	O
website	O
<	O
http://www.acm.org/women	O
>	O
.	O
A	O
final	O
report	O
is	O
also	O
available	O
via	O
the	O
ACM-W	O
Web	O
site	O
which	O
covers	O
current	O
statistics	O
on	O
women	O
in	O
computing	O
,	O
summaries	O
of	O
the	O
literature	O
in	O
the	O
database	O
,	O
and	O
a	O
set	O
of	O
recommendations	O
.	O
The	O
article	O
is	O
a	O
brief	O
synopsis	O
of	O
a	O
subset	O
of	O
the	O
literature	O
review	O
as	O
of	O
August	O
2001	O

Prediction	O
of	O
tool	O
and	O
chip	O
temperature	O
in	O
continuous	O
and	O
interrupted	B-KEY
machining	I-KEY
A	O
numerical	B-KEY
model	I-KEY
based	O
on	O
the	O
finite	B-KEY
difference	I-KEY
method	I-KEY
is	O
presented	O
to	O
predict	O
tool	O
and	O
chip	O
temperature	O
fields	O
in	O
continuous	B-KEY
machining	I-KEY
and	O
time	B-KEY
varying	I-KEY
milling	I-KEY
processes	I-KEY
.	O
Continuous	O
or	O
steady	O
state	O
machining	O
operations	O
like	O
orthogonal	B-KEY
cutting	I-KEY
are	O
studied	O
by	O
modeling	O
the	O
heat	B-KEY
transfer	I-KEY
between	O
the	O
tool	O
and	O
chip	O
at	O
the	O
tool-rake	B-KEY
face	I-KEY
contact	I-KEY
zone	I-KEY
.	O
The	O
shear	B-KEY
energy	I-KEY
created	O
in	O
the	O
primary	B-KEY
zone	I-KEY
,	O
the	O
friction	B-KEY
energy	I-KEY
produced	O
at	O
the	O
rake	O
face-chip	O
contact	O
zone	O
and	O
the	O
heat	O
balance	O
between	O
the	O
moving	O
chip	O
and	O
stationary	O
tool	O
are	O
considered	O
.	O
The	O
temperature	B-KEY
distribution	I-KEY
is	O
solved	O
using	O
the	O
finite	B-KEY
difference	I-KEY
method	I-KEY
.	O
Later	O
,	O
the	O
model	O
is	O
extended	O
to	O
milling	O
where	O
the	O
cutting	O
is	O
interrupted	O
and	O
the	O
chip	O
thickness	O
varies	O
with	O
time	O
.	O
The	O
proposed	O
model	O
combines	O
the	O
steady-state	O
temperature	O
prediction	O
in	O
continuous	B-KEY
machining	I-KEY
with	O
transient	O
temperature	O
evaluation	O
in	O
interrupted	O
cutting	O
operations	O
where	O
the	O
chip	O
and	O
the	O
process	O
change	O
in	O
a	O
discontinuous	O
manner	O
.	O
The	O
mathematical	O
models	O
and	O
simulation	O
results	O
are	O
in	O
satisfactory	O
agreement	O
with	O
experimental	O
temperature	O
measurements	O
reported	O
in	O
the	O
literature	O

A	O
high-resolution	O
high-frequency	O
monolithic	B-KEY
top-shooting	I-KEY
microinjector	I-KEY
free	O
of	O
satellite	B-KEY
drops	I-KEY
-	O
part	O
II	O
:	O
fabrication	O
,	O
implementation	O
,	O
and	O
characterization	O
For	O
pt	O
.	O
I	O
,	O
see	O
ibid.	O
,	O
vol	O
.	O
11	O
,	O
no.	O
5	O
,	O
p.	O
427-36	O
-LRB-	O
2002	O
-RRB-	O
.	O
Describes	O
the	O
fabrication	O
,	O
implementation	O
and	O
characterization	O
of	O
a	O
thermal	B-KEY
driven	I-KEY
microinjector	I-KEY
,	O
featuring	O
a	O
bubble	B-KEY
check	I-KEY
valve	I-KEY
and	O
monolithic	O
fabrication	O
.	O
Microfabrication	O
of	O
this	O
microinjector	O
is	O
based	O
on	O
bulk/surface-combined	B-KEY
micromachining	I-KEY
of	O
the	O
silicon	O
wafer	O
,	O
free	O
of	O
the	O
bonding	B-KEY
process	I-KEY
that	O
is	O
commonly	O
used	O
in	O
the	O
fabrication	O
of	O
commercial	O
printing	O
head	O
,	O
so	O
that	O
even	O
solvents	O
and	O
fuels	O
can	O
be	O
ejected	O
.	O
Droplet	O
ejection	O
sequences	O
of	O
two	O
microinjectors	O
have	O
been	O
studied	O
along	O
with	O
a	O
commercial	O
inkjet	B-KEY
printhead	I-KEY
for	O
comparison	O
.	O
The	O
droplet	O
ejection	O
of	O
our	O
microinjector	O
with	O
10	O
mu	O
m	O
diameter	O
nozzle	B-KEY
has	O
been	O
characterized	O
at	O
a	O
frequency	O
over	O
35	B-KEY
kHz	I-KEY
,	O
at	O
least	O
3	O
times	O
higher	O
than	O
those	O
of	O
commercial	O
counterparts	O
.	O
The	O
droplet	B-KEY
volume	I-KEY
from	O
this	O
device	O
is	O
smaller	O
than	O
1	O
pl	O
,	O
10	O
times	O
smaller	O
than	O
those	O
of	O
commercial	O
inkjets	O
employed	O
in	O
the	O
consumer	B-KEY
market	I-KEY
at	O
the	O
time	O
of	O
testing	O
.	O
Visualization	O
results	O
have	O
verified	O
that	O
our	O
design	O
,	O
although	O
far	O
from	O
being	O
optimized	O
,	O
operates	O
in	O
the	O
frequency	O
several	O
times	O
higher	O
than	O
those	O
of	O
commercial	O
products	O
and	O
reduces	O
the	O
crosstalk	B-KEY
among	O
neighboring	O
chambers	O

Advancements	O
during	O
the	O
past	O
quarter	O
century	O
in	O
on-line	O
monitoring	O
of	O
motor	O
and	O
generator	B-KEY
winding	I-KEY
insulation	I-KEY
Electrical	B-KEY
insulation	I-KEY
plays	O
a	O
critical	O
role	O
in	O
the	O
operation	O
of	O
motor	O
and	O
generator	O
rotor	O
and	O
stator	B-KEY
windings	I-KEY
.	O
Premature	O
failure	O
of	O
the	O
insulation	O
can	O
cost	O
millions	O
of	O
dollars	O
per	O
day	O
.	O
With	O
advancements	O
in	O
electronics	B-KEY
,	O
sensors	B-KEY
,	O
computers	B-KEY
and	O
software	B-KEY
,	O
tremendous	O
progress	O
has	O
been	O
made	O
in	O
the	O
past	O
25	O
yr	O
which	O
has	O
transformed	O
on-line	O
insulation	O
monitoring	O
from	O
a	O
rarely	O
used	O
and	O
expensive	O
tool	O
,	O
to	O
the	O
point	O
where	O
50	O
%	O
of	O
large	O
utility	O
generators	O
in	O
North	O
America	O
are	O
now	O
equipped	O
for	O
such	O
monitoring	O
.	O
This	O
review	O
paper	O
outlines	O
the	O
motivation	O
for	O
online	O
monitoring	O
,	O
discusses	O
the	O
transition	O
to	O
today	O
's	O
technology	O
,	O
and	O
describes	O
the	O
variety	O
of	O
methods	O
now	O
in	O
use	O
for	O
rotor	B-KEY
winding	I-KEY
and	O
stator	B-KEY
winding	I-KEY
monitoring	O

Distributed	B-KEY
servers	I-KEY
approach	O
for	O
large-scale	B-KEY
secure	I-KEY
multicast	I-KEY
In	O
order	O
to	O
offer	O
backward	O
and	O
forward	B-KEY
secrecy	I-KEY
for	O
multicast	B-KEY
applications	I-KEY
-LRB-	O
i.e.	O
,	O
a	O
new	O
member	O
can	O
not	O
decrypt	O
the	O
multicast	O
data	O
sent	O
before	O
its	O
joining	O
and	O
a	O
former	O
member	O
can	O
not	O
decrypt	O
the	O
data	O
sent	O
after	O
its	O
leaving	O
-RRB-	O
,	O
the	O
data	B-KEY
encryption	I-KEY
key	I-KEY
has	O
to	O
be	O
changed	O
whenever	O
a	O
user	O
joins	O
or	O
leaves	O
the	O
system	O
.	O
Such	O
a	O
change	O
has	O
to	O
be	O
made	O
known	O
to	O
all	O
the	O
current	O
users	O
.	O
The	O
bandwidth	O
used	O
for	O
such	O
re-key	B-KEY
messaging	I-KEY
can	O
be	O
high	O
when	O
the	O
user	O
pool	O
is	O
large	O
.	O
We	O
propose	O
a	O
distributed	B-KEY
servers	I-KEY
approach	O
to	O
minimize	O
the	O
overall	O
system	B-KEY
bandwidth	I-KEY
-LRB-	O
and	O
complexity	O
-RRB-	O
by	O
splitting	O
the	O
user	O
pool	O
into	O
multiple	O
groups	O
each	O
served	O
by	O
a	O
-LRB-	O
logical	O
-RRB-	O
server	O
.	O
After	O
presenting	O
an	O
analytic	O
model	O
for	O
the	O
system	O
based	O
on	O
a	O
hierarchical	B-KEY
key	I-KEY
tree	I-KEY
,	O
we	O
show	O
that	O
there	O
is	O
an	O
optimal	O
number	O
of	O
servers	O
to	O
achieve	O
minimum	O
system	B-KEY
bandwidth	I-KEY
.	O
As	O
the	O
underlying	O
user	B-KEY
traffic	I-KEY
fluctuates	O
,	O
we	O
propose	O
a	O
simple	O
dynamic	O
scheme	O
with	O
low	O
overhead	O
where	O
a	O
physical	O
server	O
adaptively	O
splits	O
and	O
merges	O
its	O
traffic	O
into	O
multiple	O
groups	O
each	O
served	O
by	O
a	O
logical	O
server	O
so	O
as	O
to	O
minimize	O
its	O
total	O
bandwidth	O
.	O
Our	O
results	O
show	O
that	O
a	O
distributed	B-KEY
servers	I-KEY
approach	O
is	O
able	O
to	O
substantially	O
reduce	O
the	O
total	O
bandwidth	O
required	O
as	O
compared	O
with	O
the	O
traditional	O
single-server	O
approach	O
,	O
especially	O
for	O
those	O
applications	O
with	O
a	O
large	O
user	O
pool	O
,	O
short	B-KEY
holding	I-KEY
time	I-KEY
,	O
and	O
relatively	O
low	O
bandwidth	O
of	O
a	O
data	O
stream	O
,	O
as	O
in	O
the	O
Internet	B-KEY
stock	I-KEY
quote	I-KEY
applications	I-KEY

Routing	B-KEY
security	I-KEY
in	O
wireless	B-KEY
ad	I-KEY
hoc	I-KEY
networks	I-KEY
A	O
mobile	B-KEY
ad	I-KEY
hoc	I-KEY
network	I-KEY
consists	O
of	O
a	O
collection	O
of	O
wireless	B-KEY
mobile	I-KEY
nodes	I-KEY
that	O
are	O
capable	O
of	O
communicating	O
with	O
each	O
other	O
without	O
the	O
use	O
of	O
a	O
network	O
infrastructure	O
or	O
any	O
centralized	O
administration	O
.	O
MANET	O
is	O
an	O
emerging	O
research	O
area	O
with	O
practical	O
applications	O
.	O
However	O
,	O
wireless	B-KEY
MANET	I-KEY
is	O
particularly	O
vulnerable	O
due	O
to	O
its	O
fundamental	O
characteristics	O
,	O
such	O
as	O
open	B-KEY
medium	I-KEY
,	O
dynamic	B-KEY
topology	I-KEY
,	O
distributed	B-KEY
cooperation	I-KEY
,	O
and	O
constrained	O
capability	O
.	O
Routing	O
plays	O
an	O
important	O
role	O
in	O
the	O
security	O
of	O
the	O
entire	O
network	O
.	O
In	O
general	O
,	O
routing	B-KEY
security	I-KEY
in	O
wireless	B-KEY
MANETs	I-KEY
appears	O
to	O
be	O
a	O
problem	O
that	O
is	O
not	O
trivial	O
to	O
solve	O
.	O
In	O
this	O
article	O
we	O
study	O
the	O
routing	B-KEY
security	I-KEY
issues	O
of	O
MANETs	O
,	O
and	O
analyze	O
in	O
detail	O
one	O
type	O
of	O
attack-the	O
``	O
black	O
hole	O
''	O
problem-that	O
can	O
easily	O
be	O
employed	O
against	O
the	O
MANETs	O
.	O
We	O
also	O
propose	O
a	O
solution	O
for	O
the	O
black	O
hole	O
problem	O
for	O
ad	O
hoc	O
on-demand	B-KEY
distance	I-KEY
vector	I-KEY
routing	I-KEY
protocol	I-KEY

How	O
to	O
drive	O
strategic	B-KEY
innovation	I-KEY
-LSB-	O
law	B-KEY
firms	I-KEY
-RSB-	O
Innovation	O
.	O
It	O
has	O
everything	O
to	O
do	O
with	O
organization	O
and	O
attitude	O
.	O
Marginal	O
improvement	O
is	O
n't	O
enough	O
anymore	O
.	O
Convert	O
your	O
problem-solving	O
skills	O
into	O
a	O
new	O
value	O
for	O
the	O
entire	O
firm	O
.	O
10	O
initiatives	O

Quantum-information	B-KEY
processing	I-KEY
by	O
nuclear	B-KEY
magnetic	I-KEY
resonance	I-KEY
:	O
Experimental	O
implementation	O
of	O
half-adder	O
and	O
subtractor	B-KEY
operations	I-KEY
using	O
an	O
oriented	O
spin-7	O
/	O
2	O
system	O
The	O
advantages	O
of	O
using	O
quantum	B-KEY
systems	I-KEY
for	O
performing	O
many	O
computational	B-KEY
tasks	I-KEY
have	O
already	O
been	O
established	O
.	O
Several	O
quantum	B-KEY
algorithms	I-KEY
have	O
been	O
developed	O
which	O
exploit	O
the	O
inherent	O
property	O
of	O
quantum	B-KEY
systems	I-KEY
such	O
as	O
superposition	O
of	O
states	O
and	O
entanglement	B-KEY
for	O
efficiently	O
performing	O
certain	O
tasks	O
.	O
The	O
experimental	O
implementation	O
has	O
been	O
achieved	O
on	O
many	O
quantum	B-KEY
systems	I-KEY
,	O
of	O
which	O
nuclear	B-KEY
magnetic	I-KEY
resonance	I-KEY
has	O
shown	O
the	O
largest	O
progress	O
in	O
terms	O
of	O
number	O
of	O
qubits	B-KEY
.	O
This	O
paper	O
describes	O
the	O
use	O
of	O
a	O
spin-7	O
/	O
2	O
as	O
a	O
three-qubit	O
system	O
and	O
experimentally	O
implements	O
the	O
half-adder	O
and	O
subtractor	O
operations	O
.	O
The	O
required	O
qubits	B-KEY
are	O
realized	O
by	O
partially	O
orienting	O
/	O
sup	O
133/Cs	O
nuclei	O
in	O
a	O
liquid-crystalline	B-KEY
medium	I-KEY
,	O
yielding	O
a	O
quadrupolar	B-KEY
split	I-KEY
well-resolved	I-KEY
septet	I-KEY
.	O
Another	O
feature	O
of	O
this	O
paper	O
is	O
the	O
proposal	O
that	O
labeling	O
of	O
quantum	B-KEY
states	I-KEY
of	O
system	O
can	O
be	O
suitably	O
chosen	O
to	O
increase	O
the	O
efficiency	O
of	O
a	O
computational	B-KEY
task	I-KEY

Improving	O
the	O
frequency	B-KEY
stability	I-KEY
of	O
microwave	B-KEY
oscillators	I-KEY
by	O
utilizing	O
the	O
dual-mode	B-KEY
sapphire-loaded	I-KEY
cavity	I-KEY
resonator	I-KEY
The	O
design	O
and	O
experimental	O
testing	O
of	O
a	O
novel	O
control	O
circuit	O
to	O
stabilize	O
the	O
temperature	O
of	O
a	O
sapphire-loaded	O
cavity	O
whispering	B-KEY
gallery	I-KEY
resonator-oscillator	I-KEY
and	O
improve	O
its	O
medium-term	O
frequency	B-KEY
stability	I-KEY
is	O
presented	O
.	O
Finite-element	O
software	O
was	O
used	O
to	O
predict	O
frequencies	O
and	O
quality	O
factors	O
of	O
WGE/sub	O
7,0,0	O
/	O
and	O
the	O
WGH/sub	O
9,0,0	O
/	O
modes	O
near	O
9	B-KEY
GHz	I-KEY
,	O
and	O
separated	O
in	O
frequency	O
by	O
approximately	O
80	O
MHz	O
.	O
Calculations	O
show	O
that	O
the	O
novel	O
temperature	B-KEY
control	I-KEY
circuits	I-KEY
from	O
the	O
difference	B-KEY
frequency	I-KEY
can	O
result	O
in	O
a	O
frequency	B-KEY
stability	I-KEY
of	O
better	O
than	O
one	O
part	O
in	O
10/sup	O
13	O
/	O
at	O
270	O
K.	O
Also	O
,	O
we	O
present	O
details	O
on	O
the	O
best	O
way	O
to	O
couple	O
orthogonally	O
to	O
two	O
modes	O
of	O
similar	O
frequency	O
but	O
different	O
polarization	O

Multiresolution	B-KEY
Markov	I-KEY
models	I-KEY
for	O
signal	O
and	O
image	O
processing	O
Reviews	O
a	O
significant	O
component	O
of	O
the	O
rich	O
field	O
of	O
statistical	O
multiresolution	O
-LRB-	O
MR	O
-RRB-	O
modeling	O
and	O
processing	O
.	O
These	O
MR	O
methods	O
have	O
found	O
application	O
and	O
permeated	O
the	O
literature	O
of	O
a	O
widely	O
scattered	O
set	O
of	O
disciplines	O
,	O
and	O
one	O
of	O
our	O
principal	O
objectives	O
is	O
to	O
present	O
a	O
single	O
,	O
coherent	O
picture	O
of	O
this	O
framework	O
.	O
A	O
second	O
goal	O
is	O
to	O
describe	O
how	O
this	O
topic	O
fits	O
into	O
the	O
even	O
larger	O
field	O
of	O
MR	O
methods	O
and	O
concepts-in	O
particular	O
,	O
making	O
ties	O
to	O
topics	O
such	O
as	O
wavelets	B-KEY
and	O
multigrid	B-KEY
methods	I-KEY
.	O
A	O
third	O
goal	O
is	O
to	O
provide	O
several	O
alternate	O
viewpoints	O
for	O
this	O
body	O
of	O
work	O
,	O
as	O
the	O
methods	O
and	O
concepts	O
we	O
describe	O
intersect	O
with	O
a	O
number	O
of	O
other	O
fields	O
.	O
The	O
principle	O
focus	O
of	O
our	O
presentation	O
is	O
the	O
class	O
of	O
MR	O
Markov	O
processes	O
defined	O
on	O
pyramidally	B-KEY
organized	I-KEY
trees	I-KEY
.	O
The	O
attractiveness	O
of	O
these	O
models	O
stems	O
from	O
both	O
the	O
very	O
efficient	O
algorithms	O
they	O
admit	O
and	O
their	O
expressive	O
power	O
and	O
broad	O
applicability	O
.	O
We	O
show	O
how	O
a	O
variety	O
of	O
methods	O
and	O
models	O
relate	O
to	O
this	O
framework	O
including	O
models	O
for	O
self-similar	O
and	O
1/f	B-KEY
processes	I-KEY
.	O
We	O
also	O
illustrate	O
how	O
these	O
methods	O
have	O
been	O
used	O
in	O
practice	O

Hybrid	O
broadcast	O
for	O
the	O
video-on-demand	B-KEY
service	O
Multicast	B-KEY
offers	O
an	O
efficient	O
means	O
of	O
distributing	O
video	O
contents/programs	O
to	O
multiple	O
clients	O
by	O
batching	O
their	O
requests	O
and	O
then	O
having	O
them	O
share	O
a	O
server	O
's	O
video	O
stream	O
.	O
Batching	O
customers	O
'	O
requests	O
is	O
either	O
client-initiated	O
or	O
server-initiated	O
.	O
Most	O
advanced	O
client-initiated	O
video	O
multicasts	B-KEY
are	O
implemented	O
by	O
patching	O
.	O
Periodic	O
broadcast	O
,	O
a	O
typical	O
server-initiated	O
approach	O
,	O
can	O
be	O
entirety-based	O
or	O
segment-based	O
.	O
This	O
paper	O
focuses	O
on	O
the	O
performance	O
of	O
the	O
VoD	O
service	O
for	O
popular	O
videos	O
.	O
First	O
,	O
we	O
analyze	O
the	O
limitation	O
of	O
conventional	B-KEY
patching	I-KEY
when	O
the	O
customer	B-KEY
request	I-KEY
rate	I-KEY
is	O
high	O
.	O
Then	O
,	O
by	O
combining	O
the	O
advantages	O
of	O
each	O
of	O
the	O
two	O
broadcast	O
schemes	O
,	O
we	O
propose	O
a	O
hybrid	B-KEY
broadcast	I-KEY
scheme	I-KEY
for	O
popular	O
videos	O
,	O
which	O
not	O
only	O
lowers	O
the	O
service	O
latency	O
but	O
also	O
improves	O
clients	O
'	O
interactivity	B-KEY
by	O
using	O
an	O
active	O
buffering	O
technique	O
.	O
This	O
is	O
shown	O
to	O
be	O
a	O
good	O
compromise	O
for	O
both	O
lowering	O
service	O
latency	O
and	O
improving	O
the	O
VCR-like	O
interactivity	B-KEY

Excess	O
energy	O
-LSB-	O
cooling	O
system	O
-RSB-	O
The	O
designers	O
retrofitting	O
a	O
comfort	B-KEY
cooling	I-KEY
system	I-KEY
to	O
offices	O
in	O
Hertfordshire	O
have	O
been	O
able	O
to	O
make	O
use	O
of	O
the	O
waste	B-KEY
heat	I-KEY
rejected	O
.	O
what	O
's	O
more	O
they	O
're	O
now	O
making	O
it	O
a	O
standard	O
solution	O
for	O
much	O
larger	O
projects	O

Parallel	B-KEY
implicit	I-KEY
predictor	I-KEY
corrector	I-KEY
methods	I-KEY
The	O
performance	O
of	O
parallel	O
codes	O
for	O
the	O
solution	O
of	O
initial	B-KEY
value	I-KEY
problems	I-KEY
is	O
usually	O
strongly	O
sensitive	O
to	O
the	O
dimension	O
of	O
the	O
continuous	O
problem	O
.	O
This	O
is	O
due	O
to	O
the	O
overhead	O
related	O
to	O
the	O
exchange	O
of	O
information	O
among	O
the	O
processors	O
and	O
motivates	O
the	O
problem	O
of	O
minimizing	O
the	O
amount	O
of	O
communications	O
.	O
According	O
to	O
this	O
principle	O
,	O
we	O
define	O
the	O
so	O
called	O
Parallel	B-KEY
Implicit	I-KEY
Predictor	I-KEY
Corrector	I-KEY
Methods	I-KEY
and	O
in	O
this	O
class	O
we	O
derive	O
A-stable	O
,	O
L-stable	O
and	O
numerically	B-KEY
zero-stable	I-KEY
formulas	I-KEY
.	O
The	O
latter	O
property	O
refers	O
to	O
the	O
zero-stability	B-KEY
condition	I-KEY
of	O
a	O
given	O
formula	O
when	O
roundoff	B-KEY
errors	I-KEY
are	O
introduced	O
in	O
its	O
coefficients	O
due	O
to	O
their	O
representation	O
in	O
finite	B-KEY
precision	I-KEY
arithmetic	I-KEY
.	O
Some	O
numerical	O
experiment	O
show	O
the	O
potentiality	O
of	O
this	O
approach	O

A	O
framework	O
for	O
image	B-KEY
deblurring	I-KEY
using	O
wavelet	B-KEY
packet	I-KEY
bases	I-KEY
We	O
show	O
that	O
the	O
average	O
over	O
translations	O
of	O
an	O
operator	B-KEY
diagonal	I-KEY
in	O
a	O
wavelet	O
packet	O
basis	O
is	O
a	O
convolution	B-KEY
.	O
We	O
also	O
show	O
that	O
an	O
operator	B-KEY
diagonal	I-KEY
in	O
a	O
wavelet	O
packet	O
basis	O
can	O
be	O
decomposed	O
into	O
several	O
operators	O
of	O
the	O
same	O
kind	O
,	O
each	O
of	O
them	O
being	O
better	O
conditioned	O
.	O
We	O
investigate	O
the	O
possibility	O
of	O
using	O
such	O
a	O
convolution	B-KEY
to	O
approximate	O
a	O
given	O
convolution	B-KEY
-LRB-	O
in	O
practice	O
an	O
image	O
blur	O
-RRB-	O
.	O
Then	O
we	O
use	O
these	O
approximations	O
to	O
deblur	O
images	O
.	O
First	O
,	O
we	O
show	O
that	O
this	O
framework	O
permits	O
us	O
to	O
redefine	O
existing	O
deblurring	O
methods	O
.	O
Then	O
,	O
we	O
show	O
that	O
it	O
permits	O
us	O
to	O
define	O
a	O
new	O
variational	O
method	O
which	O
combines	O
the	O
wavelet	O
packet	O
and	O
the	O
total	B-KEY
variation	I-KEY
approaches	I-KEY
.	O
We	O
argue	O
and	O
show	O
by	O
experiments	O
that	O
this	O
permits	O
us	O
to	O
avoid	O
the	O
drawbacks	O
of	O
both	O
approaches	O
which	O
are	O
,	O
respectively	O
,	O
ringing	B-KEY
and	O
staircasing	B-KEY

Construction	O
of	O
information	B-KEY
retrieval	I-KEY
thesaurus	B-KEY
for	O
family	B-KEY
planning	I-KEY
terms	O
using	O
CDS/ISIS	O
The	O
thesaurus	B-KEY
as	O
a	O
tool	O
for	O
information	B-KEY
retrieval	I-KEY
and	O
as	O
an	O
alternative	O
to	O
the	O
existing	O
scheme	O
of	O
classifications	B-KEY
in	O
information	B-KEY
retrieval	I-KEY
is	O
discussed	O
.	O
The	O
paper	O
considers	O
the	O
emergence	O
of	O
the	O
information	B-KEY
retrieval	I-KEY
thesaurus	B-KEY
and	O
its	O
definition	O
.	O
Family	B-KEY
planning	I-KEY
is	O
a	O
multidisciplinary	O
subject	O
covering	O
socio	O
economic	O
,	O
cultural	B-KEY
,	O
psychological	B-KEY
and	O
medical	O
fields	O
.	O
This	O
necessitated	O
the	O
construction	O
of	O
a	O
thesaurus	B-KEY
for	O
the	O
Family	B-KEY
Planning	I-KEY
discipline	O
.	O
The	O
construction	O
is	O
based	O
on	O
UNISIST	B-KEY
,	O
ISO	B-KEY
2788	I-KEY
and	O
BS	B-KEY
5723	I-KEY
guidelines	O
by	O
using	O
CDS/ISIS	B-KEY
software	I-KEY

Edge-colorings	O
with	O
no	O
large	O
polychromatic	B-KEY
stars	I-KEY
Given	O
a	O
graph	O
G	O
and	O
a	O
positive	B-KEY
integer	I-KEY
r	O
,	O
let	O
f/sub	O
r	O
/	O
-LRB-	O
G	O
-RRB-	O
denote	O
the	O
largest	O
number	O
of	O
colors	O
that	O
can	O
be	O
used	O
in	O
a	O
coloring	O
of	O
E	O
-LRB-	O
G	O
-RRB-	O
such	O
that	O
each	O
vertex	O
is	O
incident	O
to	O
at	O
most	O
r	O
colors	O
.	O
For	O
all	O
positive	B-KEY
integers	I-KEY
n	O
and	O
r	O
,	O
we	O
determine	O
f/sub	O
r	O
/	O
-LRB-	O
K/sub	O
n	O
,	O
n	O
/	O
-RRB-	O
exactly	O
and	O
f/sub	O
r	O
/	O
-LRB-	O
K/sub	O
n	O
/	O
-RRB-	O
within	O
1	O
.	O
In	O
doing	O
so	O
,	O
we	O
disprove	O
a	O
conjecture	O
by	O
Y.	O
Manoussakis	O
et	O
al.	O
-LRB-	O
1996	O
-RRB-	O

Truss	B-KEY
topology	I-KEY
optimization	I-KEY
by	O
a	O
modified	B-KEY
genetic	I-KEY
algorithm	I-KEY
This	O
paper	O
describes	O
the	O
use	O
of	O
a	O
stochastic	B-KEY
search	I-KEY
procedure	I-KEY
based	O
on	O
genetic	O
algorithms	O
for	O
developing	O
near-optimal	B-KEY
topologies	I-KEY
of	O
load-bearing	B-KEY
truss	I-KEY
structures	I-KEY
.	O
Most	O
existing	O
cases	O
these	O
publications	O
express	O
the	O
truss	O
topology	O
as	O
a	O
combination	O
of	O
members	O
.	O
These	O
methods	O
,	O
however	O
,	O
have	O
the	O
disadvantage	O
that	O
the	O
resulting	O
topology	O
may	O
include	O
needless	O
members	O
or	O
those	O
which	O
overlap	O
other	O
members	O
.	O
In	O
addition	O
to	O
these	O
problems	O
,	O
the	O
generated	O
structures	O
are	O
not	O
necessarily	O
structurally	O
stable	O
.	O
A	O
new	O
method	O
,	O
which	O
resolves	O
these	O
problems	O
by	O
expressing	O
the	O
truss	O
topology	O
as	O
a	O
combination	O
of	O
triangles	B-KEY
,	O
is	O
proposed	O
in	O
this	O
paper	O
.	O
Details	O
of	O
the	O
proposed	O
methodology	O
are	O
presented	O
as	O
well	O
as	O
the	O
results	O
of	O
numerical	O
examples	O
that	O
clearly	O
show	O
the	O
effectiveness	O
and	O
efficiency	O
of	O
the	O
method	O

Wavelet-based	O
image	B-KEY
segment	I-KEY
representation	I-KEY
An	O
efficient	O
representation	O
method	O
for	O
arbitrarily	B-KEY
shaped	I-KEY
image	I-KEY
segments	I-KEY
is	O
proposed	O
.	O
This	O
method	O
includes	O
a	O
smart	O
way	O
to	O
select	O
a	O
wavelet	B-KEY
basis	I-KEY
to	O
approximate	O
the	O
given	O
image	O
segment	O
,	O
with	O
improved	B-KEY
image	I-KEY
quality	I-KEY
and	O
reduced	B-KEY
computational	I-KEY
load	I-KEY

A	O
scanline-based	B-KEY
algorithm	I-KEY
for	O
the	O
2D	B-KEY
free-form	I-KEY
bin	I-KEY
packing	I-KEY
problem	I-KEY
This	O
paper	O
describes	O
a	O
heuristic	B-KEY
algorithm	I-KEY
for	O
the	O
2D	O
free-form	O
bin	O
packing	O
-LRB-	O
2D-FBP	O
-RRB-	O
problem	O
.	O
Given	O
a	O
set	O
of	O
2D	O
free-form	O
bins	O
and	O
a	O
set	O
of	O
2D	O
free-form	O
items	O
,	O
the	O
2D-FBP	B-KEY
problem	I-KEY
is	O
to	O
lay	O
out	O
items	O
inside	O
one	O
or	O
more	O
bins	O
in	O
such	O
a	O
way	O
that	O
the	O
number	O
of	O
bins	O
used	O
is	O
minimized	B-KEY
,	O
and	O
for	O
each	O
bin	O
,	O
the	O
yield	O
is	O
maximized	O
.	O
The	O
proposed	O
algorithm	O
handles	O
the	O
problem	O
as	O
a	O
variant	O
of	O
the	O
1D	O
problem	O
;	O
i.e.	O
,	O
items	O
and	O
bins	O
are	O
approximated	O
as	O
sets	O
of	O
scanlines	O
,	O
and	O
scanlines	O
are	O
packed	O
.	O
The	O
details	O
of	O
the	O
algorithm	O
are	O
given	O
,	O
and	O
its	O
application	O
to	O
a	O
nesting	B-KEY
problem	I-KEY
in	O
a	O
shipbuilding	B-KEY
company	I-KEY
is	O
reported	O
.	O
The	O
proposed	O
algorithm	O
consists	O
of	O
the	O
basic	O
and	O
the	O
group	B-KEY
placement	I-KEY
algorithms	I-KEY
.	O
The	O
basic	O
placement	O
algorithm	O
is	O
a	O
variant	O
of	O
the	O
first-fit	B-KEY
decreasing	I-KEY
algorithm	I-KEY
which	O
is	O
simply	O
extended	O
from	O
the	O
1D	O
case	O
to	O
the	O
2D	O
case	O
by	O
a	O
novel	O
scanline	O
approximation	O
.	O
A	O
numerical	O
study	O
with	O
real	O
instances	O
shows	O
that	O
the	O
basic	O
placement	O
algorithm	O
has	O
sufficient	O
performance	O
for	O
most	O
of	O
the	O
instances	O
,	O
however	O
,	O
the	O
group	B-KEY
placement	I-KEY
algorithm	I-KEY
is	O
required	O
when	O
items	O
must	O
be	O
aligned	O
in	O
columns	O
.	O
The	O
qualities	O
of	O
the	O
resulting	O
layouts	O
are	O
good	O
enough	O
for	O
practical	O
use	O
,	O
and	O
the	O
processing	O
times	O
are	O
good	O

A	O
leaf	B-KEY
sequencing	I-KEY
algorithm	I-KEY
to	O
enlarge	O
treatment	B-KEY
field	I-KEY
length	I-KEY
in	O
IMRT	O
With	O
MLC-based	O
IMRT	O
,	O
the	O
maximum	O
usable	O
field	O
size	O
is	O
often	O
smaller	O
than	O
the	O
maximum	O
field	O
size	O
for	O
conventional	O
treatments	O
.	O
This	O
is	O
due	O
to	O
the	O
constraints	O
of	O
the	O
overtravel	B-KEY
distances	I-KEY
of	O
MLC	O
leaves	O
and/or	O
jaws	O
.	O
Using	O
a	O
new	O
leaf	B-KEY
sequencing	I-KEY
algorithm	I-KEY
,	O
the	O
usable	O
IMRT	O
field	O
length	O
-LRB-	O
perpendicular	O
to	O
the	O
MLC	O
motion	O
-RRB-	O
can	O
be	O
mostly	O
made	O
equal	O
to	O
the	O
full	O
length	O
of	O
the	O
MLC	O
field	O
without	O
violating	O
the	O
upper	B-KEY
jaw	I-KEY
overtravel	I-KEY
limit	I-KEY
.	O
For	O
any	O
given	O
intensity	B-KEY
pattern	I-KEY
,	O
a	O
criterion	O
was	O
proposed	O
to	O
assess	O
whether	O
an	O
intensity	B-KEY
pattern	I-KEY
can	O
be	O
delivered	O
without	O
violation	O
of	O
the	O
jaw	B-KEY
position	I-KEY
constraints	I-KEY
.	O
If	O
the	O
criterion	O
is	O
met	O
,	O
the	O
new	O
algorithm	O
will	O
consider	O
the	O
jaw	B-KEY
position	I-KEY
constraints	I-KEY
during	O
the	O
segmentation	O
for	O
the	O
step	B-KEY
and	I-KEY
shoot	I-KEY
delivery	I-KEY
method	I-KEY
.	O
The	O
strategy	O
employed	O
by	O
the	O
algorithm	O
is	O
to	O
connect	O
the	O
intensity	B-KEY
elements	I-KEY
outside	O
the	O
jaw	B-KEY
overtravel	I-KEY
limits	I-KEY
with	O
those	O
inside	O
the	O
jaw	B-KEY
overtravel	I-KEY
limits	I-KEY
.	O
Several	O
methods	O
were	O
used	O
to	O
establish	O
these	O
connections	O
during	O
segmentation	O
by	O
modifying	O
a	O
previously	O
published	O
algorithm	O
-LRB-	O
areal	B-KEY
algorithm	I-KEY
-RRB-	O
,	O
including	O
changing	O
the	O
intensity	O
level	O
,	O
alternating	O
the	O
leaf-sequencing	B-KEY
direction	I-KEY
,	O
or	O
limiting	O
the	O
segment	B-KEY
field	I-KEY
size	I-KEY
.	O
The	O
algorithm	O
was	O
tested	O
with	O
1000	O
random	B-KEY
intensity	I-KEY
patterns	I-KEY
with	O
dimensions	O
of	O
21	O
*	O
27	O
cm/sup	O
2	O
/	O
,	O
800	O
intensity	O
patterns	O
with	O
higher	O
intensity	O
outside	O
the	O
jaw	O
overtravel	O
limit	O
,	O
and	O
three	O
different	O
types	O
of	O
clinical	O
treatment	O
plans	O
that	O
were	O
undeliverable	O
using	O
a	O
segmentation	O
method	O
from	O
a	O
commercial	O
treatment	O
planning	O
system	O
.	O
The	O
new	O
algorithm	O
achieved	O
a	O
success	O
rate	O
of	O
100	O
%	O
with	O
these	O
test	O
patterns	O
.	O
For	O
the	O
1000	O
random	B-KEY
patterns	I-KEY
,	O
the	O
new	O
algorithm	O
yields	O
a	O
similar	O
average	O
number	O
of	O
segments	O
of	O
36.9	O
+	O
or-2	O
.9	O
in	O
comparison	O
to	O
36.6	O
+	O
or-1	O
.3	O
when	O
using	O
the	O
areal	B-KEY
algorithm	I-KEY
.	O
For	O
the	O
800	O
patterns	O
with	O
higher	O
intensities	O
outside	O
the	O
jaw	B-KEY
overtravel	I-KEY
limits	I-KEY
,	O
the	O
new	O
algorithm	O
results	O
in	O
an	O
increase	O
of	O
25	O
%	O
in	O
the	O
average	O
number	O
of	O
segments	O
compared	O
to	O
the	O
areal	B-KEY
algorithm	I-KEY
.	O
However	O
,	O
the	O
areal	B-KEY
algorithm	I-KEY
fails	O
to	O
create	O
deliverable	B-KEY
segments	I-KEY
for	O
90	O
%	O
of	O
these	O
patterns	O
.	O
Using	O
a	O
single	B-KEY
isocenter	I-KEY
,	O
the	O
new	O
algorithm	O
provides	O
a	O
solution	O
to	O
extend	O
the	O
usable	O
IMRT	O
field	O
length	O
from	O
21	O
to	O
27	O
cm	O
for	O
IMRT	O
on	O
a	O
commercial	B-KEY
linear	I-KEY
accelerator	I-KEY
using	O
the	O
step	B-KEY
and	I-KEY
shoot	I-KEY
delivery	I-KEY
method	I-KEY

Application	O
of	O
ultrasonic	O
sensors	O
in	O
the	O
process	B-KEY
industry	I-KEY
Continuous	B-KEY
process	I-KEY
monitoring	I-KEY
in	O
gaseous	O
,	O
liquid	O
or	O
molten	O
media	O
is	O
a	O
fundamental	O
requirement	O
for	O
process	B-KEY
control	I-KEY
.	O
Besides	O
temperature	O
and	O
pressure	O
other	O
process	O
parameters	O
such	O
as	O
level	O
,	O
flow	O
,	O
concentration	O
and	O
conversion	O
are	O
of	O
special	O
interest	O
.	O
More	O
qualified	O
information	O
obtained	O
from	O
new	O
or	O
better	O
sensors	O
can	O
significantly	O
enhance	O
the	O
process	B-KEY
quality	I-KEY
and	O
thereby	O
product	O
properties	O
.	O
Ultrasonic	O
sensors	O
or	O
sensor	O
systems	O
can	O
contribute	O
to	O
this	O
development	O
.	O
The	O
state	O
of	O
the	O
art	O
of	O
ultrasonic	O
sensors	O
and	O
their	O
advantages	O
and	O
disadvantages	O
will	O
be	O
discussed	O
.	O
Commercial	O
examples	O
will	O
be	O
presented	O
.	O
Among	O
others	O
,	O
applications	O
in	O
the	O
food	O
,	O
chemical	O
and	O
pharmaceutical	B-KEY
industries	I-KEY
are	O
described	O
.	O
Possibilities	O
and	O
limitations	O
of	O
ultrasonic	O
process	O
sensors	O
are	O
discussed	O

Robust	B-KEY
self-tuning	I-KEY
PID	I-KEY
controller	I-KEY
for	O
nonlinear	B-KEY
systems	I-KEY
In	O
this	O
paper	O
,	O
we	O
propose	O
a	O
robust	B-KEY
self-tuning	I-KEY
PID	I-KEY
controller	I-KEY
suitable	O
for	O
nonlinear	B-KEY
systems	I-KEY
.	O
The	O
control	O
system	O
employs	O
a	O
preload	B-KEY
relay	I-KEY
-LRB-	O
P_Relay	O
-RRB-	O
in	O
series	O
with	O
a	O
PID	O
controller	O
.	O
The	O
P_Relay	O
ensures	O
a	O
high	O
gain	O
to	O
yield	O
a	O
robust	B-KEY
performance	I-KEY
.	O
However	O
,	O
it	O
also	O
incurs	O
a	O
chattering	B-KEY
phenomenon	I-KEY
.	O
In	O
this	O
paper	O
,	O
instead	O
of	O
viewing	O
the	O
chattering	O
as	O
an	O
undesirable	O
yet	O
inevitable	O
feature	O
,	O
we	O
use	O
it	O
as	O
a	O
naturally	B-KEY
occurring	I-KEY
signal	I-KEY
for	O
tuning	O
and	O
re-tuning	O
the	O
PID	O
controller	O
as	O
the	O
operating	B-KEY
regime	I-KEY
digresses	O
.	O
No	O
other	O
explicit	O
input	O
signal	O
is	O
required	O
.	O
Once	O
the	O
PID	O
controller	O
is	O
tuned	O
for	O
a	O
particular	O
operating	O
point	O
,	O
the	O
relay	O
may	O
be	O
disabled	O
and	O
chattering	O
ceases	O
correspondingly	O
.	O
However	O
,	O
it	O
is	O
invoked	O
when	O
there	O
is	O
a	O
change	O
in	O
setpoint	O
to	O
another	O
operating	B-KEY
regime	I-KEY
.	O
In	O
this	O
way	O
,	O
the	O
approach	O
is	O
also	O
applicable	O
to	O
time-varying	B-KEY
systems	I-KEY
as	O
the	O
PID	O
tuning	O
can	O
be	O
continuous	O
,	O
based	O
on	O
the	O
latest	O
set	O
of	O
chattering	O
characteristics	O
.	O
Analysis	O
is	O
provided	O
on	O
the	O
stability	B-KEY
properties	I-KEY
of	O
the	O
control	O
scheme	O
.	O
Simulation	B-KEY
results	I-KEY
for	O
the	O
level	O
control	O
of	O
fluid	O
in	O
a	O
spherical	B-KEY
tank	I-KEY
using	O
the	O
scheme	O
are	O
also	O
presented	O

Experimental	O
investigation	O
of	O
active	B-KEY
vibration	I-KEY
control	I-KEY
using	O
neural	O
networks	O
and	O
piezoelectric	O
actuators	O
The	O
use	O
of	O
neural	B-KEY
networks	I-KEY
for	O
identification	B-KEY
and	O
control	B-KEY
of	O
smart	B-KEY
structures	I-KEY
is	O
investigated	O
experimentally	O
.	O
Piezoelectric	B-KEY
actuators	I-KEY
are	O
employed	O
to	O
suppress	O
the	O
vibrations	O
of	O
a	O
cantilevered	B-KEY
plate	I-KEY
subject	O
to	O
impulse	O
,	O
sine	O
wave	O
and	O
band-limited	O
white	B-KEY
noise	I-KEY
disturbances	I-KEY
.	O
The	O
neural	B-KEY
networks	I-KEY
used	O
are	O
multilayer	B-KEY
perceptrons	I-KEY
trained	O
with	O
error	B-KEY
backpropagation	I-KEY
.	O
Validation	O
studies	O
show	O
that	O
the	O
identifier	O
predicts	O
the	O
system	O
dynamics	O
accurately	O
.	O
The	O
controller	B-KEY
is	O
trained	O
adaptively	O
with	O
the	O
help	O
of	O
the	O
neural	O
identifier	O
.	O
Experimental	O
results	O
demonstrate	O
excellent	O
closed-loop	B-KEY
performance	I-KEY
and	O
robustness	B-KEY
of	O
the	O
neurocontroller	B-KEY

An	O
augmented	B-KEY
spatial	I-KEY
digital	I-KEY
tree	I-KEY
algorithm	I-KEY
for	O
contact	B-KEY
detection	I-KEY
in	O
computational	B-KEY
mechanics	I-KEY
Based	O
on	O
the	O
understanding	O
of	O
existing	O
spatial	O
digital	O
tree-based	O
contact	B-KEY
detection	I-KEY
approaches	O
,	O
and	O
the	O
alternating	O
digital	O
tree	O
-LRB-	O
ADT	O
-RRB-	O
algorithm	O
in	O
particular	O
,	O
a	O
more	O
efficient	O
algorithm	O
,	O
termed	O
the	O
augmented	O
spatial	O
digital	O
tree	O
-LRB-	O
ASDT	O
-RRB-	O
algorithm	O
,	O
is	O
proposed	O
in	O
the	O
present	O
work	O
.	O
The	O
ASDT	O
algorithm	O
adopts	O
a	O
different	O
point	O
representation	O
scheme	O
that	O
uses	O
only	O
the	O
lower	O
comer	O
vertex	O
to	O
represent	O
a	O
-LRB-	O
hyper	O
-	O
-RRB-	O
rectangle	O
,	O
with	O
the	O
upper	B-KEY
comer	I-KEY
vertex	I-KEY
serving	O
as	O
the	O
augmented	O
information	O
.	O
Consequently	O
,	O
the	O
ASDT	O
algorithm	O
can	O
keep	O
the	O
working	O
space	O
the	O
same	O
as	O
the	O
original	O
n-dimensional	O
space	O
and	O
,	O
in	O
general	O
,	O
a	O
much	O
better	O
balanced	O
tree	O
can	O
be	O
expected	O
.	O
This	O
,	O
together	O
with	O
the	O
introduction	O
of	O
an	O
additional	O
bounding	O
subregion	O
for	O
the	O
rectangles	O
associated	O
with	O
each	O
tree	O
node	O
,	O
makes	O
it	O
possible	O
to	O
significantly	O
reduce	O
the	O
number	O
of	O
node	O
visits	O
in	O
the	O
region	O
search	O
,	O
although	O
each	O
node	O
visit	O
may	O
be	O
slightly	O
more	O
expensive	O
.	O
Three	O
examples	O
arising	O
in	O
computational	B-KEY
mechanics	I-KEY
are	O
presented	O
to	O
provide	O
an	O
assessment	O
of	O
the	O
performance	O
of	O
the	O
ASDT	O
.	O
The	O
numerical	O
results	O
indicate	O
that	O
the	O
ASDT	O
is	O
,	O
at	O
least	O
,	O
over	O
3.9	O
times	O
faster	O
than	O
the	O
ADT	O

An	O
efficient	O
DIPIE	B-KEY
algorithm	I-KEY
for	O
CAD	O
of	O
electrostatically	B-KEY
actuated	I-KEY
MEMS	O
devices	O
Pull-in	B-KEY
parameters	I-KEY
are	O
important	O
properties	O
of	O
electrostatic	B-KEY
actuators	I-KEY
.	O
Efficient	O
and	O
accurate	O
analysis	O
tools	O
that	O
can	O
capture	O
these	O
parameters	O
for	O
different	O
design	B-KEY
geometries	I-KEY
,	O
are	O
therefore	O
essential	O
.	O
Current	O
simulation	O
tools	O
approach	O
the	O
pull-in	O
state	O
by	O
iteratively	O
adjusting	O
the	O
voltage	O
applied	O
across	O
the	O
actuator	O
electrodes	O
.	O
The	O
convergence	B-KEY
rate	I-KEY
of	O
this	O
scheme	O
gradually	O
deteriorates	O
as	O
the	O
pull-in	O
state	O
is	O
approached	O
.	O
Moreover	O
,	O
the	O
convergence	O
is	O
inconsistent	O
and	O
requires	O
many	O
mesh	O
and	O
accuracy	O
refinements	O
to	O
assure	O
reliable	O
predictions	O
.	O
As	O
a	O
result	O
,	O
the	O
design	O
procedure	O
of	O
electrostatically	B-KEY
actuated	I-KEY
MEMS	O
devices	O
can	O
be	O
time-consuming	O
.	O
In	O
this	O
paper	O
a	O
novel	O
Displacement	B-KEY
Iteration	I-KEY
Pull-In	O
Extraction	O
-LRB-	O
DIPIE	O
-RRB-	O
scheme	O
is	O
presented	O
.	O
The	O
DIPIE	O
scheme	O
is	O
shown	O
to	O
converge	O
consistently	O
and	O
far	O
more	O
rapidly	O
than	O
the	O
Voltage	O
Iterations	O
-LRB-	O
VI	O
-RRB-	O
scheme	O
-LRB-	O
>	O
100	O
times	O
faster	O
!	O
-RRB-	O
.	O
The	O
DIPIE	O
scheme	O
requires	O
separate	O
mechanical	O
and	O
electrostatic	B-KEY
field	I-KEY
solvers	I-KEY
.	O
Therefore	O
,	O
it	O
can	O
be	O
easily	O
implemented	O
in	O
existing	O
MOEMS	B-KEY
CAD	I-KEY
packages	I-KEY
.	O
Moreover	O
,	O
using	O
the	O
DIPIE	O
scheme	O
,	O
the	O
pull-in	B-KEY
parameters	I-KEY
extraction	O
can	O
be	O
performed	O
in	O
a	O
fully	O
automated	O
mode	O
,	O
and	O
no	O
user	O
input	O
for	O
search	O
bounds	O
is	O
required	O

Design	O
and	O
implementation	O
of	O
a	O
reusable	O
and	O
extensible	O
HL7	O
encoding/decoding	O
framework	O
The	O
Health	B-KEY
Level	I-KEY
Seven	I-KEY
-LRB-	O
HL7	O
-RRB-	O
,	O
an	O
international	B-KEY
standard	I-KEY
for	O
electronic	B-KEY
data	I-KEY
exchange	I-KEY
in	O
all	O
health	B-KEY
care	I-KEY
environments	I-KEY
,	O
enables	O
disparate	O
computer	O
applications	O
to	O
exchange	O
key	O
sets	O
of	O
clinical	O
and	O
administrative	B-KEY
information	I-KEY
.	O
Above	O
all	O
,	O
it	O
defines	O
the	O
standard	O
HL7	B-KEY
message	I-KEY
formats	I-KEY
prescribed	O
by	O
the	O
standard	O
encoding	O
rules	O
.	O
In	O
this	O
paper	O
,	O
we	O
propose	O
a	O
flexible	O
,	O
reusable	O
,	O
and	O
extensible	O
HL7	O
encoding	O
and	O
decoding	O
framework	O
using	O
a	O
message	B-KEY
object	I-KEY
model	I-KEY
-LRB-	O
MOM	B-KEY
-RRB-	O
and	O
message	B-KEY
definition	I-KEY
repository	I-KEY
-LRB-	O
MDR	B-KEY
-RRB-	O
.	O
The	O
MOM	B-KEY
provides	O
an	O
abstract	O
HL7	O
message	O
form	O
represented	O
by	O
a	O
group	O
of	O
objects	O
and	O
their	O
relationships	O
.	O
It	O
reflects	O
logical	B-KEY
relationships	I-KEY
among	O
the	O
standard	O
HL7	O
message	O
elements	O
such	O
as	O
segments	O
,	O
fields	O
,	O
and	O
components	O
,	O
while	O
enforcing	O
the	O
key	O
structural	B-KEY
constraints	I-KEY
imposed	O
by	O
the	O
standard	O
.	O
Since	O
the	O
MOM	B-KEY
completely	O
eliminates	O
the	O
dependency	O
of	O
the	O
HL7	O
encoder	O
and	O
decoder	O
on	O
platform-specific	O
data	O
formats	O
,	O
it	O
makes	O
it	O
possible	O
to	O
build	O
the	O
encoder	O
and	O
decoder	O
as	O
reusable	O
standalone	B-KEY
software	I-KEY
components	I-KEY
,	O
enabling	O
the	O
interconnection	O
of	O
arbitrary	O
heterogeneous	B-KEY
hospital	I-KEY
information	I-KEY
systems	I-KEY
-LRB-	O
HIS	B-KEY
-RRB-	O
with	O
little	O
effort	O
.	O
Moreover	O
,	O
the	O
MDR	B-KEY
,	O
an	O
external	B-KEY
database	I-KEY
of	O
key	B-KEY
definitions	I-KEY
for	O
HL7	O
messages	O
,	O
helps	O
make	O
the	O
encoder	O
and	O
decoder	O
as	O
resilient	O
as	O
possible	O
to	O
future	O
modifications	O
of	O
the	O
standard	O
HL7	B-KEY
message	I-KEY
formats	I-KEY
.	O
It	O
is	O
also	O
used	O
by	O
the	O
encoder	O
and	O
decoder	O
to	O
perform	O
a	O
well-formedness	O
check	O
for	O
their	O
respective	O
inputs	O
-LRB-	O
i.e.	O
,	O
HL7	O
message	O
objects	O
expressed	O
in	O
the	O
MOM	B-KEY
and	O
encoded	O
HL7	O
message	O
strings	O
-RRB-	O
.	O
Although	O
we	O
implemented	O
a	O
prototype	O
version	O
of	O
the	O
encoder	O
and	O
decoder	O
using	O
JAVA	B-KEY
,	O
they	O
can	O
be	O
easily	O
packaged	O
and	O
delivered	O
as	O
standalone	O
components	O
using	O
the	O
standard	O
component	O
frameworks	O

Hidden	O
Markov	O
model-based	O
tool	O
wear	O
monitoring	O
in	O
turning	O
This	O
paper	O
presents	O
a	O
new	O
modeling	O
framework	O
for	O
tool	B-KEY
wear	I-KEY
monitoring	I-KEY
in	O
machining	B-KEY
processes	I-KEY
using	O
hidden	B-KEY
Markov	I-KEY
models	I-KEY
-LRB-	O
HMMs	O
-RRB-	O
.	O
Feature	O
vectors	O
are	O
extracted	O
from	O
vibration	B-KEY
signals	I-KEY
measured	O
during	O
turning	O
.	O
A	O
codebook	B-KEY
is	O
designed	O
and	O
used	O
for	O
vector	B-KEY
quantization	I-KEY
to	O
convert	O
the	O
feature	O
vectors	O
into	O
a	O
symbol	O
sequence	O
for	O
the	O
hidden	B-KEY
Markov	I-KEY
model	I-KEY
.	O
A	O
series	O
of	O
experiments	O
are	O
conducted	O
to	O
evaluate	O
the	O
effectiveness	O
of	O
the	O
approach	O
for	O
different	O
lengths	O
of	O
training	O
data	O
and	O
observation	O
sequence	O
.	O
Experimental	O
results	O
show	O
that	O
successful	O
tool	B-KEY
state	I-KEY
detection	I-KEY
rates	O
as	O
high	O
as	O
97	O
%	O
can	O
be	O
achieved	O
by	O
using	O
this	O
approach	O

Commerce	O
Department	O
plan	O
eases	O
3G	B-KEY
spectrum	I-KEY
crunch	O
The	O
federal	B-KEY
government	I-KEY
made	O
its	O
first	O
move	O
last	O
week	O
toward	O
cleaning	O
up	O
a	O
spectrum	B-KEY
allocation	I-KEY
system	I-KEY
that	O
was	O
in	O
shambles	O
just	O
a	O
year	O
ago	O
and	O
had	O
some	O
,	O
spectrum-starved	O
wireless	B-KEY
carriers	I-KEY
fearing	O
they	O
would	O
n't	O
be	O
able	O
to	O
compete	O
in	O
third-generation	O
services	O
.	O
The	O
move	O
,	O
however	O
,	O
is	O
far	O
from	O
complete	O
and	O
leaves	O
numerous	O
details	O
unsettled	O

Homogenization	B-KEY
in	O
L/sup	O
infinity	O
/	O
Homogenization	B-KEY
of	O
deterministic	B-KEY
control	I-KEY
problems	O
with	O
L/sup	B-KEY
infinity	I-KEY
/	I-KEY
running	I-KEY
cost	I-KEY
is	O
studied	O
by	O
viscosity	O
solutions	O
techniques	O
.	O
It	O
is	O
proved	O
that	O
the	O
value	B-KEY
function	I-KEY
of	O
an	O
L/sup	O
infinity	O
/	O
problem	O
in	O
a	O
medium	O
with	O
a	O
periodic	O
micro-structure	O
converges	B-KEY
uniformly	O
on	O
the	O
compact	O
sets	O
to	O
the	O
value	B-KEY
function	I-KEY
of	O
the	O
homogenized	B-KEY
problem	O
as	O
the	O
period	O
shrinks	O
to	O
0	O
.	O
Our	O
main	O
convergence	B-KEY
result	O
extends	O
that	O
of	O
Ishii	O
-LRB-	O
Stochastic	O
Analysis	O
,	O
control	O
,	O
optimization	O
and	O
applications	O
,	O
pp.	O
305-324	O
,	O
Birkhauser	O
Boston	O
,	O
Boston	O
,	O
MA	O
,	O
1999	O
.	O
-RRB-	O
to	O
the	O
case	O
of	O
a	O
discontinuous	O
Hamiltonian	O
.	O
The	O
cell	B-KEY
problem	I-KEY
is	O
solved	O
,	O
but	O
,	O
as	O
nonuniqueness	O
occurs	O
,	O
the	O
effective	O
Hamiltonian	O
must	O
be	O
selected	O
in	O
a	O
careful	O
way	O
.	O
The	O
paper	O
also	O
provides	O
a	O
representation	O
formula	O
for	O
the	O
effective	O
Hamiltonian	O
and	O
gives	O
illustrations	O
to	O
calculus	B-KEY
of	I-KEY
variations	I-KEY
,	O
averaging	B-KEY
and	O
one-dimensional	O
problems	O

Evaluating	O
the	O
performance	O
of	O
a	O
distributed	O
database	O
of	O
repetitive	B-KEY
elements	I-KEY
in	O
complete	B-KEY
genomes	I-KEY
The	O
original	O
version	O
of	O
the	O
Repeat	O
Sequence	O
Database	O
-LRB-	O
RSDB	O
-RRB-	O
was	O
created	O
based	O
on	O
centralized	O
database	O
systems	O
-LRB-	O
CDBSs	O
-RRB-	O
.	O
RSDB	O
presently	O
includes	O
an	O
enormous	O
amount	O
of	O
data	O
,	O
with	O
the	O
amount	O
of	O
biological	B-KEY
data	I-KEY
increasing	O
rapidly	O
.	O
Distributed	O
RSDB	O
-LRB-	O
DRSDB	O
-RRB-	O
is	O
developed	O
to	O
yield	O
better	O
performance	O
.	O
This	O
study	O
proposed	O
many	O
approaches	O
to	O
data	B-KEY
distribution	I-KEY
and	O
experimentally	O
determines	O
the	O
best	O
approach	O
to	O
obtain	O
good	O
performance	O
of	O
our	O
database	O
.	O
Experimental	O
results	O
indicate	O
that	O
DRSDB	O
performs	O
well	O
for	O
particular	O
types	O
of	O
query	B-KEY

Dynamics	B-KEY
and	O
control	B-KEY
of	O
initialized	B-KEY
fractional-order	I-KEY
systems	I-KEY
Due	O
to	O
the	O
importance	O
of	O
historical	O
effects	O
in	O
fractional-order	O
systems	O
,	O
this	O
paper	O
presents	O
a	O
general	O
fractional-order	O
system	O
and	O
control	B-KEY
theory	O
that	O
includes	O
the	O
time-varying	O
initialization	B-KEY
response	I-KEY
.	O
Previous	O
studies	O
have	O
not	O
properly	O
accounted	O
for	O
these	O
historical	O
effects	O
.	O
The	O
initialization	B-KEY
response	I-KEY
,	O
along	O
with	O
the	O
forced	B-KEY
response	I-KEY
,	O
for	O
fractional-order	O
systems	O
is	O
determined	O
.	O
The	O
scalar	O
fractional-order	O
impulse	B-KEY
response	I-KEY
is	O
determined	O
,	O
and	O
is	O
a	O
generalization	O
of	O
the	O
exponential	B-KEY
function	I-KEY
.	O
Stability	O
properties	O
of	O
fractional-order	O
systems	O
are	O
presented	O
in	O
the	O
complex	O
w-plane	O
,	O
which	O
is	O
a	O
transformation	O
of	O
the	O
s-plane	O
.	O
Time	O
responses	O
are	O
discussed	O
with	O
respect	O
to	O
pole	O
positions	O
in	O
the	O
complex	O
w-plane	O
and	O
frequency	O
response	O
behavior	O
is	O
included	O
.	O
A	O
fractional-order	O
vector	B-KEY
space	I-KEY
representation	I-KEY
,	O
which	O
is	O
a	O
generalization	O
of	O
the	O
state	B-KEY
space	I-KEY
concept	I-KEY
,	O
is	O
presented	O
including	O
the	O
initialization	B-KEY
response	I-KEY
.	O
Control	B-KEY
methods	O
for	O
vector	O
representations	O
of	O
initialized	B-KEY
fractional-order	I-KEY
systems	I-KEY
are	O
shown	O
.	O
Finally	O
,	O
the	O
fractional-order	B-KEY
differintegral	I-KEY
is	O
generalized	O
to	O
continuous	O
order-distributions	O
which	O
have	O
the	O
possibility	O
of	O
including	O
all	O
fractional	O
orders	O
in	O
a	O
transfer	B-KEY
function	I-KEY

An	O
accurate	O
COG	O
defuzzifier	O
design	O
using	O
Lamarckian	O
co-adaptation	O
of	O
learning	B-KEY
and	O
evolution	B-KEY
This	O
paper	O
proposes	O
a	O
design	O
technique	O
of	O
optimal	O
center	O
of	O
gravity	O
-LRB-	O
COG	O
-RRB-	O
defuzzifier	O
using	O
the	O
Lamarckian	O
co-adaptation	O
of	O
learning	B-KEY
and	O
evolution	B-KEY
.	O
The	O
proposed	O
COG	O
defuzzifier	O
is	O
specified	O
by	O
various	O
design	O
parameters	O
such	O
as	O
the	O
centers	O
,	O
widths	O
,	O
and	O
modifiers	O
of	O
MFs	O
.	O
The	O
design	O
parameters	O
are	O
adjusted	O
with	O
the	O
Lamarckian	O
co-adaptation	O
of	O
learning	B-KEY
and	O
evolution	B-KEY
,	O
where	O
the	O
learning	B-KEY
performs	O
a	O
local	B-KEY
search	I-KEY
of	O
design	O
parameters	O
in	O
an	O
individual	O
COG	O
defuzzifier	O
,	O
but	O
the	O
evolution	B-KEY
performs	O
a	O
global	O
search	O
of	O
design	O
parameters	O
among	O
a	O
population	O
of	O
various	O
COG	O
defuzzifiers	O
.	O
This	O
co-adaptation	O
scheme	O
allows	O
to	O
evolve	O
much	O
faster	O
than	O
the	O
non-learning	O
case	O
and	O
gives	O
a	O
higher	O
possibility	O
of	O
finding	O
an	O
optimal	O
solution	O
due	O
to	O
its	O
wider	O
searching	O
capability	O
.	O
An	O
application	O
to	O
the	O
truck	O
backer-upper	O
control	O
problem	O
of	O
the	O
proposed	O
co-adaptive	O
design	O
method	O
of	O
COG	O
defuzzifier	O
is	O
presented	O
.	O
The	O
approximation	O
ability	O
and	O
control	O
performance	O
are	O
compared	O
with	O
those	O
of	O
the	O
conventionally	O
simplified	O
COG	O
defuzzifier	O
in	O
terms	O
of	O
the	O
fuzzy	B-KEY
logic	I-KEY
controller	I-KEY
's	O
approximation	O
error	O
and	O
the	O
average	O
tracing	O
distance	O
,	O
respectively	O

Craigslist	B-KEY
:	O
virtual	B-KEY
community	I-KEY
maintains	O
human	O
touch	O
If	O
it	O
works	O
why	O
change	O
it	O
?	O
This	O
might	O
have	O
been	O
the	O
thought	O
on	O
the	O
minds	O
of	O
dot	O
com	O
executives	O
back	O
when	O
Internet	B-KEY
businesses	I-KEY
were	O
booming	O
,	O
and	O
most	O
of	O
the	O
Web	B-KEY
content	I-KEY
was	O
free	O
.	O
Web	O
sites	O
were	O
overflowing	O
with	O
advertisements	O
of	O
every	O
kind	O
and	O
size	O
.	O
Now	O
that	O
dot	O
com	O
principals	O
know	O
better	O
,	O
Web	O
ads	O
are	O
no	O
longer	O
the	O
only	O
path	O
to	O
revenue	B-KEY
generation	I-KEY
.	O
Community	B-KEY
portals	I-KEY
,	O
however	O
,	O
never	O
seemed	O
to	O
have	O
many	O
ads	O
to	O
begin	O
with	O
,	O
and	O
their	O
content	O
stayed	O
truer	O
to	O
who	O
they	O
served	O
.	O
Many	O
of	O
them	O
started	O
off	O
as	O
simple	O
places	O
for	O
users	O
to	O
list	O
announcements	B-KEY
,	O
local	B-KEY
events	I-KEY
,	O
want	B-KEY
ads	I-KEY
,	O
real	B-KEY
estate	I-KEY
,	O
and	O
mingle	O
with	O
other	O
local	O
users	O
.	O
The	O
author	O
saw	O
the	O
need	O
for	O
San	O
Franciscans	O
to	O
have	O
a	O
place	O
to	O
do	O
all	O
of	O
that	O
for	O
free	O
,	O
without	O
any	O
annoying	O
advertising	O
,	O
and	O
ended	O
up	O
offering	O
much	O
more	O
to	O
his	O
community	O
with	O
the	O
creation	O
of	O
craigslist	B-KEY
.	O
''	O
-LSB-	O
Polling	O
users	O
-RSB-	O
was	O
a	O
good	O
way	O
for	O
us	O
to	O
connect	O
with	O
our	O
members	O
,	O
this	O
is	O
the	O
way	O
to	O
operate	O
successfully	O
in	O
situations	O
like	O
these	O
-	O
your	O
members	O
come	O
first	O
.	O
''	O

Development	O
of	O
an	O
Internet-based	B-KEY
intelligent	I-KEY
design	I-KEY
support	I-KEY
system	I-KEY
for	O
rolling	B-KEY
element	I-KEY
bearings	I-KEY
This	O
paper	O
presents	O
a	O
novel	O
approach	O
to	O
developing	O
an	O
intelligent	B-KEY
agile	I-KEY
design	I-KEY
system	I-KEY
for	O
rolling	O
bearings	O
based	O
on	O
artificial	B-KEY
intelligence	I-KEY
-LRB-	O
AI	O
-RRB-	O
,	O
Internet	O
and	O
Web	B-KEY
technologies	I-KEY
and	O
expertise	O
.	O
The	O
underlying	O
philosophy	O
of	O
the	O
approach	O
is	O
to	O
use	O
AI	O
technology	O
and	O
Web-based	O
design	O
support	O
systems	O
as	O
smart	B-KEY
tools	I-KEY
from	O
which	O
design	O
customers	O
can	O
rapidly	O
and	O
responsively	O
access	O
the	O
systems	O
'	O
built-in	O
design	O
expertise	O
.	O
The	O
approach	O
is	O
described	O
in	O
detail	O
with	O
a	O
novel	O
AI	O
model	O
and	O
system	O
implementation	O
issues	O
.	O
The	O
major	O
issues	O
in	O
implementing	O
the	O
approach	O
are	O
discussed	O
with	O
particular	O
reference	O
to	O
using	O
AI	O
technologies	O
,	O
network	B-KEY
programming	I-KEY
,	O
client-server	B-KEY
technology	I-KEY
and	O
open	O
computing	O
of	O
bearing	B-KEY
design	I-KEY
and	O
manufacturing	B-KEY
requirements	I-KEY

Process	O
pioneers	O
-LSB-	O
agile	B-KEY
business	I-KEY
-RSB-	O
By	O
managing	B-KEY
IT	I-KEY
infrastructures	I-KEY
along	O
so-called	O
`	O
top	O
down	O
'	O
lines	O
,	O
organisations	O
can	O
streamline	O
their	O
business	B-KEY
processes	I-KEY
,	O
eliminate	O
redundant	O
tasks	O
and	O
increase	B-KEY
automation	I-KEY

Computer	B-KEY
program	I-KEY
for	O
calculating	O
the	O
p-value	O
in	O
testing	B-KEY
process	I-KEY
capability	I-KEY
index	I-KEY
C/sub	O
pmk	O
/	O
Many	O
process	O
capability	O
indices	O
,	O
including	O
C/sub	O
p	O
/	O
,	O
C/sub	O
pk	O
/	O
,	O
and	O
C/sub	O
pm	O
/	O
,	O
have	O
been	O
proposed	O
to	O
provide	O
numerical	O
measures	O
on	O
the	O
process	B-KEY
potential	I-KEY
and	O
performance	O
.	O
Combining	O
the	O
advantages	O
of	O
these	O
indices	O
,	O
Pearn	O
et	O
al.	O
-LRB-	O
1992	O
-RRB-	O
introduced	O
a	O
new	O
capability	O
index	O
called	O
C/sub	O
pmk	O
/	O
,	O
which	O
has	O
been	O
shown	O
to	O
be	O
a	O
useful	O
capability	O
index	O
for	O
processes	O
with	O
two-sided	O
specification	O
limits	O
.	O
In	O
this	O
paper	O
,	O
the	O
authors	O
implement	O
the	O
theory	O
of	O
a	O
testing	B-KEY
hypothesis	I-KEY
using	O
the	O
natural	B-KEY
estimator	I-KEY
of	O
C/sub	O
pmk	O
/	O
,	O
and	O
provide	O
an	O
efficient	O
Maple	B-KEY
computer	B-KEY
program	I-KEY
to	O
calculate	O
the	O
p-values	O
.	O
They	O
also	O
provide	O
tables	O
of	O
the	O
critical	O
values	O
for	O
some	O
commonly	O
used	O
capability	O
requirements	O
.	O
Based	O
on	O
the	O
test	O
,	O
they	O
develop	O
a	O
simple	O
step-by-step	O
procedure	O
for	O
in-plant	B-KEY
applications	I-KEY
.	O
The	O
practitioners	O
can	O
use	O
the	O
proposed	O
procedure	O
to	O
determine	O
whether	O
their	O
process	O
meets	O
the	O
preset	B-KEY
capability	I-KEY
requirement	I-KEY
,	O
and	O
make	O
reliable	B-KEY
decisions	I-KEY

A	O
new	O
merging	B-KEY
algorithm	I-KEY
for	O
constructing	O
suffix	B-KEY
trees	I-KEY
for	O
integer	B-KEY
alphabets	I-KEY
A	O
new	O
approach	O
for	O
constructing	O
a	O
suffix	B-KEY
tree	I-KEY
T/sub	O
s	O
/	O
for	O
a	O
given	O
string	O
S	O
is	O
to	O
construct	O
recursively	O
a	O
suffix	B-KEY
tree	I-KEY
T/sub	O
o	O
/	O
for	O
odd	O
positions	O
,	O
construct	O
a	O
suffix	O
,	O
tree	O
T/sub	O
e	O
/	O
for	O
even	O
positions	O
from	O
T/sub	O
o	O
/	O
and	O
then	O
merge	O
T/sub	O
o	O
/	O
and	O
T/sub	O
e	O
/	O
into	O
T/sub	O
s	O
/	O
.	O
To	O
construct	O
suffix	B-KEY
trees	I-KEY
for	O
integer	B-KEY
alphabets	I-KEY
in	O
linear	B-KEY
time	I-KEY
had	O
been	O
a	O
major	O
open	O
problem	O
on	O
index	B-KEY
data	I-KEY
structures	I-KEY
.	O
Farach	O
used	O
this	O
approach	O
and	O
gave	O
the	O
first	O
linear-time	O
algorithm	O
for	O
integer	B-KEY
alphabets	I-KEY
.	O
The	O
hardest	O
part	O
of	O
Farach	O
's	O
algorithm	O
is	O
the	O
merging	O
step	O
.	O
In	O
this	O
paper	O
we	O
present	O
a	O
new	O
and	O
simpler	O
merging	B-KEY
algorithm	I-KEY
based	O
on	O
a	O
coupled	B-KEY
BFS	I-KEY
-LRB-	O
breadth-first	B-KEY
search	I-KEY
-RRB-	O
.	O
Our	O
merging	B-KEY
algorithm	I-KEY
is	O
more	O
intuitive	O
than	O
Farach	O
's	O
coupled	O
DFS	O
-LRB-	O
depth-first	O
search	O
-RRB-	O
merging	O
,	O
and	O
thus	O
it	O
can	O
be	O
easily	O
extended	O
to	O
other	O
applications	O

An	O
efficient	O
and	O
stable	O
ray	O
tracing	O
algorithm	O
for	O
parametric	B-KEY
surfaces	I-KEY
In	O
this	O
paper	O
,	O
we	O
propose	O
an	O
efficient	O
and	O
stable	O
algorithm	O
for	O
finding	O
ray-surface	B-KEY
intersections	I-KEY
.	O
Newton	O
's	O
method	O
and	O
Bezier	B-KEY
clipping	I-KEY
are	O
adapted	O
to	O
form	O
the	O
core	O
of	O
our	O
algorithm	O
.	O
Ray	B-KEY
coherence	I-KEY
is	O
used	O
to	O
find	O
starting	O
points	O
for	O
Newton	B-KEY
iteration	I-KEY
.	O
We	O
introduce	O
an	O
obstruction	B-KEY
detection	I-KEY
technique	I-KEY
to	O
verify	O
whether	O
an	O
intersection	O
point	O
found	O
using	O
Newton	O
's	O
method	O
is	O
the	O
closest	O
.	O
When	O
Newton	O
's	O
method	O
fails	O
to	O
achieve	O
convergence	B-KEY
,	O
we	O
use	O
Bezier	B-KEY
clipping	I-KEY
substitution	O
to	O
find	O
the	O
intersection	O
points	O
.	O
This	O
combination	O
achieves	O
a	O
significant	O
improvement	O
in	O
tracing	O
primary	O
rays	O
.	O
A	O
similar	O
approach	O
successfully	O
improves	O
the	O
performance	O
of	O
tracing	O
secondary	O
rays	O

Reinventing	O
broadband	B-KEY
Many	O
believe	O
that	O
broadband	B-KEY
providers	O
need	O
to	O
change	O
their	O
whole	O
approach	O
.	O
The	O
future	O
,	O
then	O
,	O
is	O
in	O
reinventing	O
broadband	B-KEY
.	O
That	O
means	O
tiered	B-KEY
pricing	I-KEY
to	O
make	O
broadband	B-KEY
more	O
competitive	O
with	O
dial-up	O
access	O
and	O
livelier	O
,	O
more	O
distinct	O
content	O
:	O
video	B-KEY
on	I-KEY
demand	I-KEY
,	O
MP3	B-KEY
,	O
and	O
other	O
features	O
exclusive	O
to	O
the	O
fat-pipe	O
superhighway	O

Control	O
of	O
transient	O
thermal	O
response	O
during	O
sequential	O
open-die	B-KEY
forging	I-KEY
:	O
a	O
trajectory	B-KEY
optimization	I-KEY
approach	O
A	O
trajectory	B-KEY
optimization	I-KEY
approach	O
is	O
applied	O
to	O
the	O
design	O
of	O
a	O
sequence	O
of	O
open-die	B-KEY
forging	I-KEY
operations	O
in	O
order	O
to	O
control	O
the	O
transient	O
thermal	O
response	O
of	O
a	O
large	O
titanium	B-KEY
alloy	I-KEY
billet	I-KEY
.	O
The	O
amount	O
of	O
time	O
the	O
billet	O
is	O
soaked	O
in	O
furnace	O
prior	O
to	O
each	O
successive	O
forging	O
operation	O
is	O
optimized	O
to	O
minimize	O
the	O
total	O
process	O
time	O
while	O
simultaneously	O
satisfying	O
constraints	O
on	O
the	O
maximum	O
and	O
minimum	O
values	O
of	O
the	O
billet	O
temperature	B-KEY
distribution	I-KEY
to	O
avoid	O
microstructural	B-KEY
defects	I-KEY
during	O
forging	O
.	O
The	O
results	O
indicate	O
that	O
a	O
``	O
differential	O
''	O
heating	B-KEY
profile	I-KEY
is	O
the	O
most	O
effective	O
at	O
meeting	O
these	O
design	O
goals	O

A	O
new	O
algebraic	B-KEY
modelling	I-KEY
approach	I-KEY
to	O
distributed	B-KEY
problem-solving	I-KEY
in	O
MAS	O
This	O
paper	O
is	O
devoted	O
to	O
a	O
new	O
algebraic	B-KEY
modelling	I-KEY
approach	I-KEY
to	O
distributed	B-KEY
problem-solving	I-KEY
in	O
multi-agent	B-KEY
systems	I-KEY
-LRB-	O
MAS	O
-RRB-	O
,	O
which	O
is	O
featured	O
by	O
a	O
unified	B-KEY
framework	I-KEY
for	O
describing	O
and	O
treating	O
social	B-KEY
behaviors	I-KEY
,	O
social	B-KEY
dynamics	I-KEY
and	O
social	B-KEY
intelligence	I-KEY
.	O
A	O
conceptual	O
architecture	O
of	O
algebraic	O
modelling	O
is	O
presented	O
.	O
The	O
algebraic	O
modelling	O
of	O
typical	O
social	B-KEY
behaviors	I-KEY
,	O
social	O
situation	O
and	O
social	B-KEY
dynamics	I-KEY
is	O
discussed	O
in	O
the	O
context	O
of	O
distributed	B-KEY
problem-solving	I-KEY
in	O
MAS	O
.	O
The	O
comparison	O
and	O
simulation	O
on	O
distributed	B-KEY
task	I-KEY
allocations	I-KEY
and	O
resource	B-KEY
assignments	I-KEY
in	O
MAS	O
show	O
more	O
advantages	O
of	O
the	O
algebraic	O
approach	O
than	O
other	O
conventional	O
methods	O

Formula-dependent	B-KEY
equivalence	I-KEY
for	O
compositional	O
CTL	B-KEY
model	I-KEY
checking	I-KEY
We	O
present	O
a	O
polytime	B-KEY
computable	I-KEY
state	I-KEY
equivalence	I-KEY
that	O
is	O
defined	O
with	O
respect	O
to	O
a	O
given	O
CTL	B-KEY
formula	I-KEY
.	O
Since	O
it	O
does	O
not	O
attempt	O
to	O
preserve	O
all	O
CTL	B-KEY
formulas	I-KEY
,	O
like	O
bisimulation	O
does	O
,	O
we	O
can	O
expect	O
to	O
compute	O
coarser	O
equivalences	O
.	O
This	O
equivalence	O
can	O
be	O
used	O
to	O
reduce	O
the	O
complexity	O
of	O
model	O
checking	O
a	O
system	O
of	O
interacting	B-KEY
FSM	I-KEY
.	O
Additionally	O
,	O
we	O
show	O
that	O
in	O
some	O
cases	O
our	O
techniques	O
can	O
detect	O
if	O
a	O
formula	O
passes	O
or	O
fails	O
,	O
without	O
forming	O
the	O
entire	O
product	O
machine	O
.	O
The	O
method	O
is	O
exact	O
and	O
fully	O
automatic	O
,	O
and	O
handles	O
full	O
CTL	O

Faking	O
it	O
:	O
simulating	O
dependent	B-KEY
types	I-KEY
in	O
Haskell	B-KEY
Dependent	B-KEY
types	I-KEY
reflect	O
the	O
fact	O
that	O
validity	O
of	O
data	O
is	O
often	O
a	O
relative	O
notion	O
by	O
allowing	O
prior	O
data	O
to	O
affect	O
the	O
types	O
of	O
subsequent	O
data	O
.	O
Not	O
only	O
does	O
this	O
make	O
for	O
a	O
precise	B-KEY
type	I-KEY
system	I-KEY
,	O
but	O
also	O
a	O
highly	O
generic	O
one	O
:	O
both	O
the	O
type	O
and	O
the	O
program	O
for	O
each	O
instance	O
of	O
a	O
family	O
of	O
operations	O
can	O
be	O
computed	O
from	O
the	O
data	O
which	O
codes	O
for	O
that	O
instance	O
.	O
Recent	O
experimental	O
extensions	O
to	O
the	O
Haskell	B-KEY
type	B-KEY
class	I-KEY
mechanism	I-KEY
give	O
us	O
strong	O
tools	O
to	O
relativize	O
types	O
to	O
other	O
types	O
.	O
We	O
may	O
simulate	O
some	O
aspects	O
of	O
dependent	B-KEY
typing	I-KEY
by	O
making	O
counterfeit	B-KEY
type-level	I-KEY
copies	I-KEY
of	O
data	O
,	O
with	O
type	B-KEY
constructors	I-KEY
simulating	O
data	B-KEY
constructors	I-KEY
and	O
type	O
classes	O
simulating	O
datatypes	B-KEY
.	O
This	O
paper	O
gives	O
examples	O
of	O
the	O
technique	O
and	O
discusses	O
its	O
potential	O

OMS	O
battle	O
heating	O
up	O
as	O
Chicago	O
Equity	O
ousts	O
LongView	B-KEY
for	O
Macgregor	B-KEY
Chicago	B-KEY
Equity	I-KEY
Partners	I-KEY
LLC	O
has	O
gone	O
into	O
full	O
production	O
with	O
Macgregor	B-KEY
's	O
Financial	B-KEY
Trading	I-KEY
Platform	I-KEY
.	O
This	O
marks	O
a	O
concentrated	O
effort	O
to	O
achieve	O
straight-through	B-KEY
processing	I-KEY

On	O
the	O
relationship	O
between	O
parametric	B-KEY
variation	I-KEY
and	O
state	O
feedback	O
in	O
chaos	B-KEY
control	I-KEY
In	O
this	O
Letter	O
,	O
we	O
study	O
the	O
popular	O
parametric	B-KEY
variation	I-KEY
chaos	B-KEY
control	I-KEY
and	O
state-feedback	B-KEY
methodologies	O
in	O
chaos	B-KEY
control	I-KEY
,	O
and	O
point	O
out	O
for	O
the	O
first	O
time	O
that	O
they	O
are	O
actually	O
equivalent	O
in	O
the	O
sense	O
that	O
there	O
exist	O
diffeomorphisms	B-KEY
that	O
can	O
convert	O
one	O
to	O
the	O
other	O
for	O
most	O
smooth	O
chaotic	O
systems	O
.	O
Detailed	O
conversions	O
are	O
worked	O
out	O
for	O
typical	O
discrete	O
chaotic	O
maps	O
-LRB-	O
logistic	B-KEY
,	O
Henon	O
-RRB-	O
and	O
continuous	B-KEY
flows	I-KEY
-LRB-	O
Rossler	O
,	O
Lorenz	O
-RRB-	O
for	O
illustration	O
.	O
This	O
unifies	O
the	O
two	O
seemingly	O
different	O
approaches	O
from	O
the	O
physics	O
and	O
the	O
engineering	O
communities	O
on	O
chaos	B-KEY
control	I-KEY
.	O
This	O
new	O
perspective	O
reveals	O
some	O
new	O
potential	O
applications	O
such	O
as	O
chaos	O
synchronization	O
and	O
normal	O
form	O
analysis	O
from	O
a	O
unified	O
mathematical	O
point	O
of	O
view	O

On	O
fractal	B-KEY
dimension	I-KEY
in	O
information	B-KEY
systems	I-KEY
.	O
Toward	O
exact	B-KEY
sets	I-KEY
in	O
infinite	B-KEY
information	I-KEY
systems	I-KEY
The	O
notions	O
of	O
an	O
exact	O
as	O
well	O
as	O
a	O
rough	B-KEY
set	I-KEY
are	O
well-grounded	O
as	O
basic	O
notions	O
in	O
rough	B-KEY
set	I-KEY
theory	O
.	O
They	O
are	O
however	O
defined	O
in	O
the	O
setting	O
of	O
a	O
finite	O
information	B-KEY
system	I-KEY
i.e.	O
an	O
information	B-KEY
system	I-KEY
having	O
finite	O
numbers	O
of	O
objects	O
as	O
well	O
as	O
attributes	O
.	O
In	O
theoretical	O
studies	O
e.g.	O
of	O
topological	B-KEY
properties	I-KEY
of	O
rough	B-KEY
sets	I-KEY
,	O
one	O
has	O
to	O
trespass	O
this	O
limitation	O
and	O
to	O
consider	O
information	B-KEY
systems	I-KEY
with	O
potentially	O
unbound	O
number	O
of	O
attributes	O
.	O
In	O
such	O
setting	O
,	O
the	O
notions	O
of	O
rough	O
and	O
exact	B-KEY
sets	I-KEY
may	O
be	O
defined	O
in	O
terms	O
of	O
topological	O
operators	O
of	O
interior	O
and	O
closure	O
with	O
respect	O
to	O
an	O
appropriate	O
topology	O
following	O
the	O
ideas	O
from	O
the	O
finite	O
case	O
,	O
where	O
it	O
is	O
noticed	O
that	O
in	O
the	O
finite	O
case	O
rough-set-theoretic	O
operators	O
of	O
lower	O
and	O
upper	O
approximation	O
are	O
identical	O
with	O
the	O
interior	O
,	O
respectively	O
,	O
closure	B-KEY
operators	I-KEY
in	O
topology	O
induced	O
by	O
equivalence	B-KEY
classes	I-KEY
of	O
the	O
indiscernibility	O
relation	O
.	O
Extensions	O
of	O
finite	O
information	B-KEY
systems	I-KEY
are	O
also	O
desirable	O
from	O
application	O
point	O
of	O
view	O
in	O
the	O
area	O
of	O
knowledge	B-KEY
discovery	I-KEY
and	O
data	B-KEY
mining	I-KEY
,	O
when	O
demands	O
of	O
e.g.	O
mass	O
collaboration	O
and/or	O
huge	O
experimental	O
data	O
call	O
for	O
need	O
of	O
working	O
with	O
large	O
data	O
tables	O
so	O
the	O
sound	O
theoretical	O
generalization	O
of	O
these	O
cases	O
is	O
an	O
information	B-KEY
system	I-KEY
with	O
the	O
number	O
of	O
attributes	O
not	O
bound	O
in	O
advance	O
by	O
a	O
fixed	O
integer	O
i.e.	O
an	O
information	B-KEY
system	I-KEY
with	O
countably	O
but	O
infinitely	O
many	O
attributes	O
,	O
In	O
large	O
information	B-KEY
systems	I-KEY
,	O
a	O
need	O
arises	O
for	O
qualitative	B-KEY
measures	I-KEY
of	O
complexity	B-KEY
of	O
concepts	O
involved	O
free	O
of	O
parameters	O
,	O
cf.	O
e.g.	O
applications	O
for	O
the	O
Vapnik-Czervonenkis	O
dimension	O
.	O
We	O
study	O
here	O
in	O
the	O
theoretical	O
setting	O
of	O
infinite	B-KEY
information	I-KEY
system	I-KEY
a	O
proposal	O
to	O
apply	O
fractal	O
dimensions	O
suitably	O
modified	O
as	O
measures	O
of	O
concept	O
complexity	O

A	O
three-source	B-KEY
model	I-KEY
for	O
the	O
calculation	O
of	O
head	B-KEY
scatter	I-KEY
factors	I-KEY
Accurate	O
determination	O
of	O
the	O
head	B-KEY
scatter	I-KEY
factor	I-KEY
S/sub	O
c	O
/	O
is	O
an	O
important	O
issue	O
,	O
especially	O
for	O
intensity	B-KEY
modulated	I-KEY
radiation	I-KEY
therapy	I-KEY
,	O
where	O
the	O
segmented	B-KEY
fields	I-KEY
are	O
often	O
very	O
irregular	O
and	O
much	O
less	O
than	O
the	O
collimator	O
jaw	O
settings	O
.	O
In	O
this	O
work	O
,	O
we	O
report	O
an	O
S/sub	O
c	O
/	O
calculation	B-KEY
algorithm	I-KEY
for	O
symmetric	B-KEY
,	O
asymmetric	B-KEY
,	O
and	O
irregular	B-KEY
open	I-KEY
fields	I-KEY
shaped	O
by	O
the	O
tertiary	O
collimator	O
-LRB-	O
a	O
multileaf	O
collimator	O
or	O
blocks	O
-RRB-	O
at	O
different	O
source-to-chamber	O
distance	O
.	O
The	O
algorithm	O
was	O
based	O
on	O
a	O
three-source	B-KEY
model	I-KEY
,	O
in	O
which	O
the	O
photon	B-KEY
radiation	I-KEY
to	O
the	O
point	O
of	O
calculation	O
was	O
treated	O
as	O
if	O
it	O
originated	O
from	O
three	O
effective	O
sources	O
:	O
one	O
source	O
for	O
the	O
primary	O
photons	O
from	O
the	O
target	B-KEY
and	O
two	O
extra-focal	B-KEY
photon	I-KEY
sources	I-KEY
for	O
the	O
scattered	B-KEY
photons	I-KEY
from	O
the	O
primary	B-KEY
collimator	I-KEY
and	O
the	O
flattening	B-KEY
filter	I-KEY
,	O
respectively	O
.	O
The	O
field	B-KEY
mapping	O
method	O
proposed	O
by	O
Kim	O
et	O
al.	O
-LSB-	O
Phys	O
.	O
Med	O
.	O
Biol	O
.	O
43	O
,	O
1593-1604	O
-LRB-	O
1998	O
-RRB-	O
-RSB-	O
was	O
extended	O
to	O
two	O
extra-focal	B-KEY
source	I-KEY
planes	I-KEY
and	O
the	O
scatter	O
contributions	O
were	O
integrated	O
over	O
the	O
projected	O
areas	O
-LRB-	O
determined	O
by	O
the	O
detector	O
's	O
eye	O
view	O
-RRB-	O
in	O
the	O
three	O
source	O
planes	O
considering	O
the	O
source	B-KEY
intensity	I-KEY
distributions	I-KEY
.	O
The	O
algorithm	O
was	O
implemented	O
using	O
Microsoft	O
Visual	O
C/C	O
+	O
+	O
in	O
the	O
MS	B-KEY
Windows	I-KEY
environment	I-KEY
.	O
The	O
only	O
input	B-KEY
data	I-KEY
required	O
were	O
head	B-KEY
scatter	I-KEY
factors	I-KEY
for	O
symmetric	B-KEY
square	O
fields	O
,	O
which	O
are	O
normally	O
acquired	O
during	O
machine	O
commissioning	O
.	O
A	O
large	O
number	O
of	O
different	O
fields	B-KEY
were	O
used	O
to	O
evaluate	O
the	O
algorithm	O
and	O
the	O
results	O
were	O
compared	O
with	O
measurements	O
.	O
We	O
found	O
that	O
most	O
of	O
the	O
calculated	O
S/sub	O
c	O
/	O
's	O
agreed	O
with	O
the	O
measured	O
values	O
to	O
within	O
0.4	O
%	O
.	O
The	O
algorithm	O
can	O
also	O
be	O
easily	O
applied	O
to	O
deal	O
with	O
irregular	O
fields	B-KEY
shaped	O
by	O
a	O
multileaf	B-KEY
collimator	I-KEY
that	O
replaces	O
the	O
upper	O
or	O
lower	B-KEY
collimator	I-KEY
jaws	I-KEY

Women	B-KEY
of	I-KEY
color	I-KEY
in	O
computing	O
It	O
is	O
well	O
known	O
that	O
there	O
is	O
a	O
need	O
to	O
increase	O
the	O
number	O
of	O
women	O
in	O
the	O
area	O
of	O
computing	O
,	O
that	O
is	O
in	O
computer	B-KEY
science	I-KEY
and	O
computer	B-KEY
engineering	I-KEY
.	O
If	O
we	O
consider	O
women	B-KEY
of	I-KEY
color	I-KEY
,	O
that	O
is	O
women	O
of	O
under-represented	O
ethnicities	O
,	O
we	O
find	O
the	O
numbers	O
are	O
very	O
dismal	O
.	O
The	O
goal	O
of	O
this	O
article	O
is	O
to	O
bring	O
to	O
light	O
the	O
unique	O
issues	O
of	O
women	B-KEY
of	I-KEY
color	I-KEY
based	O
upon	O
the	O
personal	O
experience	O
of	O
one	O
African-American	O
woman	O
who	O
has	O
been	O
in	O
the	O
field	O
of	O
computing	O
for	O
over	O
20	O
years	O
-LRB-	O
including	O
the	O
years	O
of	O
higher	B-KEY
education	I-KEY
-RRB-	O

NARX-based	O
technique	O
for	O
the	O
modelling	B-KEY
of	O
magneto-rheological	O
damping	O
devices	O
This	O
paper	O
presents	O
a	O
methodology	O
for	O
identifying	O
variable-structure	O
nonlinear	O
models	B-KEY
of	O
magneto-rheological	O
dampers	O
-LRB-	O
MRD	O
-RRB-	O
and	O
similar	O
devices	O
.	O
Its	O
peculiarity	O
with	O
respect	O
to	O
the	O
mainstream	O
literature	O
is	O
to	O
be	O
especially	O
conceived	O
for	O
obtaining	O
models	B-KEY
that	O
are	O
structurally	O
simple	O
,	O
easy	O
to	O
estimate	O
and	O
well	O
suited	O
for	O
model-based	O
control	O
.	O
This	O
goal	O
is	O
pursued	O
by	O
adopting	O
linear-in-the-parameters	O
NARX	B-KEY
models	I-KEY
,	O
for	O
which	O
an	O
identification	O
method	O
is	O
developed	O
based	O
on	O
the	O
minimization	O
of	O
the	O
simulation	O
error	O
.	O
This	O
method	O
is	O
capable	O
of	O
selecting	O
the	O
model	B-KEY
structure	O
together	O
with	O
the	O
parameters	O
,	O
thus	O
it	O
does	O
not	O
require	O
a	O
priori	O
structural	O
information	O
.	O
A	O
set	O
of	O
validation	B-KEY
tests	O
is	O
reported	O
,	O
with	O
the	O
aim	O
of	O
demonstrating	O
the	O
technique	O
's	O
efficiency	O
by	O
comparing	O
it	O
to	O
a	O
widely	O
accepted	O
MRD	B-KEY
modelling	I-KEY
approach	O

Robust	B-KEY
model-order	I-KEY
reduction	I-KEY
of	O
complex	B-KEY
biological	I-KEY
processes	I-KEY
This	O
paper	O
addresses	O
robust	B-KEY
model-order	I-KEY
reduction	I-KEY
of	O
a	O
high	O
dimensional	O
nonlinear	O
partial	O
differential	O
equation	O
-LRB-	O
PDE	O
-RRB-	O
model	O
of	O
a	O
complex	B-KEY
biological	I-KEY
process	I-KEY
.	O
Based	O
on	O
a	O
nonlinear	O
,	O
distributed	O
parameter	O
model	O
of	O
the	O
same	O
process	O
which	O
was	O
validated	O
against	O
experimental	O
data	O
of	O
an	O
existing	O
,	O
pilot-scale	O
biological	O
nutrient	O
removal	O
-LRB-	O
BNR	O
-RRB-	O
activated	O
sludge	O
plant	O
,	O
we	O
developed	O
a	O
state-space	B-KEY
model	I-KEY
with	O
154	O
state	O
variables	O
.	O
A	O
general	O
algorithm	O
for	O
robustly	O
reducing	O
the	O
nonlinear	O
PDE	O
model	O
is	O
presented	O
and	O
,	O
based	O
on	O
an	O
investigation	O
of	O
five	O
state-of-the-art	O
model-order	O
reduction	O
techniques	O
,	O
we	O
are	O
able	O
to	O
reduce	O
the	O
original	O
model	O
to	O
a	O
model	O
with	O
only	O
30	O
states	O
without	O
incurring	O
pronounced	O
modelling	B-KEY
errors	I-KEY
.	O
The	O
singular	B-KEY
perturbation	I-KEY
approximation	I-KEY
balanced	I-KEY
truncating	I-KEY
technique	I-KEY
is	O
found	O
to	O
give	O
the	O
lowest	O
modelling	B-KEY
errors	I-KEY
in	O
low	O
frequency	O
ranges	O
and	O
hence	O
is	O
deemed	O
most	O
suitable	O
for	O
controller	B-KEY
design	I-KEY
and	O
other	O
real-time	O
applications	O

Look	O
who	O
's	O
talking	O
-LSB-	O
voice	B-KEY
recognition	I-KEY
-RSB-	O
Voice	B-KEY
recognition	I-KEY
could	O
be	O
the	O
answer	O
to	O
the	O
problem	O
of	O
financial	B-KEY
fraud	I-KEY
,	O
but	O
in	O
the	O
world	O
of	O
biometric	B-KEY
technology	O
,	O
money	O
talks	O

Portal	O
payback	O
The	O
benefits	O
of	O
deploying	O
a	O
corporate	B-KEY
portal	I-KEY
are	O
well-documented	O
:	O
access	O
to	O
applications	O
and	O
content	O
is	O
centralised	O
,	O
so	O
users	O
do	O
not	O
spend	O
hours	O
searching	O
for	O
information	O
;	O
the	O
management	O
of	O
disparate	O
applications	O
is	O
also	O
centralised	O
,	O
and	O
by	O
allowing	O
users	O
to	O
access	O
`	O
self-service	O
'	O
applications	O
in	O
areas	O
such	O
as	O
human	O
resources	O
and	O
procurement	O
,	O
organisations	O
spend	O
less	O
time	O
on	O
manual	O
processing	O
tasks	O
.	O
But	O
how	O
far	O
can	O
prospective	O
customers	O
rely	O
on	O
the	O
ROI	O
figures	O
presented	O
to	O
them	O
by	O
portal	O
technology	O
vendors	O
?	O
In	O
particular	O
,	O
how	O
reliable	O
are	O
the	O
`	O
ROI	B-KEY
calculators	I-KEY
'	O
these	O
vendors	O
supply	O
on	O
their	O
web	B-KEY
sites	I-KEY
?	O

Hybrid	B-KEY
decision	I-KEY
tree	I-KEY
In	O
this	O
paper	O
,	O
a	O
hybrid	B-KEY
learning	I-KEY
approach	I-KEY
named	O
hybrid	B-KEY
decision	I-KEY
tree	I-KEY
-LRB-	O
HDT	O
-RRB-	O
is	O
proposed	O
.	O
HDT	O
simulates	O
human	O
reasoning	B-KEY
by	O
using	O
symbolic	B-KEY
learning	I-KEY
to	O
do	O
qualitative	B-KEY
analysis	I-KEY
and	O
using	O
neural	B-KEY
learning	I-KEY
to	O
do	O
subsequent	O
quantitative	B-KEY
analysis	I-KEY
.	O
It	O
generates	O
the	O
trunk	O
of	O
a	O
binary	O
HDT	O
according	O
to	O
the	O
binary	B-KEY
information	I-KEY
gain	I-KEY
ratio	I-KEY
criterion	I-KEY
in	O
an	O
instance	O
space	O
defined	O
by	O
only	O
original	O
unordered	O
attributes	O
.	O
If	O
unordered	O
attributes	O
can	O
not	O
further	O
distinguish	O
training	O
examples	O
falling	O
into	O
a	O
leaf	O
node	O
whose	O
diversity	O
is	O
beyond	O
the	O
diversity-threshold	O
,	O
then	O
the	O
node	O
is	O
marked	O
as	O
a	O
dummy	O
node	O
.	O
After	O
all	O
those	O
dummy	O
nodes	O
are	O
marked	O
,	O
a	O
specific	O
feedforward	B-KEY
neural	I-KEY
network	I-KEY
named	O
FANNC	B-KEY
that	O
is	O
trained	O
in	O
an	O
instance	O
space	O
defined	O
by	O
only	O
original	O
ordered	O
attributes	O
is	O
exploited	O
to	O
accomplish	O
the	O
learning	O
task	O
.	O
Moreover	O
,	O
this	O
paper	O
distinguishes	O
three	O
kinds	O
of	O
incremental	B-KEY
learning	I-KEY
tasks	O
.	O
Two	O
incremental	B-KEY
learning	I-KEY
procedures	O
designed	O
for	O
example-incremental	O
learning	O
with	O
different	O
storage	O
requirements	O
are	O
provided	O
,	O
which	O
enables	O
HDT	O
to	O
deal	O
gracefully	O
with	O
data	O
sets	O
where	O
new	O
data	O
are	O
frequently	O
appended	O
.	O
Also	O
a	O
hypothesis-driven	B-KEY
constructive	I-KEY
induction	I-KEY
mechanism	O
is	O
provided	O
,	O
which	O
enables	O
HDT	O
to	O
generate	O
compact	O
concept	O
descriptions	O

Aggregate	B-KEY
bandwidth	I-KEY
estimation	I-KEY
in	O
stored	B-KEY
video	I-KEY
distribution	I-KEY
systems	I-KEY
Multimedia	B-KEY
applications	I-KEY
like	O
video	B-KEY
on	I-KEY
demand	I-KEY
,	O
distance	B-KEY
learning	I-KEY
,	O
Internet	B-KEY
video	I-KEY
broadcast	I-KEY
,	O
etc.	O
will	O
play	O
a	O
fundamental	O
role	O
in	O
future	O
broadband	B-KEY
networks	I-KEY
.	O
A	O
common	O
aspect	O
of	O
such	O
applications	O
is	O
the	O
transmission	O
of	O
video	O
streams	O
that	O
require	O
a	O
sustained	O
relatively	O
high	O
bandwidth	O
with	O
stringent	O
requirements	O
of	O
quality	B-KEY
of	I-KEY
service	I-KEY
.	O
In	O
this	O
paper	O
various	O
original	O
algorithms	O
for	O
evaluating	O
,	O
in	O
a	O
video	O
distribution	O
system	O
,	O
a	O
statistical	B-KEY
estimation	I-KEY
of	O
aggregate	O
bandwidth	O
needed	O
by	O
a	O
given	O
number	O
of	O
smoothed	O
video	O
streams	O
are	O
proposed	O
and	O
discussed	O
.	O
The	O
variable	B-KEY
bit	I-KEY
rate	I-KEY
traffic	I-KEY
generated	O
by	O
each	O
video	O
stream	O
is	O
characterized	O
by	O
its	O
marginal	B-KEY
distribution	I-KEY
and	O
by	O
conditional	B-KEY
probabilities	I-KEY
between	O
rates	O
of	O
temporary	B-KEY
closed	I-KEY
streams	I-KEY
.	O
The	O
developed	O
iterative	B-KEY
algorithms	I-KEY
evaluate	O
an	O
upper	O
and	O
lower	B-KEY
bound	I-KEY
of	O
needed	O
bandwidth	O
for	O
guaranteeing	O
a	O
given	O
loss	B-KEY
probability	I-KEY
.	O
The	O
obtained	O
results	O
are	O
compared	O
with	O
simulations	B-KEY
and	O
with	O
other	O
results	O
,	O
based	O
on	O
similar	O
assumptions	O
,	O
already	O
presented	O
in	O
the	O
literature	O
.	O
Some	O
considerations	O
on	O
the	O
developed	O
algorithms	O
are	O
made	O
,	O
in	O
order	O
to	O
evaluate	O
the	O
effectiveness	O
of	O
the	O
proposed	O
methods	O

Design	O
and	O
analysis	O
of	O
optimal	B-KEY
material	I-KEY
distribution	I-KEY
policies	I-KEY
in	O
flexible	B-KEY
manufacturing	I-KEY
systems	I-KEY
using	O
a	O
single	O
AGV	B-KEY
Modern	O
automated	O
manufacturing	O
processes	O
employ	O
automated	B-KEY
guided	I-KEY
vehicles	I-KEY
-LRB-	O
AGVs	B-KEY
-RRB-	O
for	O
material	B-KEY
handling	I-KEY
,	O
which	O
serve	O
several	O
machine	B-KEY
centres	I-KEY
-LRB-	O
MC	O
-RRB-	O
in	O
a	O
factory	O
.	O
Optimal	B-KEY
scheduling	I-KEY
of	O
AGVs	B-KEY
can	O
significantly	O
help	O
to	O
increase	O
the	O
efficiency	O
of	O
the	O
manufacturing	O
process	O
by	O
minimizing	O
the	O
idle	O
time	O
of	O
MCs	O
waiting	O
for	O
the	O
raw	O
materials	O
.	O
We	O
analyse	O
the	O
requirements	O
for	O
an	O
optimal	B-KEY
schedule	I-KEY
and	O
then	O
provide	O
a	O
mathematical	O
framework	O
for	O
an	O
efficient	O
schedule	O
of	O
material	B-KEY
delivery	I-KEY
by	O
an	O
AGV	B-KEY
.	O
A	O
mathematical	O
model	O
is	O
developed	O
and	O
then	O
a	O
strategy	O
for	O
optimal	O
material	O
distribution	O
of	O
the	O
available	O
raw	O
material	O
to	O
the	O
MCs	O
is	O
derived	O
.	O
With	O
this	O
model	O
,	O
the	O
optimal	O
number	O
of	O
MCs	O
to	O
be	O
utilized	O
is	O
also	O
determined	O
.	O
Finally	O
,	O
the	O
material	B-KEY
delivery	I-KEY
schedule	O
employing	O
multiple	O
journeys	O
to	O
the	O
MCs	O
by	O
the	O
AGV	B-KEY
is	O
carried	O
out	O
.	O
Through	O
rigorous	O
analysis	O
and	O
simulation	O
experiments	O
,	O
we	O
show	O
that	O
such	O
a	O
delivery	O
strategy	O
will	O
optimize	O
the	O
overall	O
performance	O

Approximating	O
martingales	B-KEY
for	O
variance	B-KEY
reduction	I-KEY
in	O
Markov	B-KEY
process	I-KEY
simulation	I-KEY
``	O
Knowledge	O
of	O
either	O
analytical	O
or	O
numerical	O
approximations	O
should	O
enable	O
more	O
efficient	O
simulation	O
estimators	O
to	O
be	O
constructed	O
.	O
''	O
This	O
principle	O
seems	O
intuitively	O
plausible	O
and	O
certainly	O
attractive	O
,	O
yet	O
no	O
completely	O
satisfactory	O
general	O
methodology	O
has	O
been	O
developed	O
to	O
exploit	O
it	O
.	O
The	O
authors	O
present	O
a	O
new	O
approach	O
for	O
obtaining	O
variance	B-KEY
reduction	I-KEY
in	O
Markov	B-KEY
process	I-KEY
simulation	I-KEY
that	O
is	O
applicable	O
to	O
a	O
vast	O
array	O
of	O
different	O
performance	B-KEY
measures	I-KEY
.	O
The	O
approach	O
relies	O
on	O
the	O
construction	O
of	O
a	O
martingale	B-KEY
that	O
is	O
then	O
used	O
as	O
an	O
internal	B-KEY
control	I-KEY
variate	I-KEY

Ride	B-KEY
quality	I-KEY
evaluation	I-KEY
of	O
an	O
actively-controlled	B-KEY
stretcher	I-KEY
for	O
an	O
ambulance	B-KEY
This	O
study	O
considers	O
the	O
subjective	B-KEY
evaluation	I-KEY
of	O
ride	O
quality	O
during	O
ambulance	B-KEY
transportation	O
using	O
an	O
actively-controlled	O
stretcher	O
-LRB-	O
ACS	O
-RRB-	O
.	O
The	O
ride	O
quality	O
of	O
a	O
conventional	B-KEY
stretcher	I-KEY
and	O
an	O
assistant	O
driver	O
's	O
seat	O
is	O
also	O
compared	O
.	O
Braking	B-KEY
during	O
ambulance	B-KEY
transportation	O
generates	O
negative	O
foot-to-head	O
acceleration	O
in	O
patients	O
and	O
causes	O
blood	O
pressure	O
to	O
rise	O
in	O
the	O
patient	O
's	O
head	O
.	O
The	O
ACS	O
absorbs	O
the	O
foot-to-head	O
acceleration	O
by	O
changing	O
the	O
angle	O
of	O
the	O
stretcher	O
,	O
thus	O
reducing	O
the	O
blood	B-KEY
pressure	I-KEY
variation	I-KEY
.	O
However	O
,	O
the	O
ride	O
quality	O
of	O
the	O
ACS	O
should	O
be	O
investigated	O
further	O
because	O
the	O
movement	O
of	O
the	O
ACS	O
may	O
cause	O
motion	B-KEY
sickness	I-KEY
and	O
nausea	B-KEY
.	O
Experiments	O
of	O
ambulance	B-KEY
transportation	O
,	O
including	O
rapid	O
acceleration	O
and	O
deceleration	O
,	O
are	O
performed	O
to	O
evaluate	O
the	O
effect	O
of	O
differences	O
in	O
posture	O
of	O
the	O
transported	O
subject	O
on	O
the	O
ride	O
quality	O
;	O
the	O
semantic	O
differential	O
method	O
and	O
factor	O
analysis	O
are	O
used	O
in	O
the	O
investigations	O
.	O
Subjects	O
are	O
transported	O
using	O
a	O
conventional	B-KEY
stretcher	I-KEY
with	O
head	B-KEY
forward	I-KEY
,	O
a	O
conventional	B-KEY
stretcher	I-KEY
with	O
head	B-KEY
backward	I-KEY
,	O
the	O
ACS	O
,	O
and	O
an	O
assistant	O
driver	O
's	O
seat	O
for	O
comparison	O
with	O
transportation	O
using	O
a	O
stretcher	O
.	O
Experimental	O
results	O
show	O
that	O
the	O
ACS	O
gives	O
the	O
most	O
comfortable	B-KEY
transportation	I-KEY
when	O
using	O
a	O
stretcher	O
.	O
Moreover	O
,	O
the	O
reduction	O
of	O
the	O
negative	B-KEY
foot-to-head	I-KEY
acceleration	I-KEY
at	O
frequencies	O
below	O
0.2	O
Hz	O
and	O
the	O
small	O
variation	O
of	O
the	O
foot-to-head	O
acceleration	O
result	O
in	O
more	O
comfortable	B-KEY
transportation	I-KEY
.	O
Conventional	O
transportation	O
with	O
the	O
head	B-KEY
forward	I-KEY
causes	O
the	O
worst	O
transportation	O
,	O
although	O
the	O
characteristics	O
of	O
the	O
vibration	B-KEY
of	O
the	O
conventional	B-KEY
stretcher	I-KEY
seem	O
to	O
be	O
superior	O
to	O
that	O
of	O
the	O
ACS	O

Self-testing	B-KEY
chips	I-KEY
take	O
a	O
load	O
off	O
ATE	B-KEY
Looks	O
at	O
how	O
chipmakers	O
get	O
more	O
life	O
out	O
of	O
automatic	B-KEY
test	I-KEY
equipment	I-KEY
by	O
embedding	O
innovative	B-KEY
circuits	I-KEY
in	O
silicon	O

Why	O
information	B-KEY
departments	I-KEY
are	O
becoming	O
academic	O
This	O
article	O
outlines	O
the	O
increasing	O
convergence	O
between	O
academia	B-KEY
and	O
business	B-KEY
over	O
the	O
last	O
decade	O
or	O
so	O
,	O
and	O
the	O
mutual	O
benefits	O
that	O
this	O
closer	O
association	O
has	O
brought	O
.	O
It	O
also	O
looks	O
at	O
the	O
growing	O
importance	O
of	O
the	O
information	B-KEY
profession	I-KEY
,	O
suggesting	O
that	O
this	O
is	O
leading	O
to	O
a	O
greater	O
need	O
for	O
specialist	B-KEY
skills	I-KEY
,	O
as	O
reflected	O
by	O
the	O
rise	O
in	O
academic	B-KEY
courses	I-KEY
in	O
this	O
area	O
.	O
However	O
,	O
it	O
argues	O
that	O
increasing	O
specialization	O
must	O
not	O
lead	O
to	O
insularity	O
;	O
if	O
information	O
professionals	O
are	O
truly	O
concerned	O
with	O
gaining	O
a	O
competitive	O
advantage	O
,	O
they	O
must	O
not	O
close	O
their	O
minds	O
to	O
the	O
potential	O
benefits	O
of	O
working	O
with	O
external	O
,	O
non	O
specialist	O
,	O
partners	O
.	O
The	O
benefits	O
that	O
business	B-KEY
has	O
reaped	O
from	O
academia	B-KEY
,	O
it	O
is	O
contended	O
,	O
suggest	O
that	O
this	O
may	O
also	O
be	O
a	O
fruitful	O
avenue	O
for	O
information	B-KEY
departments	I-KEY
to	O
explore	O

Stochastic	B-KEY
recurrences	I-KEY
of	O
Jackpot	B-KEY
Keno	I-KEY
We	O
describe	O
a	O
mathematical	B-KEY
model	I-KEY
and	O
simulation	B-KEY
study	O
for	O
Jackpot	B-KEY
Keno	I-KEY
,	O
as	O
implemented	O
by	O
Jupiters	B-KEY
Network	I-KEY
Gaming	I-KEY
-LRB-	O
JNG	O
-RRB-	O
in	O
the	O
Australian	O
state	O
of	O
Queensland	O
,	O
and	O
as	O
controlled	O
by	O
the	O
Queensland	O
Office	O
of	O
Gaming	O
Regulation	O
-LRB-	O
QOGR	O
-RRB-	O
-LRB-	O
http://www.qogr.qld.gov.au/keno.shtml	O
-RRB-	O
.	O
The	O
recurrences	O
for	O
the	O
house	B-KEY
net	I-KEY
hold	I-KEY
are	O
derived	O
and	O
it	O
is	O
seen	O
that	O
these	O
are	O
piecewise	B-KEY
linear	I-KEY
with	O
a	O
ternary	B-KEY
domain	I-KEY
split	I-KEY
,	O
and	O
further	O
,	O
the	O
split	O
points	O
are	O
stochastic	O
in	O
nature	O
.	O
Since	O
this	O
structure	O
is	O
intractable	O
-LRB-	O
Brockett	O
and	O
Levine	O
,	O
Statistics	O
&	O
Probability	B-KEY
&	O
their	O
Applications	O
,	O
CBS	O
College	O
Publishing	O
,	O
1984	O
-RRB-	O
,	O
estimation	O
of	O
house	B-KEY
net	I-KEY
hold	I-KEY
obtained	O
through	O
an	O
appropriately	O
designed	O
simulator	B-KEY
using	O
a	O
random	B-KEY
number	I-KEY
generator	I-KEY
with	O
desirable	O
properties	O
is	O
described	O
.	O
Since	O
the	O
model	O
and	O
simulation	B-KEY
naturally	O
derives	O
hold	O
given	O
payscale	O
,	O
but	O
JNG	O
and	O
QOGR	O
require	O
payscale	O
given	O
hold	O
,	O
an	O
inverse	B-KEY
problem	I-KEY
was	O
required	O
to	O
be	O
solved	O
.	O
This	O
required	O
development	O
of	O
a	O
special	O
algorithm	O
,	O
which	O
may	O
be	O
described	O
as	O
a	O
stochastic	B-KEY
binary	I-KEY
search	I-KEY
.	O
Experimental	B-KEY
results	I-KEY
are	O
presented	O
,	O
in	O
which	O
the	O
simulator	B-KEY
is	O
used	O
to	O
determine	O
jackpot	O
pay-scales	O
so	O
as	O
to	O
satisfy	O
legal	B-KEY
requirements	I-KEY
of	O
approximately	O
75	O
%	O
of	O
net	O
revenue	O
returned	O
to	O
the	O
players	O
,	O
i.e.	O
,	O
25	O
%	O
net	O
hold	O
for	O
the	O
house	O
-LRB-	O
JNG	O
-RRB-	O
.	O
Details	O
of	O
the	O
algorithm	O
used	O
to	O
solve	O
this	O
problem	O
are	O
presented	O
,	O
and	O
notwithstanding	O
the	O
stochastic	O
nature	O
of	O
the	O
simulation	B-KEY
,	O
convergence	O
to	O
a	O
specified	O
hold	O
for	O
the	O
inverse	B-KEY
problem	I-KEY
has	O
been	O
achieved	O
to	O
within	O
0.1	O
%	O
in	O
all	O
cases	O
of	O
interest	O
to	O
date	O

Reply	O
to	O
Carreira-Perpinan	O
and	O
Goodhill	O
-LSB-	O
mathematics	O
in	O
biology	B-KEY
-RSB-	O
In	O
a	O
paper	O
by	O
Carreira-Perpinan	O
and	O
Goodhill	O
-LRB-	O
see	O
ibid.	O
,	O
vol	O
.14	O
,	O
no.	O
7	O
,	O
p.1545-60	O
,	O
2002	O
-RRB-	O
the	O
authors	O
apply	O
mathematical	B-KEY
arguments	I-KEY
to	O
biology	B-KEY
.	O
Swindale	O
et	O
al.	O
think	O
it	O
is	O
inappropriate	O
to	O
apply	O
the	O
standards	O
of	O
proof	O
required	O
in	O
mathematics	O
to	O
the	O
acceptance	O
or	O
rejection	O
of	O
scientific	B-KEY
hypotheses	I-KEY
.	O
To	O
give	O
some	O
examples	O
,	O
showing	O
that	O
data	O
are	O
well	O
described	O
by	O
a	O
linear	B-KEY
model	I-KEY
does	O
not	O
rule	O
out	O
an	O
infinity	O
of	O
other	O
possible	O
models	O
that	O
might	O
give	O
better	O
descriptions	O
of	O
the	O
data	O
.	O
Proving	O
in	O
a	O
mathematical	O
sense	O
that	O
the	O
linear	B-KEY
model	I-KEY
was	O
correct	O
would	O
require	O
ruling	O
out	O
all	O
other	O
possible	O
models	O
,	O
a	O
hopeless	O
task	O
.	O
Similarly	O
,	O
to	O
demonstrate	O
that	O
two	O
DNA	B-KEY
samples	O
come	O
from	O
the	O
same	O
individual	O
,	O
it	O
is	O
sufficient	O
to	O
show	O
a	O
match	O
between	O
only	O
a	O
few	O
regions	O
of	O
the	O
genome	B-KEY
,	O
even	O
though	O
there	O
remains	O
a	O
very	O
large	O
number	O
of	O
additional	O
comparisons	O
that	O
could	O
be	O
done	O
,	O
any	O
one	O
of	O
which	O
might	O
potentially	O
disprove	O
the	O
match	O
.	O
This	O
is	O
unacceptable	O
in	O
mathematics	O
,	O
but	O
in	O
the	O
real	O
world	O
,	O
it	O
is	O
a	O
perfectly	O
reasonable	O
basis	O
for	O
belief	O

Physical	B-KEY
quantum	I-KEY
algorithms	I-KEY
I	O
review	O
the	O
differences	O
between	O
classical	O
and	O
quantum	O
systems	O
,	O
emphasizing	O
the	O
connection	O
between	O
no-hidden	B-KEY
variable	I-KEY
theorems	I-KEY
and	O
superior	O
computational	O
power	O
of	O
quantum	B-KEY
computers	I-KEY
.	O
Using	O
quantum	B-KEY
lattice	I-KEY
gas	I-KEY
automata	I-KEY
as	O
examples	O
,	O
I	O
describe	O
possibilities	O
for	O
efficient	O
simulation	O
of	O
quantum	O
and	O
classical	B-KEY
systems	I-KEY
with	O
a	O
quantum	B-KEY
computer	I-KEY
.	O
I	O
conclude	O
with	O
a	O
list	O
of	O
research	O
directions	O

Decomposition	O
of	O
additive	O
cellular	B-KEY
automata	I-KEY
Finite	O
additive	O
cellular	B-KEY
automata	I-KEY
with	O
fixed	O
and	O
periodic	O
boundary	O
conditions	O
are	O
considered	O
as	O
endomorphisms	B-KEY
over	O
pattern	O
spaces	O
.	O
A	O
characterization	O
of	O
the	O
nilpotent	O
and	O
regular	O
parts	O
of	O
these	O
endomorphisms	B-KEY
is	O
given	O
in	O
terms	O
of	O
their	O
minimal	O
polynomials	O
.	O
Generalized	O
eigenspace	O
decomposition	O
is	O
determined	O
and	O
relevant	O
cyclic	O
subspaces	O
are	O
described	O
in	O
terms	O
of	O
symmetries	O
.	O
As	O
an	O
application	O
,	O
the	O
lengths	O
and	O
frequencies	O
of	O
limit	O
cycles	O
in	O
the	O
transition	B-KEY
diagram	I-KEY
of	O
the	O
automaton	O
are	O
calculated	O

Developing	O
a	O
hardware	B-KEY
and	I-KEY
programming	I-KEY
curriculum	I-KEY
for	O
middle	B-KEY
school	I-KEY
girls	I-KEY
Techbridge	B-KEY
provides	O
experiences	O
and	O
resources	O
that	O
would	O
teach	O
girls	O
technology	O
skills	O
as	O
well	O
as	O
excite	O
their	O
curiosity	O
and	O
build	O
their	O
confidence	O
.	O
Funded	O
by	O
the	O
National	O
Science	O
Foundation	O
and	O
sponsored	O
by	O
Chabot	O
Space	O
and	O
Science	O
Center	O
in	O
Oakland	O
,	O
California	O
,	O
Techbridge	B-KEY
is	O
a	O
three-year	O
program	O
that	O
serves	O
approximately	O
200	O
girls	O
annually	O
.	O
Techbridge	B-KEY
is	O
hosted	O
at	O
8	O
middle	O
and	O
high	O
schools	O
in	O
Oakland	O
and	O
at	O
the	O
California	O
School	O
for	O
the	O
Blind	O
in	O
Fremont	O
,	O
California	O
generally	O
as	O
an	O
after-school	O
program	O
meeting	O
once	O
a	O
week	O
.	O
Techbridge	B-KEY
comes	O
at	O
a	O
critical	O
time	O
in	O
girls	O
'	O
development	O
when	O
girls	O
have	O
many	O
important	O
decisions	O
to	O
make	O
regarding	O
classes	O
and	O
careers	O
,	O
but	O
often	O
lack	O
the	O
confidence	O
and	O
guidance	O
to	O
make	O
the	O
best	O
choices	O
.	O
Techbridge	B-KEY
helps	O
girls	O
plan	O
for	O
the	O
next	O
steps	O
to	O
high	O
school	O
and	O
college	O
with	O
its	O
role	O
models	O
and	O
guidance	O
.	O
Techbridge	B-KEY
also	O
provides	O
training	O
and	O
resources	O
for	O
teachers	O
,	O
counselors	O
,	O
and	O
families	O

Detection	O
and	O
estimation	O
of	O
abrupt	O
changes	O
in	O
the	O
variability	O
of	O
a	O
process	O
Detection	O
of	O
change-points	O
in	O
normal	O
means	O
is	O
a	O
well-studied	O
problem	O
.	O
The	O
parallel	O
problem	O
of	O
detecting	O
changes	O
in	O
variance	O
has	O
had	O
less	O
attention	O
.	O
The	O
form	O
of	O
the	O
generalized	B-KEY
likelihood	I-KEY
ratio	I-KEY
test	I-KEY
statistic	I-KEY
has	O
long	O
been	O
known	O
,	O
but	O
its	O
null	O
distribution	B-KEY
resisted	O
exact	O
analysis	O
.	O
In	O
this	O
paper	O
,	O
we	O
formulate	O
the	O
change-point	O
problem	O
for	O
a	O
sequence	B-KEY
of	O
chi-square	B-KEY
random	I-KEY
variables	I-KEY
.	O
We	O
describe	O
a	O
procedure	O
that	O
is	O
exact	O
for	O
the	O
distribution	B-KEY
of	O
the	O
likelihood	O
ratio	O
statistic	O
for	O
all	O
even	B-KEY
degrees	I-KEY
of	I-KEY
freedom	I-KEY
,	O
and	O
gives	O
upper	O
and	O
lower	B-KEY
bounds	I-KEY
for	O
odd	O
-LRB-	O
and	O
also	O
for	O
non-integer	O
-RRB-	O
degrees	O
of	O
freedom	O
.	O
Both	O
the	O
liberal	O
and	O
conservative	B-KEY
bounds	I-KEY
for	O
chi	O
/	O
sub	O
1	O
/	O
/	O
sup	O
2	O
/	O
degrees	O
of	O
freedom	O
are	O
shown	O
through	O
simulation	B-KEY
to	O
be	O
reasonably	O
tight	O
.	O
The	O
important	O
problem	O
of	O
testing	O
for	O
change	O
in	O
the	O
normal	O
variance	O
of	O
individual	B-KEY
observations	I-KEY
corresponds	O
to	O
the	O
chi	O
/	O
sub	O
1	O
/	O
/	O
sup	O
2	O
/	O
case	O
.	O
The	O
non-null	O
case	O
is	O
also	O
covered	O
,	O
and	O
confidence	B-KEY
intervals	I-KEY
for	O
the	O
true	O
change	O
point	O
are	O
derived	O
.	O
The	O
methodology	O
is	O
illustrated	O
with	O
an	O
application	O
to	O
quality	B-KEY
control	I-KEY
in	O
a	O
deep	B-KEY
level	I-KEY
gold	I-KEY
mine	I-KEY
.	O
Other	O
applications	O
include	O
ambulatory	B-KEY
monitoring	I-KEY
of	O
medical	B-KEY
data	I-KEY
and	O
econometrics	B-KEY

The	O
impact	O
and	O
implementation	O
of	O
XML	B-KEY
on	O
business-to-business	O
commerce	O
This	O
paper	O
discusses	O
the	O
impact	O
analysis	O
of	O
the	O
Extensible	B-KEY
Markup	I-KEY
Language	I-KEY
-LRB-	O
XML	B-KEY
-RRB-	O
.	O
Each	O
business	O
partner	O
within	O
a	O
supply	O
chain	O
will	O
be	O
allowed	O
to	O
generate	O
its	O
own	O
data	O
exchange	O
format	O
by	O
adopting	O
an	O
XML	B-KEY
meta-data	O
management	O
system	O
in	O
the	O
local	O
side	O
.	O
Followed	O
after	O
a	O
brief	O
introduction	O
of	O
the	O
information	O
technology	O
for	O
Business	B-KEY
to	I-KEY
Customer	I-KEY
-LRB-	O
B2C	O
-RRB-	O
and	O
Business	B-KEY
to	I-KEY
Business	I-KEY
-LRB-	O
B2B	O
-RRB-	O
Electronic	B-KEY
Commerce	I-KEY
-LRB-	O
EC	O
-RRB-	O
,	O
the	O
impact	O
of	O
XML	B-KEY
on	O
the	O
tomorrow	O
business	O
world	O
is	O
discussed	O
.	O
A	O
real	O
case	O
study	O
for	O
impact	O
analysis	O
on	O
information	O
exchange	O
platform	O
,	O
Microsoft	O
's	O
BizTalk	B-KEY
platform	O
which	O
is	O
actually	O
an	O
XML	B-KEY
schema	O
builder	O
and	O
the	O
implementation	O
of	O
XML	O
commerce	O
application	O
will	O
provide	O
an	O
interest	O
insight	O
for	O
users	O
'	O
future	O
implementation	O

The	O
3D	B-KEY
visibility	I-KEY
complex	I-KEY
Visibility	O
problems	O
are	O
central	O
to	O
many	O
computer	B-KEY
graphics	I-KEY
applications	O
.	O
The	O
most	O
common	O
examples	O
include	O
hidden-part	B-KEY
removal	I-KEY
for	O
view	B-KEY
computation	I-KEY
,	O
shadow	B-KEY
boundaries	I-KEY
,	O
mutual	O
visibility	O
of	O
objects	O
for	O
lighting	B-KEY
simulation	I-KEY
.	O
In	O
this	O
paper	O
,	O
we	O
present	O
a	O
theoretical	O
study	O
of	O
3D	O
visibility	O
properties	O
for	O
scenes	O
of	O
smooth	B-KEY
convex	I-KEY
objects	I-KEY
.	O
We	O
work	O
in	O
the	O
space	O
of	O
light	B-KEY
rays	I-KEY
,	O
or	O
more	O
precisely	O
,	O
of	O
maximal	B-KEY
free	I-KEY
segments	I-KEY
.	O
We	O
group	O
segments	O
that	O
``	O
see	O
''	O
the	O
same	O
object	O
;	O
this	O
defines	O
the	O
3D	B-KEY
visibility	I-KEY
complex	I-KEY
.	O
The	O
boundaries	O
of	O
these	O
groups	O
of	O
segments	O
correspond	O
to	O
the	O
visual	B-KEY
events	I-KEY
of	O
the	O
scene	O
-LRB-	O
limits	O
of	O
shadows	O
,	O
disappearance	O
of	O
an	O
object	O
when	O
the	O
viewpoint	O
is	O
moved	O
,	O
etc.	O
-RRB-	O
.	O
We	O
provide	O
a	O
worst	O
case	O
analysis	O
of	O
the	O
complexity	O
of	O
the	O
visibility	O
complex	O
of	O
3D	O
scenes	O
,	O
as	O
well	O
as	O
a	O
probabilistic	B-KEY
study	I-KEY
under	O
a	O
simple	O
assumption	O
for	O
``	O
normal	O
''	O
scenes	O
.	O
We	O
extend	O
the	O
visibility	O
complex	O
to	O
handle	O
temporal	B-KEY
visibility	I-KEY
.	O
We	O
give	O
an	O
output-sensitive	B-KEY
construction	I-KEY
algorithm	I-KEY
and	O
present	O
applications	O
of	O
our	O
approach	O

Traffic	O
engineering	O
with	O
traditional	O
IP	B-KEY
routing	I-KEY
protocols	I-KEY
Traffic	O
engineering	O
involves	O
adapting	O
the	O
routing	O
of	O
traffic	O
to	O
network	B-KEY
conditions	I-KEY
,	O
with	O
the	O
joint	O
goals	O
of	O
good	O
user	B-KEY
performance	I-KEY
and	O
efficient	O
use	O
of	O
network	B-KEY
resources	I-KEY
.	O
We	O
describe	O
an	O
approach	O
to	O
intradomain	B-KEY
traffic	I-KEY
engineering	I-KEY
that	O
works	O
within	O
the	O
existing	O
deployed	O
base	O
of	O
interior	B-KEY
gateway	I-KEY
protocols	I-KEY
,	O
such	O
as	O
Open	O
Shortest	O
Path	O
First	O
and	O
Intermediate	O
System-Intermediate	O
System	O
.	O
We	O
explain	O
how	O
to	O
adapt	O
the	O
configuration	O
of	O
link	O
weights	O
,	O
based	O
on	O
a	O
networkwide	O
view	O
of	O
the	O
traffic	O
and	O
topology	O
within	O
a	O
domain	O
.	O
In	O
addition	O
,	O
we	O
summarize	O
the	O
results	O
of	O
several	O
studies	O
of	O
techniques	O
for	O
optimizing	O
OSPF/IS-IS	B-KEY
weights	I-KEY
to	O
the	O
prevailing	O
traffic	O
.	O
The	O
article	O
argues	O
that	O
traditional	O
shortest	B-KEY
path	I-KEY
routing	I-KEY
protocols	I-KEY
are	O
surprisingly	O
effective	O
for	O
engineering	O
the	O
flow	O
of	O
traffic	O
in	O
large	O
IP	B-KEY
networks	I-KEY

Multiple	B-KEY
shooting	I-KEY
using	O
a	O
dichotomically	B-KEY
stable	I-KEY
integrator	I-KEY
for	O
solving	O
differential-algebraic	B-KEY
equations	I-KEY
In	O
previous	O
work	O
by	O
the	O
first	O
author	O
,	O
it	O
has	O
been	O
established	O
that	O
a	O
dichotomically	O
stable	O
discretization	O
is	O
needed	O
when	O
solving	O
a	O
stiff	B-KEY
boundary-value	I-KEY
problem	I-KEY
in	O
ordinary	B-KEY
differential	I-KEY
equations	I-KEY
-LRB-	O
ODEs	O
-RRB-	O
,	O
when	O
sharp	O
boundary	O
layers	O
may	O
occur	O
at	O
each	O
end	O
of	O
the	O
interval	O
.	O
A	O
dichotomically	O
stable	O
implicit	B-KEY
Runge-Kutta	I-KEY
method	I-KEY
,	O
using	O
the	O
3-stage	O
,	O
fourth-order	O
,	O
Lobatto	B-KEY
IIIA	I-KEY
formulae	I-KEY
,	O
has	O
been	O
implemented	O
in	O
a	O
variable	O
step-size	O
initial-value	B-KEY
integrator	I-KEY
,	O
which	O
could	O
be	O
used	O
in	O
a	O
multiple-shooting	O
approach	O
.	O
In	O
the	O
case	O
of	O
index-one	O
differential-algebraic	B-KEY
equations	I-KEY
-LRB-	O
DAEs	O
-RRB-	O
the	O
use	O
of	O
the	O
Lobatto	B-KEY
IIIA	I-KEY
formulae	I-KEY
has	O
an	O
advantage	O
,	O
over	O
a	O
comparable	O
Gaussian	O
method	O
,	O
that	O
the	O
order	O
is	O
the	O
same	O
for	O
both	O
differential	O
and	O
algebraic	O
variables	O
,	O
and	O
there	O
is	O
no	O
need	O
to	O
treat	O
them	O
separately	O
.	O
The	O
ODE	O
integrator	O
has	O
been	O
adapted	O
for	O
the	O
solution	O
of	O
index-one	O
DAEs	O
,	O
and	O
the	O
resulting	O
integrator	O
-LRB-	O
SYMDAE	O
-RRB-	O
has	O
been	O
inserted	O
into	O
the	O
multiple-shooting	O
code	O
-LRB-	O
MSHDAE	O
-RRB-	O
previously	O
developed	O
by	O
R.	O
Lamour	O
for	O
differential-algebraic	O
boundary-value	O
problems	O
.	O
The	O
standard	O
version	O
of	O
MSHDAE	O
uses	O
a	O
BDF	O
integrator	O
,	O
which	O
is	O
not	O
dichotomically	O
stable	O
,	O
and	O
for	O
some	O
stiff	O
test	O
problems	O
this	O
fails	O
to	O
integrate	O
across	O
the	O
interval	O
of	O
interest	O
,	O
while	O
the	O
dichotomically	B-KEY
stable	I-KEY
integrator	I-KEY
SYMDAE	O
encounters	O
no	O
difficulty	O
.	O
Indeed	O
,	O
for	O
such	O
problems	O
,	O
the	O
modified	O
version	O
of	O
MSHDAE	O
produces	O
an	O
accurate	O
solution	O
,	O
and	O
within	O
limits	O
imposed	O
by	O
computer	O
word	O
length	O
,	O
the	O
efficiency	O
of	O
the	O
solution	O
process	O
improves	O
with	O
increasing	O
stiffness	O
.	O
For	O
some	O
nonstiff	O
problems	O
,	O
the	O
solution	O
is	O
also	O
entirely	O
satisfactory	O

Evolution	B-KEY
complexity	I-KEY
of	O
the	O
elementary	O
cellular	O
automaton	O
rule	O
18	O
Cellular	B-KEY
automata	I-KEY
are	O
classes	O
of	O
mathematical	O
systems	O
characterized	O
by	O
discreteness	O
-LRB-	O
in	O
space	O
,	O
time	O
,	O
and	O
state	O
values	O
-RRB-	O
,	O
determinism	O
,	O
and	O
local	O
interaction	O
.	O
Using	O
symbolic	B-KEY
dynamical	I-KEY
theory	I-KEY
,	O
we	O
coarse-grain	O
the	O
temporal	O
evolution	O
orbits	O
of	O
cellular	B-KEY
automata	I-KEY
.	O
By	O
means	O
of	O
formal	B-KEY
languages	I-KEY
and	O
automata	O
theory	O
,	O
we	O
study	O
the	O
evolution	B-KEY
complexity	I-KEY
of	O
the	O
elementary	O
cellular	O
automaton	O
with	O
local	O
rule	O
number	O
18	O
and	O
prove	O
that	O
its	O
width	O
1-evolution	O
language	O
is	O
regular	O
,	O
but	O
for	O
every	O
n	O
>	O
or	O
=	O
2	O
its	O
width	O
n-evolution	O
language	O
is	O
not	O
context	O
free	O
but	O
context	O
sensitive	O

Application	O
of	O
hybrid	B-KEY
models	I-KEY
for	O
prediction	O
and	O
optimization	B-KEY
of	O
enzyme	B-KEY
fermentation	I-KEY
process	O
.	O
A	O
comparative	O
study	O
The	O
paper	O
presents	O
a	O
comparison	O
of	O
the	O
biotechnological	O
process	O
prediction	O
and	O
optimization	B-KEY
results	O
obtained	O
by	O
using	O
different	O
structure	O
hybrid	O
mathematical	B-KEY
models	I-KEY
for	O
modeling	O
of	O
the	O
same	O
bioprocess	B-KEY
.	O
The	O
hybrid	B-KEY
models	I-KEY
under	O
investigation	O
consist	O
of	O
the	O
product	B-KEY
mass	I-KEY
balance	I-KEY
equation	I-KEY
in	O
which	O
different	O
means	O
-	O
an	O
artificial	O
neural	O
network	O
,	O
fuzzy-neural	B-KEY
network	I-KEY
and	O
cell	B-KEY
age	I-KEY
distribution	I-KEY
based	O
calculation	O
scheme	O
-	O
are	O
incorporated	O
for	O
modeling	O
the	O
specific	O
biosynthesis	B-KEY
rate	I-KEY
of	O
a	O
desired	O
product	O
.	O
Experimental	O
data	O
from	O
alpha	O
-	O
amylase	O
laboratory	O
and	O
industrial	O
fermentation	O
processes	O
are	O
used	O
for	O
model	O
parameter	O
identification	B-KEY
and	O
the	O
process	O
prediction	O
tests	O

Packet	O
spacing	O
:	O
an	O
enabling	O
mechanism	O
for	O
delivering	O
multimedia	O
content	O
in	O
computational	O
grids	O
Streaming	B-KEY
multimedia	I-KEY
with	O
UDP	B-KEY
has	O
become	O
increasingly	O
popular	O
over	O
distributed	B-KEY
systems	I-KEY
like	O
the	O
Internet	B-KEY
.	O
Scientific	O
applications	O
that	O
stream	B-KEY
multimedia	I-KEY
include	O
remote	B-KEY
computational	I-KEY
steering	I-KEY
of	O
visualization	B-KEY
data	I-KEY
and	O
video-on-demand	O
teleconferencing	O
over	O
the	O
Access	O
Grid	O
.	O
However	O
,	O
UDP	B-KEY
does	O
not	O
possess	O
a	O
self-regulating	O
,	O
congestion-control	O
mechanism	O
;	O
and	O
most	O
best-effort	O
traffic	O
is	O
served	O
by	O
congestion-controlled	O
TCP	O
.	O
Consequently	O
,	O
UDP	B-KEY
steals	O
bandwidth	O
from	O
TCP	O
such	O
that	O
TCP	O
flows	O
starve	O
for	O
network	O
resources	O
.	O
With	O
the	O
volume	O
of	O
Internet	B-KEY
traffic	O
continuing	O
to	O
increase	O
,	O
the	O
perpetuation	O
of	O
UDP-based	O
streaming	O
will	O
cause	O
the	O
Internet	O
to	O
collapse	O
as	O
it	O
did	O
in	O
the	O
mid-1980	O
's	O
due	O
to	O
the	O
use	O
of	O
non-congestion-controlled	O
TCP	O
.	O
To	O
address	O
this	O
problem	O
,	O
we	O
introduce	O
the	O
counter-intuitive	O
notion	O
of	O
inter-packet	B-KEY
spacing	I-KEY
with	O
control	O
feedback	O
to	O
enable	O
UDP-based	O
applications	O
to	O
perform	O
well	O
in	O
the	O
next-generation	O
Internet	O
and	O
computational	O
grids	O
.	O
When	O
compared	O
with	O
traditional	O
UDP-based	O
streaming	O
,	O
we	O
illustrate	O
that	O
our	O
approach	O
can	O
reduce	O
packet	O
loss	O
over	O
50	O
%	O
without	O
adversely	O
affecting	O
delivered	O
throughput	O

Defending	O
against	O
flooding-based	B-KEY
distributed	I-KEY
denial-of-service	I-KEY
attacks	I-KEY
:	O
a	O
tutorial	B-KEY
Flooding-based	O
distributed	O
denial-of-service	O
-LRB-	O
DDoS	O
-RRB-	O
attack	O
presents	O
a	O
very	O
serious	O
threat	O
to	O
the	O
stability	O
of	O
the	O
Internet	O
.	O
In	O
a	O
typical	O
DDoS	O
attack	O
,	O
a	O
large	O
number	O
of	O
compromised	O
hosts	O
are	O
amassed	O
to	O
send	O
useless	O
packets	O
to	O
jam	O
a	O
victim	O
,	O
or	O
its	O
Internet	O
connection	O
,	O
or	O
both	O
.	O
In	O
the	O
last	O
two	O
years	O
,	O
it	O
was	O
discovered	O
that	O
DDoS	B-KEY
attack	I-KEY
methods	I-KEY
and	O
tools	O
are	O
becoming	O
more	O
sophisticated	O
,	O
effective	O
,	O
and	O
also	O
more	O
difficult	O
to	O
trace	O
to	O
the	O
real	O
attackers	O
.	O
On	O
the	O
defense	O
side	O
,	O
current	O
technologies	O
are	O
still	O
unable	O
to	O
withstand	O
large-scale	B-KEY
attacks	I-KEY
.	O
The	O
main	O
purpose	O
of	O
this	O
article	O
is	O
therefore	O
twofold	O
.	O
The	O
first	O
one	O
is	O
to	O
describe	O
various	O
DDoS	B-KEY
attack	I-KEY
methods	I-KEY
,	O
and	O
to	O
present	O
a	O
systematic	O
review	O
and	O
evaluation	O
of	O
the	O
existing	O
defense	O
mechanisms	O
.	O
The	O
second	O
is	O
to	O
discuss	O
a	O
longer-term	O
solution	O
,	O
dubbed	O
the	O
Internet-firewall	O
approach	O
,	O
that	O
attempts	O
to	O
intercept	O
attack	O
packets	O
in	O
the	O
Internet	O
core	O
,	O
well	O
before	O
reaching	O
the	O
victim	O

Life	O
after	O
bankruptcy	B-KEY
-LSB-	O
telecom	B-KEY
carriers	I-KEY
-RSB-	O
How	O
comeback	O
telecom	B-KEY
carriers	I-KEY
are	O
changing	O
industry	B-KEY
economics	I-KEY
,	O
and	O
why	O
others	O
may	O
have	O
no	O
choice	O
but	O
to	O
follow	O
their	O
lead	O

Clausal	O
resolution	O
in	O
a	O
logic	O
of	O
rational	O
agency	O
A	O
resolution	B-KEY
based	I-KEY
proof	I-KEY
system	I-KEY
for	O
a	O
Temporal	B-KEY
Logic	I-KEY
of	O
Possible	O
Belief	B-KEY
is	O
presented	O
.	O
This	O
logic	O
is	O
the	O
combination	O
of	O
the	O
branching-time	O
temporal	B-KEY
logic	I-KEY
CTL	O
-LRB-	O
representing	O
change	O
over	O
time	O
-RRB-	O
with	O
the	O
modal	O
logic	O
KD45	O
-LRB-	O
representing	O
belief	O
-RRB-	O
.	O
Such	O
combinations	O
of	O
temporal	O
or	O
dynamic	B-KEY
logics	I-KEY
and	O
modal	B-KEY
logics	I-KEY
are	O
useful	O
for	O
specifying	O
complex	O
properties	O
of	O
multi-agent	B-KEY
systems	I-KEY
.	O
Proof	O
methods	O
are	O
important	O
for	O
developing	O
verification	O
techniques	O
for	O
these	O
complex	O
multi-modal	B-KEY
logics	I-KEY
.	O
Soundness	O
,	O
completeness	O
and	O
termination	O
of	O
the	O
proof	O
method	O
are	O
shown	O
and	O
simple	O
examples	O
illustrating	O
its	O
use	O
are	O
given	O

On	O
quasi-linear	O
PDAEs	O
with	O
convection	B-KEY
:	O
applications	O
,	O
indices	B-KEY
,	O
numerical	B-KEY
solution	I-KEY
For	O
a	O
class	O
of	O
partial	O
differential	O
algebraic	O
equations	O
-LRB-	O
PDAEs	O
-RRB-	O
of	O
quasi-linear	O
type	O
which	O
include	O
nonlinear	O
terms	O
of	O
convection	B-KEY
type	O
,	O
a	O
possibility	O
to	O
determine	O
a	O
time	O
and	O
spatial	B-KEY
index	I-KEY
is	O
considered	O
.	O
As	O
a	O
typical	O
example	O
we	O
investigate	O
an	O
application	O
from	O
plasma	B-KEY
physics	I-KEY
.	O
Especially	O
we	O
discuss	O
the	O
numerical	B-KEY
solution	I-KEY
of	O
initial	B-KEY
boundary	I-KEY
value	I-KEY
problems	I-KEY
by	O
means	O
of	O
a	O
corresponding	O
finite	B-KEY
difference	I-KEY
splitting	I-KEY
procedure	I-KEY
which	O
is	O
a	O
modification	O
of	O
a	O
well-known	O
fractional	B-KEY
step	I-KEY
method	I-KEY
coupled	O
with	O
a	O
matrix	B-KEY
factorization	I-KEY
.	O
The	O
convergence	O
of	O
the	O
numerical	B-KEY
solution	I-KEY
towards	O
the	O
exact	O
solution	O
of	O
the	O
corresponding	O
initial	B-KEY
boundary	I-KEY
value	I-KEY
problem	I-KEY
is	O
investigated	O
.	O
Some	O
results	O
of	O
a	O
numerical	B-KEY
solution	I-KEY
of	O
the	O
plasma	O
PDAE	O
are	O
given	O

Local	O
search	O
with	O
constraint	O
propagation	O
and	O
conflict-based	O
heuristics	O
Search	B-KEY
algorithms	I-KEY
for	O
solving	O
CSP	B-KEY
-LRB-	O
Constraint	B-KEY
Satisfaction	I-KEY
Problems	I-KEY
-RRB-	O
usually	O
fall	O
into	O
one	O
of	O
two	O
main	O
families	O
:	O
local	B-KEY
search	I-KEY
algorithms	I-KEY
and	O
systematic	O
algorithms	O
.	O
Both	O
families	O
have	O
their	O
advantages	O
.	O
Designing	O
hybrid	O
approaches	O
seems	O
promising	O
since	O
those	O
advantages	O
may	O
be	O
combined	O
into	O
a	O
single	O
approach	O
.	O
In	O
this	O
paper	O
,	O
we	O
present	O
a	O
new	O
hybrid	O
technique	O
.	O
It	O
performs	O
a	O
local	O
search	O
over	O
partial	B-KEY
assignments	I-KEY
instead	O
of	O
complete	O
assignments	O
,	O
and	O
uses	O
filtering	B-KEY
techniques	I-KEY
and	O
conflict-based	O
techniques	O
to	O
efficiently	O
guide	O
the	O
search	O
.	O
This	O
new	O
technique	O
benefits	O
from	O
both	O
classical	O
approaches	O
:	O
a	O
priori	O
pruning	O
of	O
the	O
search	O
space	O
from	O
filtering-based	O
search	O
and	O
possible	O
repair	O
of	O
early	O
mistakes	O
from	O
local	O
search	O
.	O
We	O
focus	O
on	O
a	O
specific	O
version	O
of	O
this	O
technique	O
:	O
tabu	B-KEY
decision-repair	I-KEY
.	O
Experiments	O
done	O
on	O
open-shop	O
scheduling	O
problems	O
show	O
that	O
our	O
approach	O
competes	O
well	O
with	O
the	O
best	O
highly	O
specialized	O
algorithms	O

A	O
variable-stepsize	B-KEY
variable-order	I-KEY
multistep	I-KEY
method	I-KEY
for	O
the	O
integration	O
of	O
perturbed	O
linear	O
problems	O
G.	O
Scheifele	O
-LRB-	O
1971	O
-RRB-	O
wrote	O
the	O
solution	O
of	O
a	O
perturbed	B-KEY
oscillator	I-KEY
as	O
an	O
expansion	O
in	O
terms	O
of	O
a	O
new	O
set	O
of	O
functions	O
,	O
which	O
extends	O
the	O
monomials	B-KEY
in	O
the	O
Taylor	B-KEY
series	I-KEY
of	O
the	O
solution	O
.	O
Recently	O
,	O
P.	O
Martin	O
and	O
J.M.	O
Ferrandiz	O
-LRB-	O
1997	O
-RRB-	O
constructed	O
a	O
multistep	O
code	B-KEY
based	I-KEY
on	O
the	O
Scheifele	O
technique	O
,	O
and	O
it	O
was	O
generalized	O
by	O
D.J.	O
Lopez	O
and	O
P.	O
Martin	O
-LRB-	O
1998	O
-RRB-	O
for	O
perturbed	O
linear	O
problems	O
.	O
However	O
,	O
the	O
remarked	O
codes	O
are	O
constant	B-KEY
steplength	I-KEY
methods	I-KEY
,	O
and	O
efficient	O
integrators	O
must	O
be	O
able	O
to	O
change	O
the	O
steplength	O
.	O
In	O
this	O
paper	O
we	O
extend	O
the	O
ideas	O
of	O
F.T.	O
Krogh	O
-LRB-	O
1974	O
-RRB-	O
from	O
Adams	O
methods	B-KEY
to	I-KEY
the	O
algorithm	O
proposed	O
by	O
Lopez	O
and	O
Martin	O
,	O
and	O
we	O
show	O
the	O
advantages	O
of	O
the	O
new	O
code	O
in	O
perturbed	O
problems	O

The	O
MAGNeT	B-KEY
toolkit	O
:	O
design	O
,	O
implementation	O
and	O
evaluation	O
The	O
current	O
trend	O
in	O
constructing	O
high-performance	B-KEY
computing	I-KEY
systems	O
is	O
to	O
connect	O
a	O
large	O
number	O
of	O
machines	O
via	O
a	O
fast	O
interconnect	O
or	O
a	O
large-scale	O
network	O
such	O
as	O
the	O
Internet	O
.	O
This	O
approach	O
relies	O
on	O
the	O
performance	O
of	O
the	O
interconnect	B-KEY
-LRB-	O
or	O
Internet	B-KEY
-RRB-	O
to	O
enable	O
fast	O
,	O
large-scale	O
distributed	O
computing	O
.	O
A	O
detailed	O
understanding	O
of	O
the	O
communication	O
traffic	O
is	O
required	O
in	O
order	O
to	O
optimize	O
the	O
operation	O
of	O
the	O
entire	O
system	O
.	O
Network	O
researchers	O
traditionally	O
monitor	O
traffic	O
in	O
the	O
network	O
to	O
gain	O
the	O
insight	O
necessary	O
to	O
optimize	B-KEY
network	I-KEY
operations	O
.	O
Recent	O
work	O
suggests	O
additional	O
insight	O
can	O
be	O
obtained	O
by	O
also	O
monitoring	O
traffic	O
at	O
the	O
application	O
level	O
.	O
The	O
Monitor	B-KEY
for	I-KEY
Application-Generated	I-KEY
Network	I-KEY
Traffic	I-KEY
toolkit	I-KEY
-LRB-	O
MAGNeT	B-KEY
-RRB-	O
we	O
describe	O
here	O
monitors	O
application	O
traffic	O
patterns	O
in	O
production	O
systems	O
,	O
thus	O
enabling	O
more	O
highly	O
optimized	B-KEY
networks	I-KEY
and	O
interconnects	B-KEY
for	O
the	O
next	O
generation	O
of	O
high-performance	B-KEY
computing	I-KEY
systems	O

The	O
road	O
ahead	O
-LSB-	O
supply	B-KEY
chains	I-KEY
-RSB-	O
Executive	O
supply	B-KEY
chain	I-KEY
managers	O
,	O
says	O
David	O
Metcalfe	O
of	O
Forrester	B-KEY
Research	I-KEY
,	O
need	O
the	O
skills	O
and	O
precision	O
of	O
Mongolian	O
archers	O
on	O
horseback	O
.	O
They	O
must	O
be	O
able	O
to	O
hit	O
their	O
target	O
,	O
in	O
this	O
case	O
customer	O
demand	O
,	O
while	O
moving	O
at	O
great	O
speed	O
.	O
But	O
what	O
is	O
wrong	O
with	O
the	O
supply	B-KEY
chains	I-KEY
companies	O
have	O
in	O
place	O
already	O
?	O
According	O
to	O
Metcalfe	O
,	O
current	O
manufacturing	B-KEY
models	O
are	O
too	O
inflexible	O
.	O
A	O
recent	O
survey	B-KEY
conducted	O
by	O
Forrester	B-KEY
Research	I-KEY
supports	O
this	O
claim	O
.	O
It	O
found	O
that	O
42	O
%	O
of	O
respondents	O
could	O
not	O
transfer	O
production	O
from	O
one	O
plant	O
to	O
another	O
in	O
the	O
event	O
of	O
a	O
glitch	O
in	O
the	O
supply	B-KEY
chain	I-KEY
.	O
A	O
further	O
32	O
%	O
said	O
it	O
would	O
be	O
possible	O
,	O
but	O
extremely	O
costly	O

Estimating	O
the	O
intrinsic	O
dimension	O
of	O
data	O
with	O
a	O
fractal-based	B-KEY
method	I-KEY
In	O
this	O
paper	O
,	O
the	O
problem	O
of	O
estimating	O
the	O
intrinsic	O
dimension	O
of	O
a	O
data	O
set	O
is	O
investigated	O
.	O
A	O
fractal-based	O
approach	O
using	O
the	O
Grassberger-Procaccia	O
algorithm	O
is	O
proposed	O
.	O
Since	O
the	O
Grassberger-Procaccia	O
algorithm	O
-LRB-	O
1983	O
-RRB-	O
performs	O
badly	O
on	O
sets	O
of	O
high	O
dimensionality	O
,	O
an	O
empirical	O
procedure	O
that	O
improves	O
the	O
original	O
algorithm	O
has	O
been	O
developed	O
.	O
The	O
procedure	O
has	O
been	O
tested	O
on	O
data	O
sets	O
of	O
known	O
dimensionality	O
and	O
on	O
time	B-KEY
series	I-KEY
of	O
Santa	B-KEY
Fe	I-KEY
competition	I-KEY

New	O
wrinkle	O
on	O
the	O
Web	O
?	O
Hmm	O
.	O
-LSB-	O
banking	B-KEY
-RSB-	O
The	O
financial	O
sector	O
produced	O
its	O
share	O
of	O
technology	O
hype	O
during	O
the	O
new	O
economy	O
years	O
.	O
You	O
.	O
ca	O
n't	O
blame	O
folks	O
if	O
the	O
next	O
next	O
thing	O
,	O
a	O
wave	O
of	O
Internet-related	O
innovation	O
called	O
Web	B-KEY
services	I-KEY
,	O
is	O
being	O
met	O
with	O
healthy	O
skepticism	O
.	O
Many	O
gurus	O
are	O
placing	O
their	O
bets	O
on	O
Web	B-KEY
services	I-KEY
to	O
drive	O
the	O
next	O
chapter	O
of	O
finance	O
technology	O
,	O
dramatically	O
upgrading	O
disappointing	O
automated	O
customer	O
management	O
strategies	O
by	O
electronically	O
breaking	O
down	O
barriers	O
between	O
products	O
,	O
firms	O
and	O
customers	O
,	O
and	O
perhaps	O
creating	O
a	O
whole	O
new	O
line	O
of	O
business	O
in	O
the	O
process	O
.	O
But	O
it	O
's	O
not	O
a	O
magic	O
wand	O
.	O
It	O
does	O
n't	O
change	O
the	O
need	O
for	O
a	O
bank	B-KEY
to	O
reorganize	O
and	O
streamline	O
its	O
operations	O

Numerical	B-KEY
behaviour	I-KEY
of	O
stable	O
and	O
unstable	B-KEY
solitary	I-KEY
waves	I-KEY
In	O
this	O
paper	O
we	O
analyse	O
the	O
behaviour	O
in	O
time	O
of	O
the	O
numerical	B-KEY
approximations	I-KEY
to	O
solitary	O
wave	O
solutions	O
of	O
the	O
generalized	B-KEY
Benjamin-Bona-Mahony	I-KEY
equation	I-KEY
.	O
This	O
equation	O
possesses	O
an	O
important	O
property	O
:	O
the	O
stability	O
of	O
these	O
solutions	O
depends	O
on	O
their	O
velocity	O
.	O
We	O
identify	O
the	O
error	B-KEY
propagation	I-KEY
mechanisms	I-KEY
in	O
both	O
the	O
stable	O
and	O
unstable	O
case	O
.	O
In	O
particular	O
,	O
we	O
show	O
that	O
in	O
the	O
stable	O
case	O
,	O
numerical	B-KEY
methods	I-KEY
that	O
preserve	O
some	O
conserved	O
quantities	O
of	O
the	O
problem	O
are	O
more	O
appropriate	O
for	O
the	O
simulation	O
of	O
this	O
kind	O
of	O
solutions	O

Network-centric	B-KEY
systems	I-KEY
The	O
author	O
describes	O
a	O
graduate-level	O
course	O
that	O
addresses	O
cutting-edge	O
issues	O
in	O
network-centric	B-KEY
systems	I-KEY
while	O
following	O
a	O
more	O
traditional	O
graduate	O
seminar	O
format	O

An	O
approximation	O
to	O
the	O
F	B-KEY
distribution	I-KEY
using	O
the	O
chi-square	B-KEY
distribution	I-KEY
For	O
the	O
cumulative	B-KEY
distribution	I-KEY
function	I-KEY
-LRB-	O
c.d.f.	O
-RRB-	O
of	O
the	O
F	B-KEY
distribution	I-KEY
,	O
F	O
-LRB-	O
x	O
;	O
k	O
,	O
n	O
-RRB-	O
,	O
with	O
associated	O
degrees	B-KEY
of	I-KEY
freedom	I-KEY
,	O
k	O
and	O
n	O
,	O
a	O
shrinking	B-KEY
factor	I-KEY
approximation	I-KEY
-LRB-	O
SFA	O
-RRB-	O
,	O
G	O
-LRB-	O
lambda	O
kx	O
;	O
k	O
-RRB-	O
,	O
is	O
proposed	O
for	O
large	O
n	O
and	O
any	O
fixed	O
k	O
,	O
where	O
G	O
-LRB-	O
x	O
;	O
k	O
-RRB-	O
is	O
the	O
chi-square	O
c.d.f.	O
with	O
degrees	B-KEY
of	I-KEY
freedom	I-KEY
,	O
k	O
,	O
and	O
lambda	O
=	O
lambda	O
-LRB-	O
kx	O
;	O
n	O
-RRB-	O
is	O
the	O
shrinking	O
factor	O
.	O
Numerical	B-KEY
analysis	I-KEY
indicates	O
that	O
for	O
n/k	O
>	O
or	O
=	O
3	O
,	O
approximation	O
accuracy	O
of	O
the	O
SFA	O
is	O
to	O
the	O
fourth	O
decimal	O
place	O
for	O
most	O
small	O
values	O
of	O
k	O
.	O
This	O
is	O
a	O
substantial	O
improvement	O
on	O
the	O
accuracy	O
that	O
is	O
achievable	O
using	O
the	O
normal	O
,	O
ordinary	O
chi-square	O
,	O
and	O
Scheffe-Tukey	O
approximations	O
.	O
In	O
addition	O
,	O
it	O
is	O
shown	O
that	O
the	O
theoretical	O
approximation	O
error	O
of	O
the	O
SFA	O
,	O
|	O
F	O
-LRB-	O
x	O
;	O
k	O
,	O
n	O
-RRB-	O
-	O
G	O
-LRB-	O
lambda	O
kx	O
;	O
k	O
-RRB-	O
|	O
,	O
is	O
O	O
-LRB-	O
1/n/sup	O
2	O
/	O
-RRB-	O
uniformly	O
over	O
x	O

A	O
synergic	B-KEY
analysis	I-KEY
for	O
Web-based	O
enterprise	B-KEY
resources	I-KEY
planning	I-KEY
systems	O
As	O
the	O
central	O
nervous	O
system	O
for	O
managing	O
an	O
organization	O
's	O
mission	O
and	O
critical	O
business	O
data	O
,	O
Enterprise	B-KEY
Resource	I-KEY
Planning	I-KEY
-LRB-	O
ERP	B-KEY
-RRB-	O
system	O
has	O
evolved	O
to	O
become	O
the	O
backbone	O
of	O
e-business	B-KEY
implementation	O
.	O
Since	O
an	O
ERP	B-KEY
system	O
is	O
multimodule	O
application	O
software	O
that	O
helps	O
a	O
company	O
manage	O
its	O
important	O
business	O
functions	O
,	O
it	O
should	O
be	O
versatile	O
enough	O
to	O
automate	O
every	O
aspect	O
of	O
business	O
processes	O
,	O
including	O
e-business	B-KEY

Pareto-optimal	O
formulations	O
for	O
cost	O
versus	O
colorimetric	O
accuracy	O
trade-offs	O
in	O
printer	O
color	O
management	O
Color	O
management	O
for	O
the	O
printing	O
of	O
digital	O
images	O
is	O
a	O
challenging	O
task	O
,	O
due	O
primarily	O
to	O
nonlinear	B-KEY
ink-mixing	I-KEY
behavior	I-KEY
and	O
the	O
presence	O
of	O
redundant	B-KEY
solutions	I-KEY
for	O
print	O
devices	O
with	O
more	O
than	O
three	O
inks	O
.	O
Algorithms	O
for	O
the	O
conversion	O
of	O
image	O
data	O
to	O
printer-specific	O
format	O
are	O
typically	O
designed	O
to	O
achieve	O
a	O
single	O
predetermined	O
rendering	B-KEY
intent	I-KEY
,	O
such	O
as	O
colorimetric	O
accuracy	O
.	O
We	O
present	O
two	O
CIELAB	B-KEY
to	I-KEY
CMYK	I-KEY
color	I-KEY
conversion	I-KEY
schemes	I-KEY
based	O
on	O
a	O
general	O
Pareto-optimal	O
formulation	O
for	O
printer	O
color	O
management	O
.	O
The	O
schemes	O
operate	O
using	O
a	O
149-color	O
characterization	O
data	O
set	O
selected	O
to	O
efficiently	O
capture	O
the	O
entire	O
CMYK	O
gamut	O
.	O
The	O
first	O
scheme	O
uses	O
artificial	B-KEY
neural	I-KEY
networks	I-KEY
as	O
transfer	B-KEY
functions	I-KEY
between	O
the	O
CIELAB	O
and	O
CMYK	O
spaces	O
.	O
The	O
second	O
scheme	O
is	O
based	O
on	O
a	O
reformulation	O
of	O
tetrahedral	B-KEY
interpolation	I-KEY
as	O
an	O
optimization	B-KEY
problem	O
.	O
Characterization	O
data	O
are	O
divided	O
into	O
tetrahedra	B-KEY
for	O
the	O
interpolation-based	B-KEY
approach	I-KEY
using	O
the	O
program	O
Qhull	O
,	O
which	O
removes	O
the	O
common	O
restriction	O
that	O
characterization	O
data	O
be	O
well	O
organized	O
.	O
Both	O
schemes	O
offer	O
user	B-KEY
control	I-KEY
over	O
trade-off	O
problems	O
such	O
as	O
cost	B-KEY
versus	I-KEY
reproduction	I-KEY
accuracy	I-KEY
,	O
allowing	O
for	O
user-specified	B-KEY
print	I-KEY
objectives	I-KEY
and	O
the	O
use	O
of	O
constraints	B-KEY
such	O
as	O
maximum	B-KEY
allowable	I-KEY
ink	I-KEY
and	O
maximum	O
allowable	O
AE	O
*	O
/	O
sub	O
ab	O
/	O
.	O
A	O
formulation	O
for	O
minimization	O
of	O
ink	O
is	O
shown	O
to	O
be	O
particularly	O
favorable	O
,	O
integrating	O
both	O
a	O
clipping	B-KEY
and	O
gamut	B-KEY
compression	I-KEY
features	I-KEY
into	O
a	O
single	O
methodology	O

Multi-timescale	B-KEY
Internet	I-KEY
traffic	I-KEY
engineering	I-KEY
The	O
Internet	O
is	O
a	O
collection	O
of	O
packet-based	B-KEY
hop-by-hop	I-KEY
routed	I-KEY
networks	I-KEY
.	O
Internet	O
traffic	O
engineering	O
is	O
the	O
process	O
of	O
allocating	O
resources	O
to	O
meet	O
the	O
performance	O
requirements	O
of	O
users	O
and	O
operators	O
for	O
their	O
traffic	O
.	O
Current	O
mechanisms	O
for	O
doing	O
so	O
,	O
exemplified	O
by	O
TCP	O
's	O
congestion	O
control	O
or	O
the	O
variety	O
of	O
packet	B-KEY
marking	I-KEY
disciplines	I-KEY
,	O
concentrate	O
on	O
allocating	O
resources	O
on	O
a	O
per-packet	O
basis	O
or	O
at	O
data	O
timescales	O
.	O
This	O
article	O
motivates	O
the	O
need	O
for	O
traffic	O
engineering	O
in	O
the	O
Internet	O
at	O
other	O
timescales	O
,	O
namely	O
control	O
and	O
management	O
timescales	O
,	O
and	O
presents	O
three	O
mechanisms	O
for	O
this	O
.	O
It	O
also	O
presents	O
a	O
scenario	O
to	O
show	O
how	O
these	O
mechanisms	O
increase	O
the	O
flexibility	O
of	O
operators	O
'	O
service	O
offerings	O
and	O
potentially	O
also	O
ease	O
problems	O
of	O
Internet	B-KEY
management	I-KEY

Dynamic	O
neighborhood	O
structures	O
in	O
parallel	B-KEY
evolution	O
strategies	O
Parallelizing	B-KEY
is	O
a	O
straightforward	O
approach	O
to	O
reduce	O
the	O
total	O
computation	O
time	O
of	O
evolutionary	B-KEY
algorithms	I-KEY
.	O
Finding	O
an	O
appropriate	O
communication	O
network	O
within	O
spatially	O
structured	O
populations	O
for	O
improving	O
convergence	B-KEY
speed	I-KEY
and	O
convergence	B-KEY
probability	I-KEY
is	O
a	O
difficult	O
task	O
.	O
A	O
new	O
method	O
that	O
uses	O
a	O
dynamic	O
communication	O
scheme	O
in	O
an	O
evolution	O
strategy	O
will	O
be	O
compared	O
with	O
conventional	O
static	O
and	O
dynamic	O
approaches	O
.	O
The	O
communication	O
structure	O
is	O
based	O
on	O
a	O
so-called	O
diffusion	O
model	O
approach	O
.	O
The	O
links	O
between	O
adjacent	O
individuals	O
are	O
dynamically	O
chosen	O
according	O
to	O
deterministic	O
or	O
probabilistic	O
rules	O
.	O
Due	O
to	O
self-organization	O
effects	O
,	O
efficient	O
and	O
stable	O
communication	O
structures	O
are	O
established	O
that	O
perform	O
robustly	O
and	O
quickly	O
on	O
a	O
multimodal	B-KEY
test	I-KEY
function	I-KEY

Recruiting	O
and	O
retaining	O
women	O
in	O
undergraduate	B-KEY
computing	I-KEY
majors	I-KEY
This	O
paper	O
recommends	O
methods	O
for	O
increasing	O
female	B-KEY
participation	I-KEY
in	O
undergraduate	O
computer	B-KEY
science	I-KEY
.	O
The	O
recommendations	O
are	O
based	O
on	O
recent	O
and	O
on-going	O
research	O
into	O
the	O
gender	B-KEY
gap	I-KEY
in	O
computer	B-KEY
science	I-KEY
and	O
related	O
disciplines	O
.	O
They	O
are	O
intended	O
to	O
work	O
in	O
tandem	O
with	O
the	O
Computing	O
Research	O
Association	O
's	O
recommendations	O
for	O
graduate	O
programs	O
to	O
promote	O
a	O
general	O
increase	O
in	O
women	O
's	O
participation	O
in	O
computing	O
professions	O
.	O
Most	O
of	O
the	O
suggestions	O
offered	O
could	O
improve	O
the	O
educational	O
environment	O
for	O
both	O
male	O
and	O
female	O
students	O
.	O
However	O
,	O
general	O
improvements	O
are	O
likely	O
to	O
be	O
of	O
particular	O
benefit	O
to	O
women	O
because	O
women	O
in	O
our	O
society	O
do	O
not	O
generally	O
receive	O
the	O
same	O
level	O
of	O
support	O
that	O
men	O
receive	O
for	O
entering	O
and	O
persisting	O
in	O
this	O
field	O

The	O
changing	O
landscape	O
for	O
multi	O
access	O
portals	O
Discusses	O
the	O
factors	O
that	O
have	O
made	O
life	O
difficult	O
for	O
consumer	B-KEY
portal	I-KEY
operators	I-KEY
in	O
recent	O
years	O
causing	O
them	O
,	O
like	O
others	O
in	O
the	O
telecommunications	O
,	O
media	O
and	O
technology	O
sector	O
,	O
to	O
take	O
a	O
close	O
look	O
at	O
their	O
business	O
models	O
following	O
the	O
dot.com	O
crash	O
and	O
the	O
consequent	O
reassessment	O
of	O
Internet-related	O
project	O
financing	O
by	O
the	O
venture	O
capital	O
community	O
.	O
While	O
the	O
pressure	O
is	O
on	O
to	O
generate	O
income	O
from	O
existing	O
customers	O
and	O
users	O
,	O
portal	O
operators	O
must	O
reach	O
new	O
markets	O
and	O
find	O
realistic	O
revenue	B-KEY
streams	I-KEY
.	O
This	O
search	O
for	O
real	O
revenues	O
has	O
led	O
to	O
a	O
move	O
towards	O
charging	O
for	O
content	O
,	O
a	O
strategy	O
being	O
pursued	O
by	O
a	O
large	O
number	O
of	O
horizontal	O
portal	O
players	O
,	O
including	O
MSN	O
and	O
Terra	O
Lycos	O
.	O
This	O
trend	O
is	O
particularly	O
noticeable	O
in	O
China	O
,	O
where	O
Chinadotcom	O
operates	O
a	O
mainland	O
portal	O
and	O
plans	O
a	O
range	O
of	O
fee-based	B-KEY
services	I-KEY
,	O
including	O
electronic	O
mail	O
.	O
The	O
nature	O
of	O
advertising	B-KEY
itself	O
is	O
changing	O
,	O
with	O
portals	O
seeking	O
blue-chip	B-KEY
sponsorship	I-KEY
and	O
marketing	O
deals	O
that	O
span	O
a	O
number	O
of	O
years	O
.	O
Players	O
are	O
struggling	O
to	O
redefine	O
and	O
reinvent	O
themselves	O
as	O
a	O
result	O
of	O
the	O
changing	O
environment	O
and	O
even	O
the	O
term	O
``	O
portal	O
''	O
is	O
believed	O
to	O
be	O
obsolete	O
,	O
partly	O
due	O
to	O
its	O
dot.com	O
crash	O
associations	O
.	O
Multi-access	B-KEY
portals	I-KEY
are	O
expected	O
to	O
dominate	O
the	O
consumer	O
sector	O
,	O
becoming	O
bigger	O
and	O
better	O
overall	O
than	O
their	O
predecessors	O
and	O
playing	O
a	O
more	O
powerful	O
role	O
in	O
the	O
consumer	O
environment	O

Rise	O
of	O
the	O
supercompany	O
-LSB-	O
CRM	O
-RSB-	O
All	O
the	O
thoughts	O
,	O
conversations	O
and	O
notes	O
of	O
employees	O
help	O
the	O
firm	O
create	O
a	O
wider	O
picture	O
of	O
business	O
.	O
Customer	B-KEY
relationship	I-KEY
management	I-KEY
-LRB-	O
CRM	O
-RRB-	O
feeds	O
on	O
data	O
,	O
and	O
it	O
is	O
hungry	O

Source/channel	O
coding	O
of	O
still	B-KEY
images	I-KEY
using	O
lapped	B-KEY
transforms	I-KEY
and	O
block	B-KEY
classification	I-KEY
A	O
novel	O
scheme	O
for	O
joint	O
source/channel	O
coding	O
of	O
still	B-KEY
images	I-KEY
is	O
proposed	O
.	O
By	O
using	O
efficient	O
lapped	B-KEY
transforms	I-KEY
,	O
channel-optimised	B-KEY
robust	I-KEY
quantisers	I-KEY
and	O
classification	O
methods	O
it	O
is	O
shown	O
that	O
significant	O
improvements	O
over	O
traditional	O
source/channel	O
coding	O
of	O
images	O
can	O
be	O
obtained	O
while	O
keeping	O
the	O
complexity	O
low	O

Dynamics	O
of	O
the	O
firing	B-KEY
probability	I-KEY
of	O
noisy	B-KEY
integrate-and-fire	I-KEY
neurons	I-KEY
Cortical	B-KEY
neurons	I-KEY
in	O
vivo	O
undergo	O
a	O
continuous	O
bombardment	O
due	O
to	O
synaptic	B-KEY
activity	I-KEY
,	O
which	O
acts	O
as	O
a	O
major	O
source	O
of	O
noise	O
.	O
We	O
investigate	O
the	O
effects	O
of	O
the	O
noise	B-KEY
filtering	I-KEY
by	O
synapses	O
with	O
various	O
levels	O
of	O
realism	O
on	O
integrate-and-fire	O
neuron	O
dynamics	O
.	O
The	O
noise	O
input	O
is	O
modeled	O
by	O
white	O
-LRB-	O
for	O
instantaneous	O
synapses	O
-RRB-	O
or	O
colored	O
-LRB-	O
for	O
synapses	O
with	O
a	O
finite	O
relaxation	O
time	O
-RRB-	O
noise	O
.	O
Analytical	O
results	O
for	O
the	O
modulation	O
of	O
firing	B-KEY
probability	I-KEY
in	O
response	O
to	O
an	O
oscillatory	O
input	O
current	O
are	O
obtained	O
by	O
expanding	O
a	O
Fokker-Planck	B-KEY
equation	I-KEY
for	O
small	O
parameters	O
of	O
the	O
problem-when	O
both	O
the	O
amplitude	O
of	O
the	O
modulation	O
is	O
small	O
compared	O
to	O
the	O
background	O
firing	O
rate	O
and	O
the	O
synaptic	B-KEY
time	I-KEY
constant	I-KEY
is	O
small	O
compared	O
to	O
the	O
membrane	B-KEY
time	I-KEY
constant	I-KEY
.	O
We	O
report	O
the	O
detailed	O
calculations	O
showing	O
that	O
if	O
a	O
synaptic	O
decay	O
time	O
constant	O
is	O
included	O
in	O
the	O
synaptic	O
current	O
model	O
,	O
the	O
firing-rate	O
modulation	O
of	O
the	O
neuron	O
due	O
to	O
an	O
oscillatory	O
input	O
remains	O
finite	O
in	O
the	O
high-frequency	O
limit	O
with	O
no	O
phase	B-KEY
lag	I-KEY
.	O
In	O
addition	O
,	O
we	O
characterize	O
the	O
low-frequency	O
behavior	O
and	O
the	O
behavior	O
of	O
the	O
high-frequency	O
limit	O
for	O
intermediate	O
decay	O
times	O
.	O
We	O
also	O
characterize	O
the	O
effects	O
of	O
introducing	O
a	O
rise	O
time	O
to	O
the	O
synaptic	O
currents	O
and	O
the	O
presence	O
of	O
several	O
synaptic	B-KEY
receptors	I-KEY
with	O
different	O
kinetics	O
.	O
In	O
both	O
cases	O
,	O
we	O
determine	O
,	O
using	O
numerical	B-KEY
simulations	I-KEY
,	O
an	O
effective	O
decay	O
time	O
constant	O
that	O
describes	O
the	O
neuronal	O
response	O
completely	O

Quantum	B-KEY
computation	I-KEY
for	O
physical	B-KEY
modeling	I-KEY
One	O
of	O
the	O
most	O
famous	O
American	O
physicists	O
of	O
the	O
twentieth	O
century	O
,	O
Richard	O
Feynman	O
,	O
in	O
1982	O
was	O
the	O
first	O
to	O
propose	O
using	O
a	O
quantum	B-KEY
mechanical	I-KEY
computing	I-KEY
device	O
to	O
efficiently	O
simulate	O
quantum	B-KEY
mechanical	I-KEY
many-body	I-KEY
dynamics	I-KEY
,	O
a	O
task	O
that	O
is	O
exponentially	O
complex	O
in	O
the	O
number	O
of	O
particles	O
treated	O
and	O
is	O
completely	O
intractable	O
by	O
any	O
classical	O
computing	O
means	O
for	O
large	O
systems	O
of	O
many	O
particles	O
.	O
In	O
the	O
two	O
decades	O
following	O
his	O
work	O
,	O
remarkable	O
progress	O
has	O
been	O
made	O
both	O
theoretically	O
and	O
experimentally	O
in	O
the	O
new	O
field	O
of	O
quantum	B-KEY
computation	I-KEY

Development	O
and	O
evaluation	O
of	O
a	O
case-based	B-KEY
reasoning	I-KEY
classifier	I-KEY
for	O
prediction	O
of	O
breast	B-KEY
biopsy	I-KEY
outcome	I-KEY
with	O
BI-RADS/sup	O
TM	O
/	O
lexicon	O
Approximately	O
70-85	O
%	O
of	O
breast	O
biopsies	O
are	O
performed	O
on	O
benign	B-KEY
lesions	I-KEY
.	O
To	O
reduce	O
this	O
high	O
number	O
of	O
biopsies	O
performed	O
on	O
benign	B-KEY
lesions	I-KEY
,	O
a	O
case-based	O
reasoning	O
-LRB-	O
CBR	O
-RRB-	O
classifier	O
was	O
developed	O
to	O
predict	O
biopsy	O
results	O
from	O
BI-RADS/sup	O
TM	O
/	O
findings	O
.	O
We	O
used	O
1433	O
-LRB-	O
931	O
benign	O
-RRB-	O
biopsy-proven	B-KEY
mammographic	I-KEY
cases	I-KEY
.	O
CBR	B-KEY
similarity	I-KEY
was	O
defined	O
using	O
either	O
the	O
Hamming	O
or	O
Euclidean	B-KEY
distance	I-KEY
measure	I-KEY
over	O
case	O
features	O
.	O
Ten	O
features	O
represented	O
each	O
case	O
:	O
calcification	B-KEY
distribution	I-KEY
,	O
calcification	B-KEY
morphology	I-KEY
,	O
calcification	B-KEY
number	I-KEY
,	O
mass	B-KEY
margin	I-KEY
,	O
mass	B-KEY
shape	I-KEY
,	O
mass	B-KEY
density	I-KEY
,	O
mass	B-KEY
size	I-KEY
,	O
associated	B-KEY
findings	I-KEY
,	O
special	B-KEY
cases	I-KEY
,	O
and	O
age	B-KEY
.	O
Performance	O
was	O
evaluated	O
using	O
Round	B-KEY
Robin	I-KEY
sampling	I-KEY
,	O
Receiver	O
Operating	O
Characteristic	O
-LRB-	O
ROC	O
-RRB-	O
analysis	O
,	O
and	O
bootstrap	B-KEY
.	O
To	O
determine	O
the	O
most	O
influential	B-KEY
features	I-KEY
for	O
the	O
CBR	O
,	O
an	O
exhaustive	O
feature	O
search	O
was	O
performed	O
over	O
all	O
possible	O
feature	B-KEY
combinations	I-KEY
-LRB-	O
1022	O
-RRB-	O
and	O
similarity	B-KEY
thresholds	I-KEY
.	O
Influential	B-KEY
features	I-KEY
were	O
defined	O
as	O
the	O
most	O
frequently	O
occurring	O
features	O
in	O
the	O
feature	B-KEY
subsets	I-KEY
with	O
the	O
highest	B-KEY
partial	I-KEY
ROC	I-KEY
areas	I-KEY
-LRB-	O
/	O
sub	O
0.90	O
/	O
AUC	O
-RRB-	O
.	O
For	O
CBR	O
with	O
Hamming	O
distance	O
,	O
the	O
most	O
influential	B-KEY
features	I-KEY
were	O
found	O
to	O
be	O
mass	B-KEY
margin	I-KEY
,	O
calcification	B-KEY
morphology	I-KEY
,	O
age	B-KEY
,	O
calcification	B-KEY
distribution	I-KEY
,	O
calcification	B-KEY
number	I-KEY
,	O
and	O
mass	B-KEY
shape	I-KEY
,	O
resulting	O
in	O
an	O
/	O
sub	O
0.90	O
/	O
AUC	O
of	O
0.33	O
.	O
At	O
95	O
%	O
sensitivity	O
,	O
the	O
Hamming	O
CBR	O
would	O
spare	O
from	O
biopsy	O
34	O
%	O
of	O
the	O
benign	B-KEY
lesions	I-KEY
.	O
At	O
98	O
%	O
sensitivity	O
,	O
the	O
Hamming	O
CBR	O
would	O
spare	O
27	O
%	O
benign	B-KEY
lesions	I-KEY
.	O
For	O
the	O
CBR	O
with	O
Euclidean	O
distance	O
,	O
the	O
most	O
influential	B-KEY
feature	I-KEY
subset	O
consisted	O
of	O
mass	O
margin	O
,	O
calcification	O
morphology	O
,	O
age	O
,	O
mass	O
density	O
,	O
and	O
associated	O
findings	O
,	O
resulting	O
in	O
/	O
sub	O
0.90	O
/	O
AUC	O
of	O
0.37	O
.	O
At	O
95	O
%	O
sensitivity	O
,	O
the	O
Euclidean	O
CBR	O
would	O
spare	O
from	O
biopsy	O
41	O
%	O
benign	B-KEY
lesions	I-KEY
.	O
At	O
98	O
%	O
sensitivity	O
,	O
the	O
Euclidean	O
CBR	O
would	O
spare	O
27	O
%	O
benign	B-KEY
lesions	I-KEY
.	O
The	O
profile	O
of	O
cases	O
spared	O
by	O
both	O
distance	O
measures	O
at	O
98	O
%	O
sensitivity	O
indicates	O
that	O
the	O
CBR	O
is	O
a	O
potentially	O
useful	O
diagnostic	B-KEY
tool	I-KEY
for	O
the	O
classification	O
of	O
mammographic	O
lesions	O
,	O
by	O
recommending	O
short-term	B-KEY
follow-up	I-KEY
for	O
likely	O
benign	B-KEY
lesions	I-KEY
that	O
is	O
in	O
agreement	O
with	O
final	O
biopsy	O
results	O
and	O
mammographer	O
's	O
intuition	O

A	O
fuzzy	B-KEY
logic	I-KEY
approach	I-KEY
to	O
accommodate	O
thermal	O
stress	O
and	O
improve	O
the	O
start-up	O
phase	O
in	O
combined	B-KEY
cycle	I-KEY
power	I-KEY
plants	I-KEY
Use	O
of	O
combined	O
cycle	O
power	O
generation	O
plant	O
has	O
increased	O
dramatically	O
over	O
the	O
last	O
decade	O
.	O
A	O
supervisory	B-KEY
control	I-KEY
approach	O
based	O
on	O
a	O
dynamic	B-KEY
model	I-KEY
is	O
developed	O
,	O
which	O
makes	O
use	O
of	O
proportional-integral-derivative	O
-LRB-	O
PID	O
-RRB-	O
,	O
fuzzy	O
logic	O
and	O
fuzzy	B-KEY
PID	I-KEY
schemes	I-KEY
.	O
The	O
aim	O
is	O
to	O
minimize	O
the	O
steam	O
turbine	O
plant	O
start-up	O
time	O
,	O
without	O
violating	O
maximum	B-KEY
thermal	I-KEY
stress	I-KEY
limits	I-KEY
.	O
An	O
existing	O
start-up	B-KEY
schedule	I-KEY
provides	O
the	O
benchmark	O
by	O
which	O
the	O
performance	O
of	O
candidate	O
controllers	O
is	O
assessed	O
.	O
Improvements	O
regarding	O
possible	O
reduced	O
start-up	O
times	O
and	O
satisfaction	O
of	O
maximum	O
thermal	O
stress	O
restrictions	O
have	O
been	O
realized	O
using	O
the	O
proposed	O
control	O
scheme	O

Controls	O
help	O
harmonic	B-KEY
spray	I-KEY
do	O
OK	O
removing	O
residues	O
Looks	O
at	O
how	O
innovative	O
wafer-cleaning	B-KEY
equipment	I-KEY
hit	O
the	O
market	O
in	O
a	O
timely	O
fashion	O
thanks	O
in	O
part	O
to	O
controls	O
maker	O
Rockwell	B-KEY
Automation	I-KEY

Application	O
foundations	O
-LSB-	O
application	B-KEY
servers	I-KEY
-RSB-	O
The	O
changing	O
role	O
of	O
application	B-KEY
servers	I-KEY
means	O
choosing	O
the	O
right	O
platform	O
has	O
become	O
a	O
complex	O
challenge	O

Adaptive	O
optimizing	B-KEY
compilers	O
for	O
the	O
21st	O
century	O
Historically	O
,	O
compilers	B-KEY
have	O
operated	O
by	O
applying	O
a	O
fixed	O
set	O
of	O
optimizations	B-KEY
in	O
a	O
predetermined	O
order	O
.	O
We	O
call	O
such	O
an	O
ordered	O
list	O
of	O
optimizations	B-KEY
a	O
compilation	B-KEY
sequence	O
.	O
This	O
paper	O
describes	O
a	O
prototype	O
system	O
that	O
uses	O
biased	B-KEY
random	I-KEY
search	I-KEY
to	O
discover	O
a	O
program-specific	O
compilation	B-KEY
sequence	O
that	O
minimizes	O
an	O
explicit	O
,	O
external	O
objective	O
function	O
.	O
The	O
result	O
is	O
a	O
compiler	B-KEY
framework	O
that	O
adapts	O
its	O
behavior	O
to	O
the	O
application	O
being	O
compiled	B-KEY
,	O
to	O
the	O
pool	O
of	O
available	O
transformations	O
,	O
to	O
the	O
objective	O
function	O
,	O
and	O
to	O
the	O
target	O
machine	O
.	O
This	O
paper	O
describes	O
experiments	O
that	O
attempt	O
to	O
characterize	O
the	O
space	O
that	O
the	O
adaptive	B-KEY
compiler	I-KEY
must	O
search	O
.	O
The	O
preliminary	O
results	O
suggest	O
that	O
optimal	B-KEY
solutions	O
are	O
rare	O
and	O
that	O
local	O
minima	O
are	O
frequent	O
.	O
If	O
this	O
holds	O
true	O
,	O
biased	B-KEY
random	I-KEY
searches	I-KEY
,	O
such	O
as	O
a	O
,	O
genetic	O
algorithm	O
,	O
should	O
find	O
good	O
solutions	O
more	O
quickly	O
than	O
simpler	O
strategies	O
,	O
such	O
as	O
hill	O
climbing	O

Multi-hour	B-KEY
design	I-KEY
of	O
survivable	B-KEY
classical	I-KEY
IP	I-KEY
networks	I-KEY
Most	O
of	O
Internet	B-KEY
intra-domain	I-KEY
routing	I-KEY
protocols	I-KEY
-LRB-	O
OSPF	B-KEY
,	O
RIP	B-KEY
,	O
and	O
IS-IS	B-KEY
-RRB-	O
are	O
based	O
on	O
shortest	B-KEY
path	I-KEY
routing	I-KEY
.	O
The	O
path	B-KEY
length	I-KEY
is	O
defined	O
as	O
the	O
sum	O
of	O
metrics	O
associated	O
with	O
the	O
path	B-KEY
links	I-KEY
.	O
These	O
metrics	O
are	O
often	O
managed	O
by	O
the	O
network	B-KEY
administrator	I-KEY
.	O
In	O
this	O
context	O
,	O
the	O
design	O
of	O
an	O
Internet	B-KEY
backbone	I-KEY
network	I-KEY
consists	O
in	O
dimensioning	O
the	O
network	O
-LRB-	O
routers	O
and	O
transmission	B-KEY
links	I-KEY
-RRB-	O
and	O
establishing	O
the	O
metric	O
.	O
Many	O
requirements	O
have	O
to	O
be	O
satisfied	O
.	O
First	O
,	O
Internet	B-KEY
traffic	I-KEY
is	O
not	O
static	O
as	O
significant	O
variations	O
can	O
be	O
observed	O
during	O
the	O
day	O
.	O
Second	O
,	O
many	O
failures	O
can	O
occur	O
-LRB-	O
cable	O
cuts	O
,	O
hardware	O
failures	O
,	O
software	O
failures	O
,	O
etc.	O
-RRB-	O
.	O
We	O
present	O
algorithms	O
-LRB-	O
meta-heuristics	O
and	O
greedy	O
heuristic	O
-RRB-	O
to	O
design	O
Internet	B-KEY
backbone	I-KEY
networks	I-KEY
,	O
taking	O
into	O
account	O
the	O
multi-hour	O
behaviour	O
of	O
traffic	O
and	O
some	O
survivability	B-KEY
requirements	I-KEY
.	O
Many	O
multi-hour	O
and	O
protection	O
strategies	O
are	O
studied	O
and	O
numerically	O
compared	O
.	O
Our	O
algorithms	O
can	O
be	O
extended	O
to	O
integrate	O
other	O
quality	B-KEY
of	I-KEY
service	I-KEY
constraints	I-KEY

Comparison	O
of	O
push	O
and	O
pull	B-KEY
systems	I-KEY
with	O
transporters	B-KEY
:	O
a	O
metamodelling	B-KEY
approach	I-KEY
Analyses	O
push	O
and	O
pull	B-KEY
systems	I-KEY
with	O
transportation	B-KEY
consideration	O
.	O
A	O
multiproduct	O
,	O
multiline	O
,	O
multistage	O
production	O
system	O
was	O
used	O
to	O
compare	O
the	O
two	O
systems	O
.	O
The	O
effects	O
of	O
four	O
factors	O
-LRB-	O
processing	B-KEY
time	I-KEY
variation	I-KEY
,	O
demand	B-KEY
variation	I-KEY
,	O
transporters	B-KEY
,	O
batch	B-KEY
size	I-KEY
-RRB-	O
on	O
throughput	B-KEY
rate	I-KEY
,	O
average	B-KEY
waiting	I-KEY
time	I-KEY
in	O
the	O
system	O
and	O
machine	B-KEY
utilization	I-KEY
were	O
studied	O
.	O
The	O
study	O
uses	O
metamodels	O
to	O
compare	O
the	O
two	O
systems	O
.	O
They	O
serve	O
a	O
dual	O
purpose	O
of	O
expressing	O
system	O
performance	B-KEY
measures	I-KEY
in	O
the	O
form	O
of	O
a	O
simple	O
equation	O
and	O
reducing	O
computational	O
time	O
when	O
comparing	O
the	O
two	O
systems	O
.	O
Research	O
shows	O
that	O
the	O
number	O
of	O
transporters	B-KEY
used	O
and	O
the	O
batch	B-KEY
size	I-KEY
have	O
a	O
significant	O
effect	O
on	O
the	O
performance	B-KEY
measures	I-KEY
of	O
both	O
systems	O

Fast	O
and	O
accurate	B-KEY
leaf	I-KEY
verification	I-KEY
for	O
dynamic	B-KEY
multileaf	I-KEY
collimation	I-KEY
using	O
an	O
electronic	B-KEY
portal	I-KEY
imaging	I-KEY
device	I-KEY
A	O
prerequisite	O
for	O
accurate	B-KEY
dose	I-KEY
delivery	I-KEY
of	O
IMRT	O
profiles	O
produced	O
with	O
dynamic	B-KEY
multileaf	I-KEY
collimation	I-KEY
-LRB-	O
DMLC	O
-RRB-	O
is	O
highly	O
accurate	O
leaf	B-KEY
positioning	I-KEY
.	O
In	O
our	O
institution	O
,	O
leaf	O
verification	O
for	O
DMLC	O
was	O
initially	O
done	O
with	O
film	O
and	O
ionization	B-KEY
chamber	I-KEY
.	O
To	O
overcome	O
the	O
limitations	O
of	O
these	O
methods	O
,	O
a	O
fast	O
,	O
accurate	O
and	O
two-dimensional	B-KEY
method	I-KEY
for	O
daily	O
leaf	O
verification	O
,	O
using	O
our	O
CCD-camera	O
based	O
electronic	B-KEY
portal	I-KEY
imaging	I-KEY
device	I-KEY
-LRB-	O
EPID	O
-RRB-	O
,	O
has	O
been	O
developed	O
.	O
This	O
method	O
is	O
based	O
on	O
a	O
flat	O
field	O
produced	O
with	O
a	O
0.5	O
cm	O
wide	O
sliding	B-KEY
gap	I-KEY
for	O
each	O
leaf	B-KEY
pair	I-KEY
.	O
Deviations	O
in	O
gap	B-KEY
widths	I-KEY
are	O
detected	O
as	O
deviations	O
in	O
gray	B-KEY
scale	I-KEY
value	I-KEY
profiles	I-KEY
derived	O
from	O
the	O
EPID	O
images	O
,	O
and	O
not	O
by	O
directly	O
assessing	O
leaf	B-KEY
positions	I-KEY
in	O
the	O
images	O
.	O
Dedicated	O
software	O
was	O
developed	O
to	O
reduce	O
the	O
noise	B-KEY
level	I-KEY
in	O
the	O
low	O
signal	B-KEY
images	I-KEY
produced	O
with	O
the	O
narrow	O
gaps	O
.	O
The	O
accuracy	O
of	O
this	O
quality	O
assurance	O
procedure	O
was	O
tested	O
by	O
introducing	O
known	O
leaf	B-KEY
position	I-KEY
errors	O
.	O
It	O
was	O
shown	O
that	O
errors	O
in	O
leaf	O
gap	O
as	O
small	O
as	O
0.01-0	O
.02	O
cm	O
could	O
be	O
detected	O
,	O
which	O
is	O
certainly	O
adequate	O
to	O
guarantee	O
accurate	B-KEY
dose	I-KEY
delivery	I-KEY
of	O
DMLC	O
treatments	O
,	O
even	O
for	O
strongly	O
modulated	B-KEY
beam	I-KEY
profiles	I-KEY
.	O
Using	O
this	O
method	O
,	O
it	O
was	O
demonstrated	O
that	O
both	O
short	O
and	O
long	O
term	O
reproducibility	O
in	O
leaf	B-KEY
positioning	I-KEY
were	O
within	O
0.01	O
cm	O
-LRB-	O
1	O
sigma	O
-RRB-	O
for	O
all	O
gantry	B-KEY
angles	I-KEY
,	O
and	O
that	O
the	O
effect	O
of	O
gravity	O
was	O
negligible	O

The	O
curious	O
ways	O
of	O
professional	B-KEY
cultures	I-KEY
and	O
the	O
``	O
two-body	B-KEY
opportunity	I-KEY
''	O
When	O
two	O
professionals	O
are	O
a	O
couple	O
,	O
we	O
sometimes	O
refer	O
to	O
them	O
as	O
having	O
a	O
``	O
two-body	O
problem	O
.	O
''	O
However	O
,	O
when	O
each	O
partner	O
of	O
a	O
couple	O
exists	O
in	O
the	O
same	O
cultures	O
,	O
they	O
also	O
have	O
an	O
opportunity	O
for	O
deeply	O
shared	O
understanding	O
and	O
empathy	O
,	O
simply	O
because	O
each	O
understands	O
at	O
a	O
deep	O
level	O
the	O
culture	O
in	O
which	O
the	O
other	O
works	O
.	O
I	O
explore	O
this	O
notion	O
.	O
A	O
couple	O
has	O
what	O
we	O
call	O
the	O
``	O
two-body	O
problem	O
''	O
when	O
both	O
are	O
professionals	O
who	O
are	O
qualified	O
for	O
a	O
kind	O
of	O
position	O
that	O
is	O
relatively	O
rare	O
and	O
who	O
are	O
very	O
selective	O
about	O
the	O
positions	O
that	O
they	O
accept	O
.	O
For	O
example	O
,	O
there	O
are	O
relatively	O
scant	O
numbers	O
of	O
jobs	O
as	O
a	O
computer	B-KEY
science	I-KEY
professor-at	O
any	O
level	O
.	O
An	O
individual	O
considering	O
an	O
academic	B-KEY
job	O
may	O
only	O
be	O
interested	O
in	O
research	O
universities	O
,	O
or	O
in	O
teaching	O
universities	O
,	O
restricting	O
the	O
choice	O
of	O
open	O
positions	O
substantially	O
.	O
The	O
classic	O
two-body	O
``	O
problem	O
''	O
arises	O
when	O
one	O
partner	O
wants	O
to	O
accept	O
a	O
new	O
position	O
that	O
requires	O
geographical	O
relocation	O
.	O
Then	O
,	O
the	O
other	O
partner	O
also	O
needs	O
to	O
find	O
a	O
new	O
position	O
.	O
Moreover	O
,	O
it	O
can	O
be	O
very	O
difficult	O
to	O
find	O
a	O
suitable	O
position	O
when	O
they	O
are	O
naturally	O
scarce	O

Active	O
vibration	O
control	O
of	O
composite	O
sandwich	O
beams	O
with	O
piezoelectric	B-KEY
extension-bending	O
and	O
shear	B-KEY
actuators	I-KEY
We	O
have	O
used	O
quasi-static	B-KEY
equations	I-KEY
of	O
piezoelectricity	B-KEY
to	O
derive	O
a	O
finite	B-KEY
element	I-KEY
formulation	I-KEY
capable	O
of	O
modelling	O
two	O
different	O
kinds	O
of	O
piezoelastically	B-KEY
induced	O
actuation	O
in	O
an	O
adaptive	B-KEY
composite	I-KEY
sandwich	I-KEY
beam	I-KEY
.	O
This	O
formulation	O
is	O
made	O
to	O
couple	O
certain	O
piezoelectric	B-KEY
constants	O
to	O
a	O
transverse	O
electric	O
field	O
to	O
develop	O
extension-bending	O
actuation	O
and	O
shear-induced	O
actuation	O
.	O
As	O
an	O
illustration	O
,	O
we	O
present	O
a	O
sandwich	B-KEY
model	I-KEY
of	O
three	O
sublaminates	B-KEY
:	O
face/core/face	O
.	O
We	O
develop	O
a	O
control	O
scheme	O
based	O
on	O
the	O
linear	O
quadratic	O
regulator/independent	O
modal	B-KEY
space	I-KEY
control	I-KEY
-LRB-	O
LQR/IMSC	O
-RRB-	O
method	O
and	O
use	O
this	O
to	O
estimate	O
the	O
active	B-KEY
stiffness	I-KEY
and	O
the	O
active	B-KEY
damping	I-KEY
introduced	O
by	O
shear	O
and	O
extension-bending	B-KEY
actuators	I-KEY
.	O
To	O
assess	O
the	O
performance	O
of	O
each	O
type	O
of	O
actuator	O
,	O
a	O
dynamic	B-KEY
response	I-KEY
study	O
is	O
carried	O
out	O
in	O
the	O
modal	B-KEY
domain	I-KEY
.	O
We	O
observe	O
that	O
the	O
shear	B-KEY
actuator	I-KEY
is	O
more	O
efficient	O
in	O
actively	O
controlling	O
the	O
vibration	O
than	O
the	O
extension-bending	B-KEY
actuator	I-KEY
for	O
the	O
same	O
control	O
effort	O

Conditions	O
for	O
decentralized	B-KEY
integral	I-KEY
controllability	I-KEY
The	O
term	O
decentralized	B-KEY
integral	I-KEY
controllability	I-KEY
-LRB-	O
DIC	O
-RRB-	O
pertains	O
to	O
the	O
existence	O
of	O
stable	B-KEY
decentralized	I-KEY
controllers	I-KEY
with	O
integral	B-KEY
action	I-KEY
that	O
have	O
closed-loop	B-KEY
properties	I-KEY
such	O
as	O
stable	B-KEY
independent	I-KEY
detuning	I-KEY
.	O
It	O
is	O
especially	O
useful	O
to	O
select	O
control	O
structures	O
systematically	O
at	O
the	O
early	O
stage	O
of	O
control	B-KEY
system	I-KEY
design	I-KEY
because	O
the	O
only	O
information	O
needed	O
for	O
DIC	O
is	O
the	O
steady-state	B-KEY
process	I-KEY
gain	I-KEY
matrix	I-KEY
.	O
Here	O
,	O
a	O
necessary	O
and	O
sufficient	O
condition	O
conjectured	O
in	O
the	O
literature	O
is	O
proved	O
.	O
The	O
real	B-KEY
structured	I-KEY
singular	I-KEY
value	I-KEY
which	O
can	O
exploit	O
realness	O
of	O
the	O
controller	O
gain	O
is	O
used	O
to	O
describe	O
computable	O
conditions	O
for	O
DIC	O
.	O
The	O
primary	O
usage	O
of	O
DIC	O
is	O
to	O
eliminate	O
unworkable	O
pairings	O
.	O
For	O
this	O
,	O
two	O
other	O
simple	O
necessary	O
conditions	O
are	O
proposed	O
.	O
Examples	O
are	O
given	O
to	O
illustrate	O
the	O
effectiveness	O
of	O
the	O
proposed	O
conditions	O
for	O
DIC	O

A	O
discontinuous	O
Galerkin	O
method	O
for	O
transient	B-KEY
analysis	I-KEY
of	O
wave	O
propagation	O
in	O
unbounded	B-KEY
domains	I-KEY
A	O
technique	O
based	O
on	O
the	O
discontinuous	B-KEY
Galerkin	I-KEY
finite	I-KEY
element	I-KEY
method	I-KEY
is	O
developed	O
and	O
applied	O
to	O
the	O
derivation	O
of	O
an	O
absorbing	B-KEY
boundary	I-KEY
condition	I-KEY
for	O
the	O
analysis	O
of	O
transient	B-KEY
wave	I-KEY
propagation	I-KEY
.	O
The	O
condition	O
is	O
exact	O
in	O
that	O
only	O
discretization	B-KEY
error	I-KEY
is	O
involved	O
.	O
Furthermore	O
,	O
the	O
computational	B-KEY
cost	I-KEY
associated	O
with	O
use	O
of	O
the	O
condition	O
is	O
an	O
order	O
of	O
magnitude	O
lower	O
than	O
for	O
conditions	O
based	O
on	O
Green	O
functions	O
.	O
The	O
time-stepping	B-KEY
scheme	I-KEY
resulting	O
from	O
an	O
implicit	B-KEY
method	I-KEY
in	O
conjunction	O
with	O
this	O
boundary	O
condition	O
appears	O
to	O
be	O
unconditionally	O
stable	O

A	O
linear	B-KEY
time	I-KEY
special	I-KEY
case	I-KEY
for	O
MC	B-KEY
games	I-KEY
MC	B-KEY
games	I-KEY
are	O
infinite	O
duration	O
two-player	B-KEY
games	I-KEY
played	O
on	O
graphs	O
.	O
Deciding	O
the	O
winner	O
in	O
MC	B-KEY
games	I-KEY
is	O
equivalent	O
to	O
the	O
the	O
modal	B-KEY
mu-calculus	I-KEY
model	I-KEY
checking	I-KEY
.	O
In	O
this	O
article	O
we	O
provide	O
a	O
linear	B-KEY
time	I-KEY
algorithm	I-KEY
for	O
a	O
class	O
of	O
MC	B-KEY
games	I-KEY
.	O
We	O
show	O
that	O
,	O
if	O
all	O
cycles	O
in	O
each	O
strongly	O
connected	O
component	O
of	O
the	O
game	O
graph	O
have	O
at	O
least	O
one	O
common	O
vertex	O
,	O
the	O
winner	O
can	O
be	O
found	O
in	O
linear	O
time	O
.	O
Our	O
results	O
hold	O
also	O
for	O
parity	O
games	O
,	O
which	O
are	O
equivalent	O
to	O
MC	B-KEY
games	I-KEY

Applying	O
genetic	B-KEY
algorithms	I-KEY
to	O
solve	O
the	O
fuzzy	B-KEY
optimal	I-KEY
profit	I-KEY
problem	I-KEY
This	O
study	O
investigated	O
the	O
application	O
of	O
genetic	B-KEY
algorithms	I-KEY
in	O
solving	O
a	O
fuzzy	B-KEY
optimization	I-KEY
problem	I-KEY
that	O
arises	O
in	O
business	B-KEY
and	O
economics	B-KEY
.	O
In	O
this	O
problem	O
,	O
a	O
fuzzy	B-KEY
price	I-KEY
is	O
determined	O
using	O
a	O
linear	O
or	O
a	O
quadratic	B-KEY
fuzzy	I-KEY
demand	I-KEY
function	I-KEY
as	O
well	O
as	O
a	O
linear	B-KEY
cost	I-KEY
function	I-KEY
.	O
The	O
objective	O
is	O
to	O
find	O
the	O
optimal	O
fuzzy	O
profit	O
,	O
which	O
is	O
derived	O
from	O
the	O
fuzzy	B-KEY
price	I-KEY
and	O
fuzzy	O
cost	O
.	O
Traditional	O
methods	O
for	O
solving	O
this	O
problem	O
are	O
-LRB-	O
1	O
-RRB-	O
the	O
extension	O
principle	O
,	O
and	O
-LRB-	O
2	O
-RRB-	O
using	O
interval	O
arithmetic	O
and	O
alpha	O
-	O
cuts	O
.	O
However	O
,	O
we	O
argue	O
that	O
traditional	O
methods	O
for	O
solving	O
this	O
problem	O
are	O
too	O
restrictive	O
to	O
produce	O
an	O
optimal	O
solution	O
,	O
and	O
that	O
an	O
alternative	O
approach	O
is	O
possibly	O
needed	O
.	O
We	O
use	O
genetic	B-KEY
algorithms	I-KEY
to	O
obtain	O
an	O
approximate	B-KEY
solution	I-KEY
for	O
this	O
fuzzy	B-KEY
optimal	I-KEY
profit	I-KEY
problem	I-KEY
without	O
using	O
membership	O
functions	O
.	O
We	O
not	O
only	O
give	O
empirical	O
examples	O
to	O
show	O
the	O
effectiveness	O
of	O
this	O
approach	O
,	O
but	O
also	O
give	O
theoretical	B-KEY
proofs	I-KEY
to	O
validate	O
correctness	O
of	O
the	O
algorithm	O
.	O
We	O
conclude	O
that	O
genetic	B-KEY
algorithms	I-KEY
can	O
produce	O
good	O
approximate	B-KEY
solutions	I-KEY
when	O
applied	O
to	O
solve	O
fuzzy	B-KEY
optimization	I-KEY
problems	I-KEY

A	O
dynamic	B-KEY
checkpoint	I-KEY
scheduling	I-KEY
scheme	I-KEY
for	O
fault	O
tolerant	O
distributed	B-KEY
computing	I-KEY
systems	I-KEY
The	O
selection	O
of	O
the	O
optimal	B-KEY
checkpointing	I-KEY
interval	I-KEY
has	O
been	O
a	O
very	O
critical	O
issue	O
in	O
implementing	O
checkpointing-recovery	O
schemes	O
for	O
fault	O
tolerant	O
distributed	O
systems	O
.	O
This	O
paper	O
presents	O
a	O
new	O
scheme	O
that	O
allows	O
a	O
process	O
to	O
select	O
the	O
proper	O
checkpointing	O
interval	O
dynamically	O
.	O
A	O
process	O
in	O
the	O
system	O
evaluates	O
the	O
cost	O
of	O
checkpointing	O
and	O
possible	O
rollback	O
for	O
each	O
checkpointing	O
interval	O
and	O
selects	O
the	O
proper	O
time	O
interval	O
for	O
the	O
next	O
checkpointing	O
.	O
Unlike	O
the	O
other	O
schemes	O
,	O
the	O
overhead	O
incurred	O
by	O
both	O
the	O
checkpointing	O
and	O
rollback	O
activities	O
are	O
considered	O
for	O
the	O
cost	B-KEY
evaluation	I-KEY
,	O
and	O
the	O
current	O
communication	B-KEY
pattern	I-KEY
is	O
reflected	O
in	O
the	O
selection	O
of	O
the	O
checkpointing	O
interval	O
.	O
Moreover	O
,	O
the	O
proposed	O
scheme	O
requires	O
no	O
extra	O
message	O
communication	O
for	O
the	O
checkpointing	O
interval	O
selection	O
and	O
can	O
easily	O
be	O
incorporated	O
into	O
the	O
existing	O
checkpointing	O
coordination	O
schemes	O

Estimation	O
of	O
thermal	B-KEY
coefficients	I-KEY
of	O
magneto-optical	B-KEY
media	I-KEY
Previously	O
we	O
described	O
a	O
method	O
for	O
estimating	O
the	O
thermal	B-KEY
conductivity	I-KEY
of	O
magneto-optic	B-KEY
recording	I-KEY
media	I-KEY
.	O
The	O
method	O
relies	O
on	O
identifying	O
the	O
laser	B-KEY
power	I-KEY
that	O
brings	O
the	O
maximum	B-KEY
temperature	I-KEY
of	O
the	O
TbFeCo	B-KEY
layer	O
to	O
as	O
high	O
as	O
the	O
Curie	O
temperature	O
.	O
We	O
extensively	O
use	O
a	O
similar	O
method	O
to	O
estimate	O
the	O
heat	B-KEY
capacity	I-KEY
of	O
a	O
dielectric	B-KEY
layer	I-KEY
,	O
a	O
TbFeCo	B-KEY
layer	O
,	O
and	O
an	O
aluminum	O
alloy	O
layer	O
of	O
magneto-optic	O
recording	O
media	O
.	O
Measurements	O
are	O
conducted	O
on	O
static	B-KEY
disks	I-KEY
with	O
a	O
beam	O
of	O
light	B-KEY
focused	I-KEY
on	O
a	O
TbFeCo	B-KEY
layer	O
.	O
The	O
method	O
has	O
the	O
advantage	O
of	O
thermal	B-KEY
diffusion	I-KEY
depending	O
on	O
a	O
multilayer	B-KEY
structure	I-KEY
and	O
irradiation	B-KEY
time	I-KEY

Prediction	O
and	O
compensation	B-KEY
of	O
dynamic	B-KEY
errors	I-KEY
for	O
coordinate	B-KEY
measuring	I-KEY
machines	I-KEY
Coordinate	B-KEY
measuring	I-KEY
machines	I-KEY
-LRB-	O
CMMs	O
-RRB-	O
are	O
already	O
widely	O
utilized	O
as	O
measuring	O
tools	O
in	O
the	O
modem	O
manufacturing	B-KEY
industry	I-KEY
.	O
Rapidly	O
approaching	O
now	O
is	O
the	O
trend	O
for	O
next-generation	O
CMMs	O
.	O
However	O
,	O
the	O
increases	O
in	O
measuring	O
velocity	O
of	O
CMM	O
applications	O
are	O
limited	O
by	O
dynamic	B-KEY
errors	I-KEY
that	O
occur	O
in	O
CMMs	O
.	O
In	O
this	O
paper	O
a	O
systematic	O
approach	O
for	O
modeling	O
the	O
dynamic	B-KEY
errors	I-KEY
of	O
a	O
touch-trigger	B-KEY
probe	I-KEY
CMM	O
is	O
developed	O
through	O
theoretical	O
analysis	O
and	O
experimental	O
study	O
.	O
An	O
overall	O
analysis	O
of	O
the	O
dynamic	B-KEY
errors	I-KEY
of	O
CMMs	O
is	O
conducted	O
,	O
with	O
weak	O
components	O
of	O
the	O
CMM	O
identified	O
by	O
a	O
laser	B-KEY
interferometer	I-KEY
.	O
The	O
probing	O
process	O
,	O
as	O
conducted	O
with	O
a	O
touch-trigger	B-KEY
probe	I-KEY
,	O
is	O
analyzed	O
.	O
The	O
dynamic	B-KEY
errors	I-KEY
are	O
measured	O
,	O
modeled	O
,	O
and	O
predicted	O
using	O
neural	B-KEY
networks	I-KEY
.	O
The	O
results	O
indicate	O
that	O
,	O
using	O
this	O
mode	O
,	O
it	O
is	O
possible	O
to	O
compensate	B-KEY
for	O
the	O
dynamic	B-KEY
errors	I-KEY
of	O
CMMs	O

Four-point	B-KEY
wavelets	I-KEY
and	O
their	O
applications	O
Multiresolution	B-KEY
analysis	I-KEY
-LRB-	O
MRA	O
-RRB-	O
and	O
wavelets	O
provide	O
useful	O
and	O
efficient	O
tools	O
for	O
representing	O
functions	O
at	O
multiple	O
levels	O
of	O
details	O
.	O
Wavelet	B-KEY
representations	I-KEY
have	O
been	O
used	O
in	O
a	O
broad	O
range	O
of	O
applications	O
,	O
including	O
image	B-KEY
compression	I-KEY
,	O
physical	B-KEY
simulation	I-KEY
and	O
numerical	B-KEY
analysis	I-KEY
.	O
In	O
this	O
paper	O
,	O
the	O
authors	O
construct	O
a	O
new	O
class	O
of	O
wavelets	O
,	O
called	O
four-point	B-KEY
wavelets	I-KEY
,	O
based	O
on	O
an	O
interpolatory	B-KEY
four-point	I-KEY
subdivision	I-KEY
scheme	I-KEY
.	O
They	O
are	O
of	O
local	O
support	O
,	O
symmetric	O
and	O
stable	O
.	O
The	O
analysis	O
and	O
synthesis	O
algorithms	O
have	O
linear	B-KEY
time	I-KEY
complexity	I-KEY
.	O
Depending	O
on	O
different	O
weight	B-KEY
parameters	I-KEY
w	O
,	O
the	O
scaling	B-KEY
functions	I-KEY
and	O
wavelets	O
generated	O
by	O
the	O
four-point	O
subdivision	O
scheme	O
are	O
of	O
different	O
degrees	O
of	O
smoothness	O
.	O
Therefore	O
the	O
user	O
can	O
select	O
better	O
wavelets	O
relevant	O
to	O
the	O
practice	O
among	O
the	O
classes	O
of	O
wavelets	O
.	O
The	O
authors	O
apply	O
the	O
four-point	B-KEY
wavelets	I-KEY
in	O
signal	O
compression	O
.	O
The	O
results	O
show	O
that	O
the	O
four-point	B-KEY
wavelets	I-KEY
behave	O
much	O
better	O
than	O
B-spline	B-KEY
wavelets	I-KEY
in	O
many	O
situations	O

Self-organized	B-KEY
critical	I-KEY
traffic	I-KEY
in	O
parallel	B-KEY
computer	I-KEY
networks	I-KEY
In	O
a	O
recent	O
paper	O
,	O
we	O
analysed	O
the	O
dynamics	O
of	O
traffic	O
flow	O
in	O
a	O
simple	O
,	O
square	B-KEY
lattice	I-KEY
architecture	I-KEY
.	O
It	O
was	O
shown	O
that	O
a	O
phase	B-KEY
transition	I-KEY
takes	O
place	O
between	O
a	O
free	O
and	O
a	O
congested	B-KEY
phase	I-KEY
.	O
The	O
transition	B-KEY
point	I-KEY
was	O
shown	O
to	O
exhibit	O
optimal	B-KEY
information	I-KEY
transfer	I-KEY
and	O
wide	B-KEY
fluctuations	I-KEY
in	O
time	O
,	O
with	O
scale-free	B-KEY
properties	I-KEY
.	O
In	O
this	O
paper	O
,	O
we	O
further	O
extend	O
our	O
analysis	O
by	O
considering	O
a	O
generalization	B-KEY
of	O
the	O
previous	O
model	O
in	O
which	O
the	O
rate	O
of	O
packet	B-KEY
emission	I-KEY
is	O
regulated	O
by	O
the	O
local	O
congestion	O
perceived	O
by	O
each	O
node	O
.	O
As	O
a	O
result	O
of	O
the	O
feedback	O
between	O
traffic	O
congestion	O
and	O
packet	B-KEY
release	I-KEY
,	O
the	O
system	O
is	O
poised	O
at	O
criticality	O
.	O
Many	O
well-known	O
statistical	B-KEY
features	I-KEY
displayed	O
by	O
Internet	B-KEY
traffic	I-KEY
are	O
recovered	O
from	O
our	O
model	O
in	O
a	O
natural	O
way	O

Secrets	O
of	O
the	O
Glasgow	B-KEY
Haskell	I-KEY
compiler	I-KEY
inliner	I-KEY
Higher-order	B-KEY
languages	I-KEY
such	O
as	O
Haskell	O
encourage	O
the	O
programmer	O
to	O
build	O
abstractions	B-KEY
by	O
composing	O
functions	O
.	O
A	O
good	O
compiler	O
must	O
inline	O
many	O
of	O
these	O
calls	O
to	O
recover	O
an	O
efficiently	O
executable	B-KEY
program	I-KEY
.	O
In	O
principle	O
,	O
inlining	O
is	O
dead	O
simple	O
:	O
just	O
replace	O
the	O
call	O
of	O
a	O
function	O
by	O
an	O
instance	O
of	O
its	O
body	O
.	O
But	O
any	O
compiler-writer	O
will	O
tell	O
you	O
that	O
inlining	O
is	O
a	O
black	O
art	O
,	O
full	O
of	O
delicate	O
compromises	O
that	O
work	O
together	O
to	O
give	O
good	O
performance	B-KEY
without	O
unnecessary	O
code	O
bloat	O
.	O
The	O
purpose	O
of	O
this	O
paper	O
is	O
,	O
therefore	O
,	O
to	O
articulate	O
the	O
key	O
lessons	O
we	O
learned	O
from	O
a	O
full-scale	O
``	O
production	O
''	O
inliner	O
,	O
the	O
one	O
used	O
in	O
the	O
Glasgow	O
Haskell	O
compiler	O
.	O
We	O
focus	O
mainly	O
on	O
the	O
algorithmic	B-KEY
aspects	I-KEY
,	O
but	O
we	O
also	O
provide	O
some	O
indicative	O
measurements	O
to	O
substantiate	O
the	O
importance	O
of	O
various	O
aspects	O
of	O
the	O
inliner	O

Controlling	O
in	O
between	O
the	O
Lorenz	O
and	O
the	O
Chen	B-KEY
systems	I-KEY
This	O
letter	O
investigates	O
a	O
new	O
chaotic	O
system	O
and	O
its	O
role	O
as	O
a	O
joint	O
function	O
between	O
two	O
complex	O
chaotic	O
systems	O
,	O
the	O
Lorenz	O
and	O
the	O
Chen	B-KEY
systems	I-KEY
,	O
using	O
a	O
simple	O
variable	O
constant	O
controller	O
.	O
With	O
the	O
gradual	O
tuning	B-KEY
of	O
the	O
controller	O
,	O
the	O
controlled	O
system	O
evolves	O
from	O
the	O
canonical	O
Lorenz	B-KEY
attractor	I-KEY
to	O
the	O
Chen	B-KEY
attractor	I-KEY
through	O
the	O
new	O
transition	B-KEY
chaotic	I-KEY
attractor	I-KEY
.	O
This	O
evolving	O
procedure	O
reveals	O
the	O
forming	O
mechanisms	O
of	O
all	O
similar	O
and	O
closely	O
related	O
chaotic	O
systems	O
,	O
and	O
demonstrates	O
that	O
a	O
simple	O
control	O
technique	O
can	O
be	O
very	O
useful	O
in	O
generating	O
and	O
analyzing	O
some	O
complex	O
chaotic	O
dynamical	O
phenomena	O

Philadelphia	B-KEY
stock	I-KEY
exchange	I-KEY
taps	O
TimesTen	B-KEY
for	O
database	O
technology	O
PHLX	O
rolls	O
out	O
Equity	B-KEY
Options	I-KEY
AutoQuote	I-KEY
System	I-KEY
to	O
traders	O
as	O
the	O
first	O
application	O
to	O
leverage	O
its	O
enhanced	O
data	B-KEY
architecture	I-KEY

Web-based	B-KEY
intelligent	I-KEY
helpdesk-support	I-KEY
environment	I-KEY
With	O
the	O
advent	O
of	O
Internet	B-KEY
technology	I-KEY
,	O
it	O
is	O
now	O
feasible	O
to	O
provide	O
effective	O
and	O
efficient	O
helpdesk	O
service	O
over	O
the	O
global	O
Internet	O
to	O
meet	O
customers	O
'	O
requirements	O
and	O
satisfaction	O
.	O
In	O
this	O
research	O
,	O
we	O
have	O
designed	O
and	O
developed	O
a	O
Web-based	B-KEY
intelligent	I-KEY
helpdesk-support	I-KEY
environment	I-KEY
,	O
WebHotLine	B-KEY
,	O
to	O
support	O
the	O
customer	B-KEY
service	I-KEY
centre	I-KEY
of	O
a	O
large	O
multinational	O
corporation	O
in	O
the	O
electronics	O
industry	O
.	O
The	O
paper	O
describes	O
the	O
basic	O
architecture	O
of	O
the	O
environment	O
that	O
supports	O
the	O
major	O
functions	O
of	O
Web-based	B-KEY
fault	I-KEY
information	I-KEY
retrieval	I-KEY
,	O
online	B-KEY
multilingual	I-KEY
translation	I-KEY
capability	I-KEY
,	O
different	O
operating	O
modes	O
of	O
video-conferencing	O
for	O
enhanced	O
support	O
and	O
direct	O
intelligent	O
fault	O
diagnosis	O
by	O
customers	O
or	O
customer	O
support	O
engineers	O
.	O
As	O
a	O
result	O
,	O
WebHotLine	B-KEY
helps	O
to	O
save	O
cost	O
in	O
eliminating	O
the	O
expensive	O
overseas	O
telephone	O
charges	O
,	O
reduction	O
in	O
machine	O
down	O
time	O
and	O
number	O
of	O
on-site	O
visits	O
by	O
service	O
engineers	O
as	O
in	O
traditional	O
helpdesk	O
environment	O

Construction	O
of	O
double	B-KEY
sampling	I-KEY
s-control	I-KEY
charts	I-KEY
for	O
agile	B-KEY
manufacturing	I-KEY
Double	O
sampling	O
-LRB-	O
DS	O
-RRB-	O
X-control	O
charts	O
are	O
designed	O
to	O
allow	O
quick	O
detection	O
of	O
a	O
small	O
shift	O
of	O
process	O
mean	O
and	O
provides	O
a	O
quick	O
response	O
in	O
an	O
agile	B-KEY
manufacturing	I-KEY
environment	O
.	O
However	O
,	O
the	O
DS	O
X-control	O
charts	O
assume	O
that	O
the	O
process	B-KEY
standard	I-KEY
deviation	I-KEY
remains	O
unchanged	O
throughout	O
the	O
entire	O
course	O
of	O
the	O
statistical	B-KEY
process	I-KEY
control	I-KEY
.	O
Therefore	O
,	O
a	O
complementary	O
DS	O
chart	O
that	O
can	O
be	O
used	O
to	O
monitor	O
the	O
process	O
variation	O
caused	O
by	O
changes	O
in	O
process	B-KEY
standard	I-KEY
deviation	I-KEY
should	O
be	O
developed	O
.	O
In	O
this	O
paper	O
,	O
the	O
development	O
of	O
the	O
DS	O
s-charts	O
for	O
quickly	O
detecting	O
small	O
shift	O
in	O
process	B-KEY
standard	I-KEY
deviation	I-KEY
for	O
agile	B-KEY
manufacturing	I-KEY
is	O
presented	O
.	O
The	O
construction	O
of	O
the	O
DS	O
s-charts	O
is	O
based	O
on	O
the	O
same	O
concepts	O
in	O
constructing	O
the	O
DS	O
X-charts	O
and	O
is	O
formulated	O
as	O
an	O
optimization	O
problem	O
and	O
solved	O
with	O
a	O
genetic	B-KEY
algorithm	I-KEY
.	O
The	O
efficiency	O
of	O
the	O
DS	O
s-control	O
chart	O
is	O
compared	O
with	O
that	O
of	O
the	O
traditional	O
s-control	O
chart	O
.	O
The	O
results	O
show	O
that	O
the	O
DS	O
s-control	O
charts	O
can	O
be	O
a	O
more	O
economically	O
preferable	O
alternative	O
in	O
detecting	O
small	O
shifts	O
than	O
traditional	O
s-control	O
charts	O

Integration	O
-	O
no	O
longer	O
a	O
barrier	O
?	O
-LSB-	O
agile	B-KEY
business	I-KEY
-RSB-	O
Web	B-KEY
services	I-KEY
will	O
be	O
a	O
critical	O
technology	O
for	O
enabling	O
the	O
`	O
agile	B-KEY
business	I-KEY
'	O

Design	O
of	O
1-D	O
and	O
2-D	O
variable	B-KEY
fractional	I-KEY
delay	I-KEY
allpass	I-KEY
filters	I-KEY
using	O
weighted	B-KEY
least-squares	I-KEY
method	I-KEY
In	O
this	O
paper	O
,	O
a	O
weighted	B-KEY
least-squares	I-KEY
method	I-KEY
is	O
presented	O
to	O
design	O
one-dimensional	O
and	O
two-dimensional	O
variable	B-KEY
fractional	I-KEY
delay	I-KEY
allpass	I-KEY
filters	I-KEY
.	O
First	O
,	O
each	O
coefficient	O
of	O
the	O
variable	O
allpass	O
filter	O
is	O
expressed	O
as	O
the	O
polynomial	O
of	O
the	O
fractional	B-KEY
delay	I-KEY
parameter	I-KEY
.	O
Then	O
,	O
the	O
nonlinear	O
phase	O
error	O
is	O
approximated	O
by	O
a	O
weighted	B-KEY
equation	I-KEY
error	I-KEY
such	O
that	O
the	O
cost	B-KEY
function	I-KEY
can	O
be	O
converted	O
into	O
a	O
quadratic	O
form	O
.	O
Next	O
,	O
by	O
minimizing	O
the	O
weighted	B-KEY
equation	I-KEY
error	I-KEY
,	O
the	O
optimal	B-KEY
polynomial	I-KEY
coefficients	I-KEY
can	O
be	O
obtained	O
iteratively	O
by	O
solving	O
a	O
set	O
of	O
linear	B-KEY
simultaneous	I-KEY
equations	I-KEY
at	O
each	O
iteration	O
.	O
Finally	O
,	O
the	O
design	O
examples	O
are	O
demonstrated	O
to	O
illustrate	O
the	O
effectiveness	O
of	O
the	O
proposed	O
approach	O

New	B-KEY
age	I-KEY
computing	I-KEY
-LSB-	O
autonomic	B-KEY
computing	I-KEY
-RSB-	O
Autonomic	B-KEY
computing	I-KEY
-LRB-	O
AC	B-KEY
-RRB-	O
,	O
sometimes	O
called	O
self-managed	B-KEY
computing	I-KEY
,	O
is	O
the	O
name	O
chosen	O
by	O
IBM	O
to	O
describe	O
the	O
company	O
's	O
new	O
initiative	O
aimed	O
at	O
making	O
computing	O
more	O
reliable	O
and	O
problem-free	O
.	O
It	O
is	O
a	O
response	O
to	O
a	O
growing	O
realization	O
that	O
the	O
problem	O
today	O
with	O
computers	O
is	O
not	O
that	O
they	O
need	O
more	O
speed	O
or	O
have	O
too	O
little	O
memory	O
,	O
but	O
that	O
they	O
crash	O
all	O
too	O
often	O
.	O
This	O
article	O
reviews	O
current	O
initiatives	O
being	O
carried	O
out	O
in	O
the	O
AC	B-KEY
field	O
by	O
the	O
IT	O
industry	O
,	O
followed	O
by	O
key	O
challenges	O
which	O
require	O
to	O
be	O
addressed	O
in	O
its	O
development	O
and	O
implementation	O

Sensing	O
and	O
control	O
of	O
double-sided	B-KEY
arc	I-KEY
welding	I-KEY
process	O
The	O
welding	O
industry	O
is	O
driven	O
to	O
improve	O
productivity	O
without	O
sacrificing	O
quality	O
.	O
For	O
thick	B-KEY
material	I-KEY
welding	I-KEY
,	O
the	O
current	O
practice	O
is	O
to	O
use	O
backing	O
or	O
multiple	O
passes	O
.	O
The	O
laser	B-KEY
welding	I-KEY
process	I-KEY
,	O
capable	O
of	O
achieving	O
deep	O
narrow	O
penetration	O
,	O
can	O
significantly	O
improve	O
welding	O
productivity	O
for	O
such	O
applications	O
by	O
reducing	O
the	O
number	O
of	O
passes	O
.	O
However	O
,	O
its	O
competitiveness	O
in	O
comparison	O
with	O
traditional	O
arc	O
welding	O
is	O
weakened	O
by	O
its	O
high	O
cost	O
,	O
strict	O
fit-up	O
requirement	O
,	O
and	O
difficulty	O
in	O
welding	O
large	O
structures	O
.	O
In	O
this	O
work	O
,	O
a	O
different	O
method	O
,	O
referred	O
to	O
as	O
double-sided	B-KEY
arc	I-KEY
welding	I-KEY
-LRB-	O
DSAW	O
-RRB-	O
is	O
developed	O
to	O
improve	O
the	O
arc	O
concentration	O
for	O
arc	O
welding	O
.	O
A	O
sensing	O
and	O
control	B-KEY
system	I-KEY
is	O
developed	O
to	O
achieve	O
deep	O
narrow	O
penetration	O
under	O
variations	O
in	O
welding	O
conditions	O
.	O
Experiments	O
verified	O
that	O
the	O
pulsed	O
keyhole	O
DSAW	O
system	O
developed	O
is	O
capable	O
of	O
achieving	O
deep	O
narrow	O
penetration	O
on	O
a	O
1/2	O
inch	O
thick	O
square	O
butt	O
joint	O
in	O
a	O
single	O
pass	O

Pioneering	B-KEY
women	I-KEY
in	O
computer	O
science	O
Although	O
their	O
contributions	O
are	O
not	O
well	O
documented	O
,	O
women	O
have	O
played	O
an	O
important	O
role	O
in	O
the	O
development	O
of	O
computer	O
science	O
.	O
A	O
survey	O
of	O
women	O
pioneers	O
demonstrates	O
their	O
influence	O
in	O
designing	O
and	O
programming	O
the	O
first	O
electronic	B-KEY
computers	I-KEY
and	O
languages	O
,	O
while	O
laying	O
the	O
groundwork	O
for	O
women	O
's	O
expanding	O
involvement	O
in	O
science	O

A	O
scalable	O
and	O
efficient	O
systolic	B-KEY
algorithm	I-KEY
for	O
the	O
longest	B-KEY
common	I-KEY
subsequence	I-KEY
problem	I-KEY
A	O
longest	O
common	O
subsequence	O
-LRB-	O
LCS	O
-RRB-	O
of	O
two	O
strings	O
is	O
a	O
common	O
subsequence	O
of	O
two	O
strings	O
of	O
maximal	O
length	O
.	O
The	O
LCS	O
problem	O
is	O
that	O
of	O
finding	O
an	O
LCS	O
of	O
two	O
given	O
strings	O
and	O
the	O
length	O
of	O
the	O
LCS	O
.	O
This	O
problem	O
has	O
been	O
the	O
subject	O
of	O
much	O
research	O
because	O
its	O
solution	O
can	O
be	O
applied	O
in	O
many	O
areas	O
.	O
In	O
this	O
paper	O
,	O
a	O
scalable	O
and	O
efficient	O
systolic	B-KEY
algorithm	I-KEY
is	O
presented	O
.	O
For	O
two	O
given	O
strings	O
of	O
length	O
m	O
and	O
n	O
,	O
where	O
m	O
>	O
or	O
=	O
n	O
,	O
the	O
algorithm	O
can	O
solve	O
the	O
LCS	O
problem	O
in	O
m	O
+2	O
r-1	O
-LRB-	O
respectively	O
n	O
+2	O
r-1	O
-RRB-	O
time	O
steps	O
with	O
r	O
<	O
n/2	O
-LRB-	O
respectively	O
r	O
<	O
m/2	O
-RRB-	O
processors	O
.	O
Experimental	O
results	O
show	O
that	O
the	O
algorithm	O
can	O
be	O
faster	O
on	O
multicomputers	O
than	O
all	O
the	O
previous	O
systolic	B-KEY
algorithms	I-KEY
for	O
the	O
same	O
problem	O

Complex	B-KEY
dynamics	I-KEY
in	O
nearly	B-KEY
symmetric	I-KEY
three-cell	I-KEY
cellular	I-KEY
neural	I-KEY
networks	I-KEY
The	O
paper	O
introduces	O
a	O
class	O
of	O
third-order	O
nonsymmetric	O
Cellular	O
Neural	O
Networks	O
-LRB-	O
CNNs	B-KEY
-RRB-	O
,	O
and	O
shows	O
through	O
computer	O
simulations	O
that	O
they	O
undergo	O
a	O
cascade	O
of	O
period	B-KEY
doubling	I-KEY
bifurcations	I-KEY
which	O
leads	O
to	O
the	O
birth	O
of	O
a	O
large-size	B-KEY
complex	I-KEY
attractor	I-KEY
.	O
A	O
major	O
point	O
is	O
that	O
these	O
bifurcations	O
and	O
complex	B-KEY
dynamics	I-KEY
happen	O
in	O
a	O
small	O
neighborhood	O
of	O
a	O
particular	O
CNN	B-KEY
with	O
a	O
symmetric	B-KEY
interconnection	I-KEY
matrix	I-KEY

Hierarchical	B-KEY
neuro-fuzzy	I-KEY
quadtree	I-KEY
models	O
Hybrid	O
neuro-fuzzy	B-KEY
systems	I-KEY
have	O
been	O
in	O
evidence	O
during	O
the	O
past	O
few	O
years	O
,	O
due	O
to	O
its	O
attractive	O
combination	O
of	O
the	O
learning	O
capacity	O
of	O
artificial	O
neural	O
networks	O
with	O
the	O
interpretability	O
of	O
the	O
fuzzy	O
systems	O
.	O
This	O
article	O
proposes	O
a	O
new	O
hybrid	O
neuro-fuzzy	O
model	O
,	O
named	O
hierarchical	B-KEY
neuro-fuzzy	I-KEY
quadtree	I-KEY
-LRB-	O
HNFQ	O
-RRB-	O
,	O
which	O
is	O
based	O
on	O
a	O
recursive	O
partitioning	O
method	O
of	O
the	O
input	O
space	O
named	O
quadtree	O
.	O
The	O
article	O
describes	O
the	O
architecture	O
of	O
this	O
new	O
model	O
,	O
presenting	O
its	O
basic	O
cell	O
and	O
its	O
learning	B-KEY
algorithm	I-KEY
.	O
The	O
HNFQ	O
system	O
is	O
evaluated	O
in	O
three	O
well	O
known	O
benchmark	O
applications	O
:	O
the	O
sinc	O
-LRB-	O
x	O
,	O
y	O
-RRB-	O
function	O
approximation	O
,	O
the	O
Mackey	B-KEY
Glass	I-KEY
chaotic	I-KEY
series	I-KEY
forecast	O
and	O
the	O
two	O
spirals	O
problem	O
.	O
When	O
compared	O
to	O
other	O
neuro-fuzzy	B-KEY
systems	I-KEY
,	O
the	O
HNFQ	O
exhibits	O
competing	O
results	O
,	O
with	O
two	O
major	O
advantages	O
it	O
automatically	O
creates	O
its	O
own	O
structure	O
and	O
it	O
is	O
not	O
limited	O
to	O
few	O
input	O
variables	O

A	O
suggestion	O
of	O
fractional-order	B-KEY
controller	I-KEY
for	O
flexible	B-KEY
spacecraft	I-KEY
attitude	I-KEY
control	I-KEY
A	O
controller	O
design	O
method	O
for	O
flexible	B-KEY
spacecraft	I-KEY
attitude	I-KEY
control	I-KEY
is	O
proposed	O
.	O
The	O
system	O
is	O
first	O
described	O
by	O
a	O
partial	B-KEY
differential	I-KEY
equation	I-KEY
with	O
internal	B-KEY
damping	I-KEY
.	O
Then	O
the	O
frequency	B-KEY
response	I-KEY
is	O
analyzed	O
,	O
and	O
the	O
three	O
basic	O
characteristics	O
of	O
the	O
flexible	O
system	O
,	O
namely	O
,	O
average	O
function	O
,	O
lower	O
bound	O
and	O
upper	O
bound	O
are	O
defined	O
.	O
On	O
this	O
basis	O
,	O
a	O
fractional-order	B-KEY
controller	I-KEY
is	O
proposed	O
,	O
which	O
functions	O
as	O
phase	B-KEY
stabilization	I-KEY
control	I-KEY
for	O
lower	O
frequency	O
and	O
smoothly	O
enters	O
to	O
amplitude	B-KEY
stabilization	I-KEY
at	O
higher	O
frequency	O
by	O
proper	O
amplitude	O
attenuation	O
.	O
It	O
is	O
shown	O
that	O
the	O
equivalent	O
damping	B-KEY
ratio	I-KEY
increases	O
in	O
proportion	O
to	O
the	O
square	O
of	O
frequency	O

Look	O
into	O
the	O
future	O
of	O
content	B-KEY
management	I-KEY
Predictions	O
of	O
consolidation	O
in	O
the	O
Content	B-KEY
Management	I-KEY
-LRB-	O
CM	O
-RRB-	O
vendor	O
arena	O
have	O
appeared	O
in	O
nearly	O
every	O
major	O
industry	O
prognosis	O
over	O
the	O
past	O
two	O
years	O
.	O
Gartner	O
Group	O
,	O
for	O
example	O
,	O
recently	O
reiterated	O
its	O
prediction	O
that	O
half	O
the	O
CM	O
vendors	O
in	O
existence	O
in	O
mid-2001	O
would	O
leave	O
the	O
marketplace	O
by	O
the	O
end	O
of	O
2002	O
.	O
Analysts	O
consistently	O
advise	O
prospective	O
CM	O
buyers	O
to	O
tread	O
carefully	O
because	O
their	O
vendor	O
may	O
not	O
stick	O
around	O
.	O
But	O
fortunately	O
,	O
the	O
story	O
goes	O
,	O
fewer	O
vendor	O
choices	O
will	O
finally	O
bring	O
greater	O
clarity	O
and	O
sharper	O
differentiators	O
to	O
this	O
otherwise	O
very	O
messy	O
product	O
landscape	O
.	O
In	O
fact	O
,	O
the	O
number	O
of	O
CM	O
vendors	O
continues	O
to	O
rise	O
.	O
Industry	O
growth	O
has	O
come	O
through	O
greater	O
demand	O
among	O
CM	O
buyers	O
,	O
but	O
also	O
expanding	O
product	B-KEY
functionality	I-KEY
as	O
well	O
as	O
successful	O
partnerships	B-KEY
.	O
The	O
marketplace	O
certainly	O
can	O
not	O
sustain	O
its	O
current	O
breadth	O
of	O
vendors	O
in	O
the	O
long	O
run	O
,	O
yet	O
it	O
remains	O
unclear	O
when	O
and	O
how	O
any	O
serious	O
industry	B-KEY
consolidation	I-KEY
will	O
occur	O
.	O
In	O
the	O
meantime	O
,	O
evolving	O
business	O
models	O
and	O
feature	O
sets	O
have	O
created	O
just	O
the	O
kind	O
of	O
clearer	O
segmentation	O
and	O
transparent	O
product	O
differences	O
that	O
were	O
supposed	O
to	O
emerge	O
following	O
an	O
industry	O
contraction	O

A	O
feature-preserving	B-KEY
volumetric	I-KEY
technique	I-KEY
to	O
merge	B-KEY
surface	I-KEY
triangulations	I-KEY
Several	O
extensions	O
and	O
improvements	O
to	O
surface	B-KEY
merging	I-KEY
procedures	I-KEY
based	O
on	O
the	O
extraction	O
of	O
isosurfaces	O
from	O
a	O
distance	O
map	O
defined	O
on	O
an	O
adaptive	B-KEY
background	I-KEY
grid	I-KEY
are	O
presented	O
.	O
The	O
main	O
objective	O
is	O
to	O
extend	O
the	O
application	O
of	O
these	O
algorithms	O
to	O
surfaces	O
with	O
sharp	B-KEY
edges	I-KEY
and	O
comers	O
.	O
In	O
order	O
to	O
deal	O
with	O
objects	O
of	O
different	O
length	O
scales	O
,	O
the	O
initial	O
background	O
grids	O
are	O
created	O
using	O
a	O
Delaunay	B-KEY
triangulation	I-KEY
method	I-KEY
and	O
local	B-KEY
voxelizations	I-KEY
.	O
A	O
point	O
enrichment	O
technique	O
that	O
introduces	O
points	O
into	O
the	O
background	O
grid	O
along	O
detected	O
surface	O
features	O
such	O
as	O
ridges	B-KEY
is	O
used	O
to	O
ensure	O
that	O
these	O
features	O
are	O
preserved	O
in	O
the	O
final	O
merged	O
surface	O
.	O
The	O
surface	O
merging	O
methodology	O
is	O
extended	O
to	O
include	O
other	O
Boolean	B-KEY
operations	I-KEY
between	O
surface	B-KEY
triangulations	I-KEY
.	O
The	O
iso-surface	B-KEY
extraction	I-KEY
algorithms	O
are	O
modified	O
to	O
obtain	O
the	O
correct	O
iso-surface	O
for	O
multi-component	B-KEY
objects	I-KEY
.	O
The	O
procedures	O
are	O
demonstrated	O
with	O
various	O
examples	O
,	O
ranging	O
from	O
simple	B-KEY
geometrical	I-KEY
entities	I-KEY
to	O
complex	B-KEY
engineering	I-KEY
applications	I-KEY
.	O
The	O
present	O
algorithms	O
allow	O
realistic	O
modelling	O
of	O
a	O
large	O
number	O
of	O
complex	O
engineering	O
geometries	O
using	O
overlapping	B-KEY
components	I-KEY
defined	O
discretely	O
,	O
i.e.	O
via	O
surface	B-KEY
triangulations	I-KEY
.	O
This	O
capability	O
is	O
very	O
useful	O
for	O
grid	O
generation	O
starting	O
from	O
data	O
originated	O
in	O
measurements	O
or	O
images	B-KEY

Two	O
efficient	O
algorithms	O
for	O
the	O
generalized	B-KEY
maximum	I-KEY
balanced	I-KEY
flow	I-KEY
problem	I-KEY
Minoux	O
-LRB-	O
1976	O
-RRB-	O
considered	O
the	O
maximum	O
balanced	O
flow	O
problem	O
,	O
i.e.	O
the	O
problem	O
of	O
finding	O
a	O
maximum	O
flow	O
in	O
a	O
two-terminal	B-KEY
network	I-KEY
N	O
=	O
-LRB-	O
V	O
,	O
A	O
-RRB-	O
with	O
source	O
s	O
and	O
sink	O
t	O
satisfying	O
the	O
constraint	O
that	O
any	O
arc-flow	O
of	O
N	O
is	O
bounded	O
by	O
a	O
fixed	O
proportion	O
of	O
the	O
total	O
flow	O
value	O
from	O
s	O
to	O
t	O
,	O
where	O
V	O
is	O
vertex	O
set	O
and	O
A	O
is	O
arc	O
set	O
.	O
As	O
a	O
generalization	O
,	O
we	O
focus	O
on	O
the	O
problem	O
of	O
maximizing	O
the	O
total	O
flow	O
value	O
of	O
a	O
generalized	O
flow	O
in	O
N	O
with	O
gains	O
gamma	O
-LRB-	O
a	O
-RRB-	O
>	O
0	O
-LRB-	O
a	O
in	O
A	O
-RRB-	O
where	O
any	O
arc-flow	O
is	O
bounded	O
by	O
a	O
fixed	O
proportion	O
of	O
the	O
total	O
flow	O
value	O
,	O
where	O
gamma	O
-LRB-	O
a	O
-RRB-	O
f	O
-LRB-	O
a	O
-RRB-	O
units	O
arrive	O
at	O
the	O
vertex	O
w	O
for	O
each	O
arc-flow	O
f	O
-LRB-	O
a	O
-RRB-	O
-LRB-	O
a	O
identical	O
to	O
-LRB-	O
upsilon	O
,	O
w	O
-RRB-	O
in	O
A	O
-RRB-	O
entering	O
vertex	O
upsilon	O
in	O
a	O
generalized	O
flow	O
.	O
Our	O
main	O
results	O
are	O
to	O
propose	O
two	O
polynomial	B-KEY
algorithms	I-KEY
for	O
this	O
problem	O
.	O
The	O
first	O
algorithm	O
runs	O
in	O
O	O
-LRB-	O
mM	O
-LRB-	O
n	O
,	O
m	O
,	O
B	O
'	O
-RRB-	O
log	O
B	O
-RRB-	O
time	O
,	O
where	O
B	O
is	O
the	O
maximum	O
absolute	O
value	O
among	O
integral	O
values	O
used	O
by	O
an	O
instance	O
of	O
the	O
problem	O
,	O
and	O
M	O
-LRB-	O
n	O
,	O
m	O
,	O
B	O
'	O
-RRB-	O
denotes	O
the	O
complexity	O
of	O
solving	O
a	O
generalized	O
maximum	O
flow	O
problem	O
in	O
a	O
network	O
with	O
n	O
vertices	O
,	O
and	O
m	O
arcs	O
,	O
and	O
a	O
rational	O
instance	O
expressed	O
with	O
integers	O
between	O
1	O
and	O
B	O
'	O
.	O
In	O
the	O
second	O
algorithm	O
,	O
using	O
a	O
parameterized	B-KEY
technique	I-KEY
,	O
runs	O
in	O
O	O
-LRB-	O
-LCB-	O
M	O
-LRB-	O
n	O
,	O
m	O
,	O
B	O
'	O
-RRB-	O
-RCB-	O
/	O
sup	O
2	O
/	O
-RRB-	O
time	O

The	O
effect	O
of	O
voxel	B-KEY
size	I-KEY
on	O
the	O
accuracy	O
of	O
dose-volume	O
histograms	O
of	O
prostate	O
/	O
sup	O
125/I	O
seed	O
implants	O
Cumulative	B-KEY
dose-volume	I-KEY
histograms	I-KEY
-LRB-	O
DVH	O
-RRB-	O
are	O
crucial	O
in	O
evaluating	O
the	O
quality	O
of	O
radioactive	B-KEY
seed	I-KEY
prostate	I-KEY
implants	I-KEY
.	O
When	O
calculating	O
DVHs	O
,	O
the	O
choice	O
of	O
voxel	B-KEY
size	I-KEY
is	O
a	O
compromise	O
between	O
computational	B-KEY
speed	I-KEY
-LRB-	O
larger	O
voxels	O
-RRB-	O
and	O
accuracy	O
-LRB-	O
smaller	O
voxels	O
-RRB-	O
.	O
We	O
quantified	O
the	O
effect	O
of	O
voxel	B-KEY
size	I-KEY
on	O
the	O
accuracy	O
of	O
DVHs	O
using	O
an	O
in-house	B-KEY
computer	I-KEY
program	I-KEY
.	O
The	O
program	O
was	O
validated	O
by	O
comparison	O
with	O
a	O
hand-calculated	O
DVH	O
for	O
a	O
single	O
0.4-U	O
iodine-125	O
model	O
6711	O
seed	O
.	O
We	O
used	O
the	O
program	O
to	O
find	O
the	O
voxel	B-KEY
size	I-KEY
required	O
to	O
obtain	O
accurate	O
DVHs	O
of	O
five	O
iodine-125	O
prostate	O
implant	O
patients	O
at	O
our	O
institution	O
.	O
One-millimeter	O
cubes	O
were	O
sufficient	O
to	O
obtain	O
DVHs	O
that	O
are	O
accurate	O
within	O
5	O
%	O
up	O
to	O
200	O
%	O
of	O
the	O
prescription	O
dose	O
.	O
For	O
the	O
five	O
patient	O
plans	O
,	O
we	O
obtained	O
good	O
agreement	O
with	O
the	O
VariSeed	O
-LRB-	O
version	O
6.7	O
,	O
Varian	O
,	O
USA	O
-RRB-	O
treatment	O
planning	O
software	O
's	O
DVH	O
algorithm	O
by	O
using	O
voxels	O
with	O
a	O
sup-inf	O
dimension	O
equal	O
to	O
the	O
spacing	O
between	O
successive	O
transverse	O
seed	O
implant	O
planes	O
-LRB-	O
5	O
mm	O
-RRB-	O
.	O
The	O
volume	O
that	O
receives	O
at	O
least	O
200	O
%	O
of	O
the	O
target	O
dose	O
,	O
V/sub	O
200	O
/	O
,	O
calculated	O
by	O
VariSeed	O
was	O
30	O
%	O
to	O
43	O
%	O
larger	O
than	O
that	O
calculated	O
by	O
our	O
program	O
with	O
small	O
voxels	O
.	O
The	O
single-seed	O
DVH	O
calculated	O
by	O
VariSeed	O
fell	O
below	O
the	O
hand	O
calculation	O
by	O
up	O
to	O
50	O
%	O
at	O
low	O
doses	O
-LRB-	O
30	O
Gy	O
-RRB-	O
,	O
and	O
above	O
it	O
by	O
over	O
50	O
%	O
at	O
high	O
doses	O
-LRB-	O
>	O
250	O
Gy	O
-RRB-	O

Handles	B-KEY
and	O
exception	B-KEY
safety	O
,	O
Part	O
1	O
.	O
A	O
simple	O
handle	B-KEY
class	O
Every	O
C++	B-KEY
program	I-KEY
that	O
uses	O
inheritance	O
must	O
manage	O
memory	O
somehow	O
.	O
The	O
most	O
obvious	O
way	O
to	O
do	O
so	O
is	O
directly	O
,	O
but	O
programmers	O
who	O
create	O
complicated	O
data	O
structures	O
often	O
have	O
trouble	O
figuring	O
out	O
what	O
parts	O
of	O
those	O
data	O
structures	O
are	O
safe	O
to	O
delete	O
when	O
.	O
The	O
classical	O
method	O
of	O
dealing	O
with	O
such	O
complexity	O
is	O
to	O
hide	O
it	O
in	O
a	O
class	O
.	O
Such	O
classes	O
are	O
typically	O
called	O
handles	B-KEY
;	O
the	O
idea	O
is	O
to	O
attach	O
a	O
handle	B-KEY
object	O
to	O
another	O
object	O
that	O
contains	O
the	O
actual	O
data	O
.	O
The	O
simplest	O
form	O
of	O
a	O
handle	B-KEY
,	O
which	O
we	O
have	O
discussed	O
in	O
this	O
article	O
,	O
is	O
one	O
in	O
which	O
each	O
handle	B-KEY
object	O
corresponds	O
to	O
a	O
single	O
object	O
from	O
the	O
inheritance	B-KEY
hierarchy	I-KEY
.	O
Such	O
handles	B-KEY
are	O
straightforward	O
to	O
use	O
and	O
to	O
implement	O
and	O
tend	O
to	O
be	O
intrinsically	O
exception	B-KEY
safe	O
in	O
almost	O
all	O
respects	O
.	O
The	O
one	O
exception	B-KEY
hazard	O
in	O
such	O
a	O
class	O
is	O
typically	O
the	O
assignment	B-KEY
operator	I-KEY
.	O
Assignment	B-KEY
operators	I-KEY
often	O
test	O
for	O
self-assignment	B-KEY
to	O
avoid	O
aliasing	B-KEY
problems	I-KEY
.	O
As	O
Herb	O
Sutter	O
has	O
observed	O
-LRB-	O
2000	O
-RRB-	O
,	O
programs	O
that	O
need	O
such	O
tests	O
are	O
almost	O
always	O
exception	B-KEY
unsafe	O
.	O
By	O
rewriting	O
the	O
assignment	B-KEY
operator	I-KEY
,	O
we	O
ensure	O
that	O
we	O
do	O
not	O
do	O
anything	O
irrevocable	O
until	O
the	O
possibility	O
of	O
throwing	O
an	O
exception	B-KEY
has	O
passed	O
.	O
This	O
strategy	O
ensures	O
that	O
if	O
an	O
exception	B-KEY
occurs	O
while	O
our	O
assignment	B-KEY
operator	I-KEY
is	O
executing	O
,	O
we	O
do	O
not	O
corrupt	O
the	O
rest	O
of	O
our	O
system	O

A	O
self-organizing	B-KEY
context-based	I-KEY
approach	I-KEY
to	O
the	O
tracking	O
of	O
multiple	O
robot	B-KEY
trajectories	I-KEY
We	O
have	O
combined	O
competitive	O
and	O
Hebbian	B-KEY
learning	I-KEY
in	O
a	O
neural	O
network	O
designed	O
to	O
learn	O
and	O
recall	O
complex	B-KEY
spatiotemporal	I-KEY
sequences	I-KEY
.	O
In	O
such	O
sequences	O
,	O
a	O
particular	O
item	O
may	O
occur	O
more	O
than	O
once	O
or	O
the	O
sequence	O
may	O
share	B-KEY
states	I-KEY
with	O
another	O
sequence	O
.	O
Processing	O
of	O
repeated/shared	O
states	O
is	O
a	O
hard	O
problem	O
that	O
occurs	O
very	O
often	O
in	O
the	O
domain	O
of	O
robotics	O
.	O
The	O
proposed	O
model	O
consists	O
of	O
two	O
groups	O
of	O
synaptic	B-KEY
weights	I-KEY
:	O
competitive	O
interlayer	O
and	O
Hebbian	B-KEY
intralayer	I-KEY
connections	I-KEY
,	O
which	O
are	O
responsible	O
for	O
encoding	O
respectively	O
the	O
spatial	O
and	O
temporal	O
features	O
of	O
the	O
input	O
sequence	O
.	O
Three	O
additional	O
mechanisms	O
allow	O
the	O
network	O
to	O
deal	O
with	O
shared	B-KEY
states	I-KEY
:	O
context	B-KEY
units	I-KEY
,	O
neurons	O
disabled	O
from	O
learning	O
,	O
and	O
redundancy	O
used	O
to	O
encode	O
sequence	B-KEY
states	I-KEY
.	O
The	O
network	O
operates	O
by	O
determining	O
the	O
current	O
and	O
the	O
next	O
state	O
of	O
the	O
learned	O
sequences	O
.	O
The	O
model	O
is	O
simulated	O
over	O
various	O
sets	O
of	O
robot	B-KEY
trajectories	I-KEY
in	O
order	O
to	O
evaluate	O
its	O
storage	O
and	O
retrieval	B-KEY
abilities	I-KEY
;	O
its	O
sequence	B-KEY
sampling	I-KEY
effects	I-KEY
;	O
its	O
robustness	O
to	O
noise	O
and	O
its	O
tolerance	O
to	O
fault	O

Autonomous	O
detection	O
of	O
crack	O
initiation	O
using	O
surface-mounted	O
piezotransducers	O
In	O
this	O
paper	O
we	O
report	O
on	O
the	O
application	O
of	O
an	O
in	B-KEY
situ	I-KEY
health	I-KEY
monitoring	I-KEY
system	O
,	O
comprising	O
an	O
array	O
of	O
piezoceramic	B-KEY
wafer	I-KEY
elements	I-KEY
,	O
to	O
the	O
detection	O
of	O
fatigue	B-KEY
degradation	I-KEY
in	O
metallic	B-KEY
specimens	I-KEY
exposed	O
to	O
cyclic	B-KEY
loading	I-KEY
.	O
Lamb	B-KEY
waves	I-KEY
,	O
transmitted	O
through	O
a	O
beam	O
test	O
coupon	O
,	O
are	O
sensed	O
using	O
small	O
surface-mounted	B-KEY
piezotransducer	I-KEY
elements	I-KEY
,	O
and	O
the	O
signals	O
are	O
then	O
autonomously	O
analysed	O
for	O
indications	O
relating	O
to	O
the	O
onset	O
of	O
structural	B-KEY
degradation	I-KEY
.	O
The	O
experimental	O
results	O
confirm	O
the	O
efficacy	O
of	O
the	O
approach	O
and	O
provide	O
a	O
demonstration	O
of	O
good	O
robustness	B-KEY
under	O
realistic	O
loading	B-KEY
conditions	I-KEY
,	O
emphasizing	O
the	O
great	O
potential	O
for	O
developing	O
an	O
automated	B-KEY
in	I-KEY
situ	I-KEY
structural	I-KEY
health	I-KEY
monitoring	I-KEY
system	O
for	O
application	O
to	O
fatigue-prone	O
operational	O
structures	O
,	O
such	O
as	O
aircraft	B-KEY

The	O
chromatic	B-KEY
spectrum	I-KEY
of	O
mixed	B-KEY
hypergraphs	I-KEY
A	O
mixed	B-KEY
hypergraph	I-KEY
is	O
a	O
triple	O
H	O
=	O
-LRB-	O
X	O
,	O
C	O
,	O
D	O
-RRB-	O
,	O
where	O
X	O
is	O
the	O
vertex	B-KEY
set	I-KEY
,	O
and	O
each	O
of	O
C	O
,	O
D	O
is	O
a	O
list	O
of	O
subsets	O
of	O
X	O
.	O
A	O
strict	B-KEY
k-coloring	I-KEY
of	O
H	O
is	O
a	O
surjection	O
c	O
:	O
X	O
-LCB-	O
1	O
,	O
...	O
,	O
k	O
-RCB-	O
such	O
that	O
each	O
member	O
of	O
le	O
has	O
two	O
vertices	O
assigned	O
a	O
common	O
value	O
and	O
each	O
member	O
of	O
D	O
has	O
two	O
vertices	O
assigned	O
distinct	O
values	O
.	O
The	O
feasible	O
set	O
of	O
H	O
is	O
-LCB-	O
k	O
:	O
H	O
has	O
a	O
strict	B-KEY
k-coloring	I-KEY
-RCB-	O
.	O
Among	O
other	O
results	O
,	O
we	O
prove	O
that	O
a	O
finite	O
set	O
of	O
positive	B-KEY
integers	I-KEY
is	O
the	O
feasible	O
set	O
of	O
some	O
mixed	B-KEY
hypergraph	I-KEY
if	O
and	O
only	O
if	O
it	O
omits	O
the	O
number	O
I	O
or	O
is	O
an	O
interval	O
starting	O
with	O
1	O
.	O
For	O
the	O
set	O
-LCB-	O
s	O
,	O
t	O
-RCB-	O
with	O
2	O
<	O
or	O
=	O
s	O
<	O
or	O
=	O
t	O
-	O
2	O
,	O
the	O
smallest	O
realization	O
has	O
2t	O
-	O
s	O
vertices	O
.	O
When	O
every	O
member	O
of	O
C	O
union	O
D	O
is	O
a	O
single	O
interval	O
in	O
an	O
underlying	O
linear	O
order	O
on	O
the	O
vertices	O
,	O
the	O
feasible	O
set	O
is	O
also	O
a	O
single	O
interval	O
of	O
integers	O

The	O
heat	B-KEY
is	O
on	O
-LSB-	O
building	B-KEY
automation	I-KEY
systems	I-KEY
-RSB-	O
Integrating	O
building	B-KEY
automation	I-KEY
systems	I-KEY
-LRB-	O
BASs	O
-RRB-	O
can	O
result	O
in	O
systems	O
that	O
have	O
the	O
ability	O
to	O
sense	O
changes	O
in	O
the	O
air	O
temperature	O
through	O
a	O
building	O
's	O
heating	B-KEY
,	O
ventilation	B-KEY
,	O
and	O
air	B-KEY
conditioning	I-KEY
-LRB-	O
HVAC	B-KEY
-RRB-	O
systems	O
.	O
Taking	O
advantages	O
of	O
the	O
Internet	B-KEY
,	O
using	O
remote	B-KEY
monitoring	I-KEY
,	O
and	O
building	O
interoperability	B-KEY
through	O
open	O
protocol	O
systems	O
are	O
some	O
of	O
the	O
issues	O
discussed	O
throughout	O
the	O
BAS/HVAC	O
community	O
.	O
By	O
putting	O
information	O
over	O
the	O
Internet	B-KEY
,	O
facility	O
managers	O
get	O
real-time	B-KEY
data	I-KEY
on	O
energy	O
usage	O
and	O
performance	O
issues	O

Blind	B-KEY
source	I-KEY
separation	I-KEY
applied	O
to	O
image	B-KEY
cryptosystems	I-KEY
with	O
dual	B-KEY
encryption	I-KEY
Blind	B-KEY
source	I-KEY
separation	I-KEY
-LRB-	O
BSS	O
-RRB-	O
is	O
explored	O
to	O
add	O
another	O
encryption	O
level	O
besides	O
the	O
existing	O
encryption	O
methods	O
for	O
image	B-KEY
cryptosystems	I-KEY
.	O
The	O
transmitted	B-KEY
images	I-KEY
are	O
covered	O
with	O
a	O
noise	B-KEY
image	I-KEY
by	O
specific	O
mixing	O
before	O
encryption	O
and	O
then	O
recovered	O
through	O
BSS	O
after	O
decryption	O
.	O
Simulation	O
results	O
illustrate	O
the	O
validity	O
of	O
the	O
proposed	O
method	O

Information	B-KEY
needs	I-KEY
of	O
the	O
working	B-KEY
journalists	I-KEY
in	O
Orissa	O
:	O
a	O
study	O
Provides	O
an	O
insight	O
into	O
the	O
various	O
information	B-KEY
needs	I-KEY
of	O
working	B-KEY
journalists	I-KEY
in	O
Orissa	O
.	O
Analyses	O
data	O
received	O
from	O
226	O
working	B-KEY
journalists	I-KEY
representing	O
40	O
newspaper	B-KEY
organisations	I-KEY
.	O
Also	O
depicts	O
the	O
specialisation	O
of	O
working	B-KEY
journalists	I-KEY
,	O
their	O
frequency	O
of	O
information	B-KEY
requirement	I-KEY
,	O
mode	O
of	O
dissemination	O
preferred	O
,	O
information	B-KEY
sources	I-KEY
explored	O
,	O
mode	O
of	O
services	O
opted	O
,	O
and	O
their	O
information	O
privations	O
.	O
The	O
study	O
asserts	O
that	O
subjects	O
primarily	O
concerned	O
with	O
the	O
professional	B-KEY
work	I-KEY
and	O
image	O
of	O
the	O
working	B-KEY
journalists	I-KEY
are	O
rated	O
utmost	O
significant	O

Quantum	B-KEY
sensitive	I-KEY
dependence	I-KEY
Wave	B-KEY
functions	I-KEY
of	O
bounded	B-KEY
quantum	I-KEY
systems	I-KEY
with	O
time-independent	B-KEY
potentials	I-KEY
,	O
being	O
almost	O
periodic	B-KEY
functions	I-KEY
,	O
can	O
not	O
have	O
time	B-KEY
asymptotics	I-KEY
as	O
in	O
classical	B-KEY
chaos	I-KEY
.	O
However	O
,	O
bounded	B-KEY
quantum	I-KEY
systems	I-KEY
with	O
time-dependent	B-KEY
interactions	I-KEY
,	O
as	O
used	O
in	O
quantum	B-KEY
control	I-KEY
,	O
may	O
have	O
continuous	O
spectrum	O
and	O
the	O
rate	O
of	O
growth	O
of	O
observables	O
is	O
an	O
issue	O
of	O
both	O
theoretical	O
and	O
practical	O
concern	O
.	O
Rates	O
of	O
growth	O
in	O
quantum	O
mechanics	O
are	O
discussed	O
by	O
constructing	O
quantities	O
with	O
the	O
same	O
physical	O
meaning	O
as	O
those	O
involved	O
in	O
the	O
classical	B-KEY
Lyapunov	I-KEY
exponent	I-KEY
.	O
A	O
generalized	O
notion	O
of	O
quantum	B-KEY
sensitive	I-KEY
dependence	I-KEY
is	O
introduced	O
and	O
the	O
mathematical	O
structure	O
of	O
the	O
operator	B-KEY
matrix	I-KEY
elements	I-KEY
that	O
correspond	O
to	O
different	O
types	O
of	O
growth	O
is	O
characterized	O

Application	O
of	O
artificial	B-KEY
intelligence	I-KEY
to	O
search	O
ground-state	B-KEY
geometry	I-KEY
of	O
clusters	O
We	O
introduce	O
a	O
global	B-KEY
optimization	I-KEY
procedure	I-KEY
,	O
the	O
neural-assisted	B-KEY
genetic	I-KEY
algorithm	I-KEY
-LRB-	O
NAGA	O
-RRB-	O
.	O
It	O
combines	O
the	O
power	O
of	O
an	O
artificial	B-KEY
neural	I-KEY
network	I-KEY
-LRB-	O
ANN	O
-RRB-	O
with	O
the	O
versatility	O
of	O
the	O
genetic	O
algorithm	O
.	O
This	O
method	O
is	O
suitable	O
to	O
solve	O
optimization	O
problems	O
that	O
depend	O
on	O
some	O
kind	O
of	O
heuristics	O
to	O
limit	O
the	O
search	O
space	O
.	O
If	O
a	O
reasonable	O
amount	O
of	O
data	O
is	O
available	O
,	O
the	O
ANN	O
can	O
``	O
understand	O
''	O
the	O
problem	O
and	O
provide	O
the	O
genetic	O
algorithm	O
with	O
a	O
selected	O
population	B-KEY
of	O
elements	O
that	O
will	O
speed	O
up	O
the	O
search	O
for	O
the	O
optimum	B-KEY
solution	I-KEY
.	O
We	O
tested	O
the	O
method	O
in	O
a	O
search	O
for	O
the	O
ground-state	B-KEY
geometry	I-KEY
of	O
silicon	B-KEY
clusters	I-KEY
.	O
We	O
trained	O
the	O
ANN	O
with	O
information	O
about	O
the	O
geometry	O
and	O
energetics	O
of	O
small	O
silicon	B-KEY
clusters	I-KEY
.	O
Next	O
,	O
the	O
ANN	O
learned	O
how	O
to	O
restrict	O
the	O
configurational	O
space	O
for	O
larger	O
silicon	B-KEY
clusters	I-KEY
.	O
For	O
Si/sub	O
10	O
/	O
and	O
Si/sub	O
20	O
/	O
,	O
we	O
noticed	O
that	O
the	O
NAGA	O
is	O
at	O
least	O
three	O
times	O
faster	O
than	O
the	O
``	O
pure	O
''	O
genetic	O
algorithm	O
.	O
As	O
the	O
size	O
of	O
the	O
cluster	O
increases	O
,	O
it	O
is	O
expected	O
that	O
the	O
gain	O
in	O
terms	O
of	O
time	O
will	O
increase	O
as	O
well	O

Modelling	O
tomographic	B-KEY
cone-beam	I-KEY
projection	I-KEY
data	I-KEY
from	O
a	O
polyhedral	B-KEY
phantom	I-KEY
Analytical	O
phantoms	O
are	O
used	O
to	O
generate	O
projection	O
data	O
for	O
testing	O
reconstruction	B-KEY
accuracy	I-KEY
in	O
computed	B-KEY
axial	I-KEY
tomography	I-KEY
.	O
A	O
circular	O
source	O
locus	O
-LRB-	O
equivalent	O
to	O
rotating	O
specimen	O
with	O
a	O
fixed	O
source	O
-RRB-	O
provides	O
insufficient	O
data	O
for	O
`	O
exact	O
'	O
reconstruction	O
in	O
cone-beam	B-KEY
transmission	I-KEY
tomography	I-KEY
,	O
thus	O
phantom	O
data	O
are	O
useful	O
for	O
studying	O
the	O
consequent	O
errors	O
and	O
also	O
for	O
investigating	O
alternative	B-KEY
scanning	I-KEY
loci	I-KEY
and	O
reconstruction	O
techniques	O
.	O
We	O
present	O
an	O
algorithm	O
that	O
can	O
compute	O
phantom	O
cone-beam	O
projection	O
data	O
from	O
a	O
phantom	O
comprising	O
geometrically	B-KEY
defined	I-KEY
polyhedra	I-KEY
.	O
Each	O
polyhedron	O
is	O
defined	O
as	O
a	O
set	O
of	O
polygons	O
enclosing	O
a	O
volume	O
of	O
fixed	O
linear	B-KEY
attenuation	I-KEY
coefficient	I-KEY
.	O
The	O
algorithm	O
works	O
by	O
projecting	O
each	O
polygon	O
in	O
turn	O
onto	O
the	O
modelled	O
detector	O
array	O
,	O
which	O
accumulates	O
the	O
product	O
of	O
source	O
to	O
polygon	O
intersection	O
distance	O
-LRB-	O
for	O
the	O
rays	O
intersecting	O
each	O
detector	O
element	O
-RRB-	O
,	O
linear	B-KEY
attenuation	I-KEY
coefficient	I-KEY
and	O
sign	O
of	O
projected	O
polygon	O
area	O
-LRB-	O
indicating	O
whether	O
rays	O
enter	O
or	O
exit	O
the	O
polyhedron	O
at	O
this	O
face	O
-RRB-	O
.	O
The	O
phantom	O
data	O
are	O
rotated	O
according	O
to	O
the	O
projection	O
angle	O
,	O
whilst	O
the	O
source	O
location	O
and	O
detector	O
plane	O
remain	O
fixed	O
.	O
Polyhedra	O
can	O
be	O
of	O
simple	O
geometric	O
form	O
,	O
or	O
complex	O
surfaces	O
derived	O
from	O
3D	O
images	O
of	O
real	O
specimens	O
.	O
This	O
algorithm	O
is	O
illustrated	O
using	O
a	O
phantom	O
comprising	O
989	O
238	O
polygons	O
,	O
representing	O
an	O
iso-surface	O
generated	O
from	O
a	O
microtomographic	B-KEY
reconstruction	I-KEY
of	O
a	O
piece	O
of	O
walrus	B-KEY
tusk	I-KEY

Frontier	O
between	O
separability	B-KEY
and	O
quantum	B-KEY
entanglement	I-KEY
in	O
a	O
many	B-KEY
spin	I-KEY
system	I-KEY
We	O
discuss	O
the	O
critical	B-KEY
point	I-KEY
x/sub	O
c	O
/	O
separating	B-KEY
the	O
quantum	B-KEY
entangled	I-KEY
and	O
separable	B-KEY
states	O
in	O
two	O
series	O
of	O
N	O
spins	O
S	O
in	O
the	O
simple	O
mixed	O
state	O
characterized	O
by	O
the	O
matrix	O
operator	O
rho	O
=	O
x	O
|	O
phi	O
>	O
<	O
phi	O
|	O
+1	O
-	O
x/D/sup	O
N/I/sub	O
D/N	O
,	O
where	O
x	O
in	O
-LSB-	O
0	O
,	O
1	O
-RSB-	O
,	O
D	O
=	O
2S	O
+	O
1	O
,	O
I/sub	O
D/N	O
is	O
the	O
D/sup	O
N	O
/	O
*	O
D/sup	O
N	O
/	O
unity	O
matrix	O
and	O
|	O
phi	O
>	O
is	O
a	O
special	O
entangled	O
state	O
.	O
The	O
cases	O
x	O
=	O
0	O
and	O
x	O
=	O
1	O
correspond	O
respectively	O
to	O
fully	O
random	B-KEY
spins	I-KEY
and	O
to	O
a	O
fully	O
entangled	B-KEY
state	I-KEY
.	O
In	O
the	O
first	O
of	O
these	O
series	O
we	O
consider	O
special	O
states	O
|	O
phi	O
>	O
invariant	O
under	O
charge	B-KEY
conjugation	I-KEY
,	O
that	O
generalizes	O
the	O
N	O
=	O
2	O
spin	O
S	O
=	O
1/2	O
Einstein-Podolsky-Rosen	B-KEY
state	I-KEY
,	O
and	O
in	O
the	O
second	O
one	O
we	O
consider	O
generalizations	O
of	O
the	O
Werner	O
-LRB-	O
1989	O
-RRB-	O
density	O
matrices	O
.	O
The	O
evaluation	O
of	O
the	O
critical	B-KEY
point	I-KEY
x/sub	O
c	O
/	O
was	O
done	O
through	O
bounds	O
coming	O
from	O
the	O
partial	B-KEY
transposition	I-KEY
method	I-KEY
of	O
Peres	O
-LRB-	O
1996	O
-RRB-	O
and	O
the	O
conditional	O
nonextensive	B-KEY
entropy	I-KEY
criterion	I-KEY
.	O
Our	O
results	O
suggest	O
the	O
conjecture	O
that	O
whenever	O
the	O
bounds	O
coming	O
from	O
both	O
methods	O
coincide	O
the	O
result	O
of	O
x/sub	O
c	O
/	O
is	O
the	O
exact	O
one	O
.	O
The	O
results	O
we	O
present	O
are	O
relevant	O
for	O
the	O
discussion	O
of	O
quantum	B-KEY
computing	I-KEY
,	O
teleportation	B-KEY
and	O
cryptography	B-KEY

A	O
transactional	B-KEY
asynchronous	O
replication	O
scheme	O
for	O
mobile	B-KEY
database	I-KEY
systems	O
In	O
mobile	B-KEY
database	I-KEY
systems	O
,	O
mobility	O
of	O
users	O
has	O
a	O
significant	O
impact	O
on	O
data	B-KEY
replication	I-KEY
.	O
As	O
a	O
result	O
,	O
the	O
various	O
replica	O
control	O
protocols	O
that	O
exist	O
today	O
in	O
traditional	O
distributed	O
and	O
multidatabase	B-KEY
environments	O
are	O
no	O
longer	O
suitable	O
.	O
To	O
solve	O
this	O
problem	O
,	O
a	O
new	O
mobile	B-KEY
database	I-KEY
replication	O
scheme	O
,	O
the	O
Transaction-Level	O
Result-Set	O
Propagation	O
-LRB-	O
TLRSP	O
-RRB-	O
model	O
,	O
is	O
put	O
forward	O
in	O
this	O
paper	O
.	O
The	O
conflict	O
detection	O
and	O
resolution	O
strategy	O
based	O
on	O
TLRSP	O
is	O
discussed	O
in	O
detail	O
,	O
and	O
the	O
implementation	O
algorithm	O
is	O
proposed	O
.	O
In	O
order	O
to	O
compare	O
the	O
performance	O
of	O
the	O
TLRSP	O
model	O
with	O
that	O
of	O
other	O
mobile	O
replication	O
schemes	O
,	O
we	O
have	O
developed	O
a	O
detailed	O
simulation	O
model	O
.	O
Experimental	O
results	O
show	O
that	O
the	O
TLRSP	O
model	O
provides	O
an	O
efficient	O
support	O
for	O
replicated	O
mobile	B-KEY
database	I-KEY
systems	O
by	O
reducing	O
reprocessing	O
overhead	O
and	O
maintaining	O
database	O
consistency	O

Accuracy	O
and	O
stability	B-KEY
of	O
splitting	O
with	O
Stabilizing	B-KEY
Corrections	O
This	O
paper	O
contains	O
a	O
convergence	B-KEY
analysis	I-KEY
for	O
the	O
method	O
of	O
stabilizing	B-KEY
corrections	O
,	O
which	O
is	O
an	O
internally	O
consistent	O
splitting	O
scheme	O
for	O
initial-boundary	O
value	O
problems	O
.	O
To	O
obtain	O
more	O
accuracy	O
and	O
a	O
better	O
treatment	O
of	O
explicit	O
terms	O
several	O
extensions	O
are	O
regarded	O
and	O
analyzed	O
.	O
The	O
relevance	O
of	O
the	O
theoretical	O
results	O
is	O
tested	O
for	O
convection-diffusion-reaction	B-KEY
equations	I-KEY

Real-time	B-KEY
quasi-2-D	I-KEY
inversion	I-KEY
of	O
array	B-KEY
resistivity	I-KEY
logging	I-KEY
data	I-KEY
using	O
neural	B-KEY
network	I-KEY
We	O
present	O
a	O
quasi-2-D	O
real-time	B-KEY
inversion	I-KEY
algorithm	I-KEY
for	O
a	O
modern	O
galvanic	B-KEY
array	I-KEY
tool	I-KEY
via	O
dimensional	B-KEY
reduction	I-KEY
and	O
neural	B-KEY
network	I-KEY
simulation	O
.	O
Using	O
reciprocity	B-KEY
and	O
superposition	B-KEY
,	O
we	O
apply	O
a	O
numerical	B-KEY
focusing	I-KEY
technique	I-KEY
to	O
the	O
unfocused	B-KEY
data	I-KEY
.	O
The	O
numerically	O
focused	B-KEY
data	I-KEY
are	O
much	O
less	O
subject	O
to	O
2-D	O
and	O
layering	O
effects	O
and	O
can	O
be	O
approximated	O
as	O
from	O
a	O
cylindrical	O
1-D	O
Earth	O
.	O
We	O
then	O
perform	O
1-D	B-KEY
inversion	I-KEY
on	O
the	O
focused	B-KEY
data	I-KEY
to	O
provide	O
approximate	O
information	O
about	O
the	O
2-D	O
resistivity	O
structure	O
.	O
A	O
neural	B-KEY
network	I-KEY
is	O
used	O
to	O
perform	O
forward	B-KEY
modeling	I-KEY
in	O
the	O
1-D	B-KEY
inversion	I-KEY
,	O
which	O
is	O
several	O
hundred	O
times	O
faster	O
than	O
conventional	O
numerical	O
forward	O
solutions	O
.	O
Testing	O
our	O
inversion	O
algorithm	O
on	O
both	O
synthetic	O
and	O
field	O
data	O
shows	O
that	O
this	O
fast	O
inversion	O
algorithm	O
is	O
useful	O
for	O
providing	O
formation	B-KEY
resistivity	I-KEY
information	O
at	O
a	O
well	B-KEY
site	I-KEY

The	O
canonical	B-KEY
dual	I-KEY
frame	I-KEY
of	O
a	O
wavelet	B-KEY
frame	I-KEY
We	O
show	O
that	O
there	O
exist	O
wavelet	B-KEY
frames	I-KEY
that	O
have	O
nice	O
dual	O
wavelet	B-KEY
frames	I-KEY
,	O
but	O
for	O
which	O
the	O
canonical	B-KEY
dual	I-KEY
frame	I-KEY
does	O
not	O
consist	O
of	O
wavelets	O
,	O
i.e.	O
,	O
can	O
not	O
be	O
generated	O
by	O
the	O
translates	O
and	O
dilates	O
of	O
a	O
single	O
function	O

Remember	O
e-commerce	B-KEY
?	O
Yeah	O
,	O
well	O
,	O
it	O
's	O
still	O
here	O
Sandy	O
Kemper	O
,	O
the	O
always	O
outspoken	O
CEO	O
of	O
successful	O
e-commerce	B-KEY
company	O
eScout	B-KEY
,	O
offers	O
his	O
views	O
on	O
the	O
purported	O
demise	O
of	O
``	O
commerce	O
''	O
in	O
e-commerce	B-KEY
,	O
and	O
what	O
opportunities	O
lie	O
ahead	O
for	O
those	O
bankers	B-KEY
bold	O
enough	O
to	O
act	O
in	O
a	O
market	O
turned	O
tentative	O
by	O
early	O
excesses	O

Scalable	B-KEY
secure	I-KEY
group	I-KEY
communication	I-KEY
over	O
IP	B-KEY
multicast	I-KEY
We	O
introduce	O
and	O
analyze	O
a	O
scalable	O
rekeying	O
scheme	O
for	O
implementing	O
secure	O
group	O
communications	O
Internet	B-KEY
protocol	I-KEY
multicast	I-KEY
.	O
We	O
show	O
that	O
our	O
scheme	O
incurs	O
constant	O
processing	O
,	O
message	O
,	O
and	O
storage	B-KEY
overhead	I-KEY
for	O
a	O
rekey	O
operation	O
when	O
a	O
single	O
member	O
joins	O
or	O
leaves	O
the	O
group	O
,	O
and	O
logarithmic	O
overhead	O
for	O
bulk	O
simultaneous	O
changes	O
to	O
the	O
group	O
membership	O
.	O
These	O
bounds	O
hold	O
even	O
when	O
group	B-KEY
dynamics	I-KEY
are	O
not	O
known	O
a	O
priori	O
.	O
Our	O
rekeying	B-KEY
algorithm	I-KEY
requires	O
a	O
particular	O
clustering	O
of	O
the	O
members	O
of	O
the	O
secure	B-KEY
multicast	I-KEY
group	I-KEY
.	O
We	O
describe	O
a	O
protocol	O
to	O
achieve	O
such	O
clustering	O
and	O
show	O
that	O
it	O
is	O
feasible	O
to	O
efficiently	O
cluster	O
members	O
over	O
realistic	O
Internet-like	B-KEY
topologies	I-KEY
.	O
We	O
evaluate	O
the	O
overhead	B-KEY
of	O
our	O
own	O
rekeying	O
scheme	O
and	O
also	O
of	O
previously	O
published	O
schemes	O
via	O
simulation	B-KEY
over	O
an	O
Internet	B-KEY
topology	I-KEY
map	I-KEY
containing	O
over	O
280	O
000	O
routers	O
.	O
Through	O
analysis	O
and	O
detailed	O
simulations	B-KEY
,	O
we	O
show	O
that	O
this	O
rekeying	O
scheme	O
performs	O
better	O
than	O
previous	O
schemes	O
for	O
a	O
single	O
change	O
to	O
group	B-KEY
membership	I-KEY
.	O
Further	O
,	O
for	O
bulk	O
group	O
changes	O
,	O
our	O
algorithm	O
outperforms	O
all	O
previously	O
known	O
schemes	O
by	O
several	O
orders	O
of	O
magnitude	O
in	O
terms	O
of	O
actual	O
bandwidth	B-KEY
usage	I-KEY
,	O
processing	B-KEY
costs	I-KEY
,	O
and	O
storage	B-KEY
requirements	I-KEY

Network	B-KEY
intrusion	I-KEY
and	O
fault	B-KEY
detection	I-KEY
:	O
a	O
statistical	O
anomaly	O
approach	O
With	O
the	O
advent	O
and	O
explosive	O
growth	O
of	O
the	O
global	O
Internet	B-KEY
and	O
electronic	B-KEY
commerce	I-KEY
environments	I-KEY
,	O
adaptive/automatic	B-KEY
network/service	I-KEY
intrusion	I-KEY
and	O
anomaly	O
detection	O
in	O
wide	B-KEY
area	I-KEY
data	I-KEY
networks	I-KEY
and	O
e-commerce	B-KEY
infrastructures	I-KEY
is	O
fast	O
gaining	O
critical	O
research	O
and	O
practical	O
importance	O
.	O
We	O
present	O
and	O
demonstrate	O
the	O
use	O
of	O
a	O
general-purpose	O
hierarchical	O
multitier	O
multiwindow	O
statistical	O
anomaly	O
detection	O
technology	O
and	O
system	O
that	O
operates	O
automatically	O
,	O
adaptively	O
,	O
and	O
proactively	O
,	O
and	O
can	O
be	O
applied	O
to	O
various	O
networking	O
technologies	O
,	O
including	O
both	O
wired	O
and	O
wireless	B-KEY
ad	I-KEY
hoc	I-KEY
networks	I-KEY
.	O
Our	O
method	O
uses	O
statistical	B-KEY
models	I-KEY
and	O
multivariate	B-KEY
classifiers	I-KEY
to	O
detect	O
anomalous	O
network	O
conditions	O
.	O
Some	O
numerical	O
results	O
are	O
also	O
presented	O
that	O
demonstrate	O
that	O
our	O
proposed	O
methodology	O
can	O
reliably	O
detect	O
attacks	O
with	O
traffic	B-KEY
anomaly	I-KEY
intensity	I-KEY
as	O
low	O
as	O
3-5	O
percent	O
of	O
the	O
typical	O
background	B-KEY
traffic	I-KEY
intensity	I-KEY
,	O
thus	O
promising	O
to	O
generate	O
an	O
effective	O
early	O
warning	O

Noise-constrained	B-KEY
hyperspectral	I-KEY
data	I-KEY
compression	I-KEY
Storage	O
and	O
transmission	B-KEY
requirements	I-KEY
for	O
hyperspectral	B-KEY
data	I-KEY
sets	I-KEY
are	O
significant	O
.	O
To	O
reduce	O
hardware	B-KEY
costs	I-KEY
,	O
well-designed	O
compression	O
techniques	O
are	O
needed	O
to	O
preserve	O
information	B-KEY
content	I-KEY
while	O
maximizing	O
compression	B-KEY
ratios	I-KEY
.	O
Lossless	B-KEY
compression	I-KEY
techniques	I-KEY
maintain	O
data	B-KEY
integrity	I-KEY
,	O
but	O
yield	O
small	O
compression	B-KEY
ratios	I-KEY
.	O
We	O
present	O
a	O
slightly	B-KEY
lossy	I-KEY
compression	I-KEY
algorithm	I-KEY
that	O
uses	O
the	O
noise	B-KEY
statistics	I-KEY
of	O
the	O
data	O
to	O
preserve	O
information	B-KEY
content	I-KEY
while	O
maximizing	O
compression	B-KEY
ratios	I-KEY
.	O
The	O
adaptive	O
principal	O
components	O
analysis	O
-LRB-	O
APCA	O
-RRB-	O
algorithm	O
uses	O
noise	B-KEY
statistics	I-KEY
to	O
determine	O
the	O
number	O
of	O
significant	O
principal	O
components	O
and	O
selects	O
only	O
those	O
that	O
are	O
required	O
to	O
represent	O
each	O
pixel	O
to	O
within	O
the	O
noise	B-KEY
level	I-KEY
.	O
We	O
demonstrate	O
the	O
effectiveness	O
of	O
these	O
methods	O
with	O
airborne	O
visible/infrared	O
spectrometer	O
-LRB-	O
AVIRIS	O
-RRB-	O
,	O
hyperspectral	O
digital	O
imagery	O
collection	O
experiment	O
-LRB-	O
HYDICE	O
-RRB-	O
,	O
hyperspectral	B-KEY
mapper	I-KEY
-LRB-	O
HYMAP	B-KEY
-RRB-	O
,	O
and	O
Hyperion	B-KEY
datasets	I-KEY

Development	O
of	O
an	O
integrated	O
and	O
open-architecture	B-KEY
precision	B-KEY
motion	B-KEY
control	I-KEY
system	O
In	O
this	O
paper	O
,	O
the	O
development	O
of	O
an	O
integrated	O
and	O
open-architecture	B-KEY
precision	B-KEY
motion	B-KEY
control	I-KEY
system	O
is	O
presented	O
.	O
The	O
control	O
system	O
is	O
generally	O
applicable	O
,	O
but	O
it	O
is	O
developed	O
with	O
a	O
particular	O
focus	O
on	O
direct	B-KEY
drive	I-KEY
servo	I-KEY
systems	I-KEY
based	O
on	O
linear	B-KEY
motors	I-KEY
.	O
The	O
overall	O
control	O
system	O
is	O
comprehensive	O
,	O
comprising	O
of	O
various	O
selected	O
control	O
and	O
instrumentation	O
components	O
,	O
integrated	O
within	O
a	O
configuration	O
of	O
hardware	O
architecture	O
centred	O
around	O
a	O
dSPACE	O
DS1004	O
DSP	O
processor	O
board	O
.	O
These	O
components	O
include	O
a	O
precision	B-KEY
composite	B-KEY
controller	I-KEY
-LRB-	O
comprising	O
of	O
feedforward	B-KEY
and	O
feedback	B-KEY
control	O
-RRB-	O
,	O
a	O
disturbance	O
observer	O
,	O
an	O
adaptive	B-KEY
notch	I-KEY
filter	I-KEY
,	O
and	O
a	O
geometrical	B-KEY
error	I-KEY
compensator	I-KEY
.	O
The	O
hardware	O
architecture	O
,	O
software	O
development	O
platform	O
,	O
user	O
interface	O
,	O
and	O
all	O
constituent	O
control	O
components	O
are	O
described	O

The	O
incredible	O
shrinking	O
pipeline	O
We	O
look	O
at	O
the	O
harsh	O
facts	O
concerning	O
the	O
percentage	O
of	O
degrees	O
awarded	O
in	O
CS	O
to	O
women	B-KEY
.	O
We	O
study	O
the	O
trend	O
of	O
degrees	O
awarded	O
in	O
CS	O
since	O
1980	O
,	O
and	O
compare	O
the	O
trend	O
in	O
CS	O
to	O
other	O
science	B-KEY
and	O
engineering	B-KEY
disciplines	O
.	O
We	O
consider	O
the	O
relationship	O
between	O
the	O
percentage	O
of	O
degrees	O
awarded	O
to	O
women	B-KEY
by	O
a	O
CS	O
department	O
and	O
the	O
college	O
the	O
CS	O
department	O
is	O
within	O
.	O
We	O
find	O
that	O
CS	O
departments	O
in	O
engineering	B-KEY
colleges	O
graduate	O
,	O
on	O
average	O
,	O
proportionately	O
fewer	O
women	B-KEY
than	O
CS	O
departments	O
in	O
non-engineering	O
colleges	O
.	O
We	O
request	O
that	O
the	O
community	O
respond	O
to	O
the	O
facts	O
and	O
speculations	O
presented	O
in	O
this	O
article	O

Optimization	O
of	O
cutting	O
conditions	O
for	O
single	B-KEY
pass	I-KEY
turning	I-KEY
operations	I-KEY
using	O
a	O
deterministic	B-KEY
approach	I-KEY
An	O
optimization	O
analysis	O
,	O
strategy	O
and	O
CAM	B-KEY
software	I-KEY
for	O
the	O
selection	O
of	O
economic	B-KEY
cutting	I-KEY
conditions	I-KEY
in	O
single	B-KEY
pass	I-KEY
turning	I-KEY
operations	I-KEY
are	O
presented	O
using	O
a	O
deterministic	B-KEY
approach	I-KEY
.	O
The	O
optimization	O
is	O
based	O
on	O
criteria	O
typified	O
by	O
the	O
maximum	B-KEY
production	I-KEY
rate	I-KEY
and	O
includes	O
a	O
host	O
of	O
practical	O
constraints	O
.	O
It	O
is	O
shown	O
that	O
the	O
deterministic	O
optimization	O
approach	O
involving	O
mathematical	B-KEY
analyses	I-KEY
of	O
constrained	B-KEY
economic	I-KEY
trends	I-KEY
and	O
graphical	O
representation	O
on	O
the	O
feed-speed	O
domain	O
provides	O
a	O
clearly	O
defined	O
strategy	O
that	O
not	O
only	O
provides	O
a	O
unique	O
global	O
optimum	O
solution	O
,	O
but	O
also	O
the	O
software	O
that	O
is	O
suitable	O
for	O
on-line	O
CAM	O
applications	O
.	O
A	O
numerical	O
study	O
has	O
verified	O
the	O
developed	O
optimization	O
strategies	O
and	O
software	O
and	O
has	O
shown	O
the	O
economic	O
benefits	O
of	O
using	O
optimization	O

A	O
high-resolution	O
high-frequency	O
monolithic	B-KEY
top-shooting	I-KEY
microinjector	I-KEY
free	O
of	O
satellite	O
drops	O
-	O
part	O
I	O
:	O
concept	O
,	O
design	O
,	O
and	O
model	O
Introduces	O
an	O
innovative	O
microinjector	O
design	O
,	O
featuring	O
a	O
bubble	B-KEY
valve	I-KEY
,	O
which	O
entails	O
superior	O
droplet	B-KEY
ejection	I-KEY
characteristics	I-KEY
and	O
monolithic	O
fabrication	O
,	O
which	O
allows	O
handling	O
of	O
a	O
wide	O
range	O
of	O
liquids	O
.	O
This	O
new	O
microinjector	O
uses	O
asymmetric	B-KEY
bubbles	I-KEY
to	O
reduce	O
crosstalk	B-KEY
,	O
increase	O
frequency	B-KEY
response	I-KEY
and	O
eliminate	O
satellite	B-KEY
droplets	I-KEY
.	O
During	O
a	O
firing	O
,	O
i.e.	O
,	O
droplet	O
ejection	O
,	O
the	O
``	O
virtual	B-KEY
valve	I-KEY
''	O
closes	O
,	O
by	O
growing	O
a	O
thermal	O
bubble	O
in	O
the	O
microchannel	O
,	O
to	O
isolate	O
the	O
microchamber	O
from	O
the	O
liquid	O
supply	O
and	O
neighboring	O
chambers	O
.	O
Between	O
firings	O
,	O
however	O
,	O
the	O
virtual	B-KEY
valve	I-KEY
opens	O
,	O
by	O
collapsing	O
the	O
bubble	O
,	O
to	O
reduce	O
flow	B-KEY
restriction	I-KEY
for	O
fast	O
refilling	O
of	O
the	O
microchamber	O
.	O
The	O
use	O
of	O
bubble	B-KEY
valves	I-KEY
brings	O
about	O
fast	O
and	O
reliable	O
device	O
operation	O
without	O
imposing	O
the	O
significant	O
complication	O
fabrication	O
of	O
physical	O
microvalves	O
would	O
call	O
for	O
.	O
In	O
addition	O
,	O
through	O
a	O
special	O
heater	O
configuration	O
and	O
chamber	B-KEY
designs	I-KEY
,	O
bubbles	O
surrounding	O
the	O
nozzle	O
cut	O
off	O
the	O
tail	O
of	O
the	O
droplets	O
being	O
ejected	O
and	O
completely	O
eliminate	O
satellite	B-KEY
droplets	I-KEY
.	O
A	O
simple	O
one-dimensional	O
model	O
of	O
the	O
operation	O
of	O
the	O
microinjector	O
is	O
used	O
to	O
estimate	O
the	O
bubble	O
formation	O
and	O
liquid	B-KEY
refilling	I-KEY

Where	O
tech	O
is	O
cheap	O
-LSB-	O
servers	O
-RSB-	O
Talk	O
,	O
consultancy	O
,	O
support	O
,	O
not	O
tech	O
is	O
the	O
expensive	O
part	O
of	O
network	B-KEY
installations	I-KEY
.	O
It	O
's	O
a	O
good	O
job	O
that	O
small-scale	B-KEY
servers	I-KEY
can	O
either	O
be	O
remotely	O
managed	B-KEY
,	O
or	O
require	O
little	O
actual	O
management	B-KEY

Scheduling	B-KEY
schemes	O
for	O
an	O
integrated	O
flight	O
and	O
propulsion	B-KEY
control	I-KEY
system	O
We	O
describe	O
two	O
schemes	O
for	O
scheduling	B-KEY
an	O
integrated	O
flight	O
and	O
propulsion	B-KEY
control	I-KEY
system	O
for	O
an	O
experimental	O
vertical/short	O
take-off	O
and	O
landing	O
-LRB-	O
V/STOL	O
-RRB-	O
aircraft	O
concept	O
in	O
the	O
acceleration	O
from	O
hover	O
-LRB-	O
0-120	O
kn	O
-RRB-	O
flight	O
phase	O
.	O
Multivariable	O
integrated	O
flight	O
and	O
propulsion	B-KEY
controllers	I-KEY
are	O
designed	O
at	O
several	O
points	O
over	O
the	O
V/STOL	O
envelope	O
and	O
implemented	O
as	O
exact	O
plant	O
observers	B-KEY
with	O
state	B-KEY
feedback	I-KEY
.	O
In	O
the	O
first	O
scheduling	B-KEY
scheme	O
,	O
the	O
values	O
of	O
the	O
state	B-KEY
feedback	I-KEY
and	O
observer	B-KEY
gain	O
matrices	O
are	O
interpolated	O
between	O
the	O
fixed-point	O
designs	O
as	O
a	O
function	O
of	O
aircraft	O
speed	O
.	O
In	O
the	O
second	O
approach	O
,	O
the	O
control	O
signals	O
produced	O
by	O
the	O
different	O
fixed-point	B-KEY
controllers	I-KEY
are	O
blended	O
,	O
allowing	O
a	O
significant	O
reduction	O
in	O
the	O
order	O
of	O
the	O
scheduled	B-KEY
controllers	O
.	O
Both	O
scheduling	B-KEY
schemes	O
are	O
shown	O
in	O
nonlinear	O
simulation	O
to	O
provide	O
excellent	O
handling	O
qualities	O
as	O
the	O
aircraft	O
accelerates	O
from	O
the	O
hover	O

Extracting	O
straight	B-KEY
road	I-KEY
structure	I-KEY
in	O
urban	B-KEY
environments	I-KEY
using	O
IKONOS	B-KEY
satellite	I-KEY
imagery	I-KEY
We	O
discuss	O
a	O
fully	B-KEY
automatic	I-KEY
technique	I-KEY
for	O
extracting	O
roads	O
in	O
urban	B-KEY
environments	I-KEY
.	O
The	O
method	O
has	O
its	O
bases	O
in	O
a	O
vegetation	B-KEY
mask	I-KEY
derived	O
from	O
multispectral	O
IKONOS	O
data	O
and	O
in	O
texture	B-KEY
derived	O
from	O
panchromatic	B-KEY
IKONOS	I-KEY
data	I-KEY
.	O
These	O
two	O
techniques	O
together	O
are	O
used	O
to	O
distinguish	O
road	B-KEY
pixels	I-KEY
.	O
We	O
then	O
move	O
from	O
individual	O
pixels	O
to	O
an	O
object-based	B-KEY
representation	I-KEY
that	O
allows	O
reasoning	O
on	O
a	O
higher	O
level	O
.	O
Recognition	O
of	O
individual	O
segments	O
and	O
intersections	O
and	O
the	O
relationships	O
among	O
them	O
are	O
used	O
to	O
determine	O
underlying	O
road	O
structure	O
and	O
to	O
then	O
logically	O
hypothesize	O
the	O
existence	O
of	O
additional	O
road	B-KEY
network	I-KEY
components	I-KEY
.	O
We	O
show	O
results	O
on	O
an	O
image	O
of	O
San	B-KEY
Diego	I-KEY
,	O
California	O
.	O
The	O
object-based	B-KEY
processing	I-KEY
component	I-KEY
may	O
be	O
adapted	O
to	O
utilize	O
other	O
basis	O
techniques	O
as	O
well	O
,	O
and	O
could	O
be	O
used	O
to	O
build	O
a	O
road	O
network	O
in	O
any	O
scene	O
having	O
a	O
straight-line	B-KEY
structured	I-KEY
topology	I-KEY

Recognition	O
of	O
finite	O
simple	O
groups	O
S/sub	O
4	O
/	O
-LRB-	O
q	O
-RRB-	O
by	O
their	O
element	B-KEY
orders	I-KEY
It	O
is	O
proved	O
that	O
among	O
simple	O
groups	O
S/sub	O
4	O
/	O
-LRB-	O
q	O
-RRB-	O
in	O
the	O
class	O
of	O
finite-groups	O
,	O
only	O
the	O
groups	O
S/sub	O
4	O
/	O
-LRB-	O
3/sup	O
n	O
/	O
-RRB-	O
,	O
where	O
n	O
is	O
an	O
odd	O
number	O
greater	O
than	O
unity	O
,	O
are	O
recognizable	O
by	O
a	O
set	O
of	O
their	O
element	B-KEY
orders	I-KEY
.	O
It	O
is	O
also	O
shown	O
that	O
simple	O
groups	O
U/sub	O
3	O
/	O
-LRB-	O
9	O
-RRB-	O
,	O
/	O
sup	O
3/D/sub	O
4	O
/	O
-LRB-	O
2	O
-RRB-	O
,	O
G/sub	O
2	O
/	O
-LRB-	O
4	O
-RRB-	O
,	O
S/sub	O
6	O
/	O
-LRB-	O
3	O
-RRB-	O
,	O
F/sub	O
4	O
/	O
-LRB-	O
2	O
-RRB-	O
,	O
and	O
/	O
sup	O
2/E/sub	O
6	O
/	O
-LRB-	O
2	O
-RRB-	O
are	O
recognizable	O
,	O
but	O
L/sub	O
3	O
/	O
-LRB-	O
3	O
-RRB-	O
is	O
not	O

A	O
conference	B-KEY
's	O
impact	O
on	O
undergraduate	B-KEY
female	I-KEY
students	I-KEY
In	O
September	O
of	O
2000	O
,	O
the	O
3rd	O
Grace	O
Hopper	O
Celebration	O
of	O
Women	O
in	O
Computing	O
was	O
held	O
in	O
Cape	O
Cod	O
,	O
Massachusetts	O
.	O
Along	O
with	O
a	O
colleague	O
from	O
a	O
nearby	O
university	O
,	O
we	O
accompanied	O
seven	O
of	O
our	O
female	O
undergraduate	O
students	O
to	O
this	O
conference	B-KEY
.	O
This	O
paper	O
reports	O
on	O
how	O
the	O
conference	B-KEY
experience	O
immediately	O
affected	O
these	O
students	O
-	O
what	O
impressed	O
them	O
,	O
what	O
scared	O
them	O
,	O
what	O
it	O
clarified	O
for	O
them	O
.	O
It	O
also	O
reports	O
on	O
how	O
the	O
context	O
in	O
which	O
these	O
students	O
currently	O
evaluate	O
their	O
ability	O
,	O
potential	O
and	O
opportunity	O
in	O
computer	O
science	O
is	O
different	O
now	O
from	O
what	O
it	O
was	O
before	O
the	O
conference	B-KEY
.	O
Hopefully	O
,	O
by	O
understanding	O
their	O
experience	O
,	O
we	O
can	O
gain	O
some	O
insight	O
into	O
things	O
we	O
can	O
do	O
for	O
all	O
of	O
our	O
undergraduate	B-KEY
female	I-KEY
students	I-KEY
to	O
better	O
support	O
their	O
computer	O
science	O
and	O
engineering	B-KEY
education	I-KEY

Himalayan	O
information	O
system	O
:	O
a	O
proposed	O
model	O
The	O
information	B-KEY
explosion	I-KEY
and	O
the	O
development	O
in	O
information	B-KEY
technology	I-KEY
force	O
us	O
to	O
develop	O
information	O
systems	O
in	O
various	O
fields	O
.	O
The	O
research	O
on	O
Himalaya	O
has	O
achieved	O
phenomenal	O
growth	O
in	O
recent	O
years	O
in	O
India	B-KEY
.	O
The	O
information	B-KEY
requirements	I-KEY
of	O
Himalayan	O
researchers	O
are	O
divergent	O
in	O
nature	O
.	O
In	O
order	O
to	O
meet	O
these	O
divergent	O
needs	O
,	O
all	O
information	O
generated	O
in	O
various	O
Himalayan	O
research	O
institutions	O
has	O
to	O
be	O
collected	O
and	O
organized	O
to	O
facilitate	O
free	O
flow	O
of	O
information	O
.	O
This	O
paper	O
describes	O
the	O
need	O
for	O
a	O
system	O
for	O
Himalayan	O
information	O
.	O
It	O
also	O
presents	O
the	O
objectives	O
of	O
Himalayan	O
information	O
system	O
-LRB-	O
HIMIS	B-KEY
-RRB-	O
.	O
It	O
discusses	O
in	O
brief	O
the	O
idea	O
of	O
setting	O
up	O
a	O
HIMIS	B-KEY
and	O
explains	O
its	O
utility	O
to	O
the	O
users	O
.	O
It	O
appeals	O
to	O
the	O
government	B-KEY
for	O
supporting	O
the	O
development	O
of	O
such	O
system	O

Design	O
and	O
implementation	O
of	O
a	O
new	O
sliding-mode	B-KEY
observer	I-KEY
for	O
speed-sensorless	O
control	O
of	O
induction	O
machine	O
In	O
this	O
letter	O
,	O
a	O
new	O
sliding-mode-sensorless	O
control	O
algorithm	O
is	O
proposed	O
for	O
the	O
field-oriented	O
induction	O
machine	O
drive	O
.	O
In	O
the	O
proposed	O
algorithm	O
,	O
the	O
terms	O
containing	O
flux	B-KEY
,	O
speed	B-KEY
,	O
and	O
rotor	B-KEY
time	I-KEY
constant	I-KEY
,	O
which	O
are	O
common	O
in	O
both	O
current	O
and	O
flux	B-KEY
equations	O
,	O
in	O
the	O
current	O
model	O
of	O
the	O
induction	O
machine	O
are	O
estimated	O
by	O
a	O
sliding	O
function	O
.	O
The	O
flux	B-KEY
and	O
speed	B-KEY
estimation	O
accuracy	O
is	O
guaranteed	O
when	O
the	O
error	O
between	O
the	O
actual	O
current	O
and	O
observed	O
current	O
converges	O
to	O
zero	O
.	O
Hence	O
,	O
the	O
fourth-order	O
system	O
is	O
reduced	O
to	O
two	O
second-order	O
systems	O
,	O
and	O
the	O
speed	B-KEY
estimation	O
becomes	O
very	O
simple	O
and	O
robust	O
to	O
the	O
parameter	B-KEY
uncertainties	I-KEY
.	O
The	O
new	O
approach	O
is	O
verified	O
by	O
simulation	O
and	O
experimental	O
results	O

On	O
fuzzy	O
and	O
probabilistic	B-KEY
control	I-KEY
charts	I-KEY
In	O
this	O
article	O
,	O
different	O
procedures	O
of	O
constructing	O
control	O
charts	O
for	O
linguistic	B-KEY
data	I-KEY
,	O
based	O
on	O
fuzzy	O
and	O
probability	O
theory	O
,	O
are	O
discussed	O
.	O
Three	O
sets	O
of	O
membership	B-KEY
functions	I-KEY
,	O
with	O
different	O
degrees	O
of	O
fuzziness	O
,	O
are	O
proposed	O
for	O
fuzzy	O
approaches	O
.	O
A	O
comparison	O
between	O
fuzzy	O
and	O
probability	O
approaches	O
,	O
based	O
on	O
the	O
Average	B-KEY
Run	I-KEY
Length	I-KEY
and	O
samples	O
under	O
control	O
,	O
is	O
conducted	O
for	O
real	O
data	O
.	O
Contrary	O
to	O
the	O
conclusions	O
of	O
Raz	O
and	O
Wang	O
-LRB-	O
1990	O
-RRB-	O
the	O
choice	O
of	O
degree	O
of	O
fuzziness	O
affected	O
the	O
sensitivity	B-KEY
of	O
control	O
charts	O

PDF	B-KEY
subscriptions	I-KEY
bolster	O
revenue	O
In	O
1999	O
SD	B-KEY
Times	I-KEY
offered	O
prospective	O
subscribers	O
the	O
option	O
of	O
receiving	O
their	O
issues	O
as	O
Adobe	B-KEY
Acrobat	I-KEY
PDF	I-KEY
files	I-KEY
.	O
What	O
set	O
the	O
proposal	O
apart	O
from	O
what	O
other	O
publishers	O
were	O
doing	O
electronically	O
on	O
the	O
Web	O
was	O
that	O
readers	O
would	O
get	O
the	O
entire	O
version	O
of	O
the	O
paper-including	O
both	O
advertising	O
and	O
editorial	O
just	O
as	O
it	O
looked	O
when	O
it	O
was	O
laid	O
out	O
and	O
went	O
to	O
press	O
.	O
SD	B-KEY
Times	I-KEY
is	O
only	O
one	O
of	O
a	O
small	O
,	O
but	O
growing	O
,	O
number	O
of	O
publications	O
that	O
are	O
taking	O
on	O
the	O
electronic	O
world	O
and	O
finding	O
success	O
.	O
In	O
the	O
past	O
six	O
months	O
alone	O
,	O
the	O
New	O
York	O
Times	O
,	O
Popular	O
Mechanics	O
,	O
trade	O
magazine	B-KEY
Electronic	O
Buyers	O
'	O
News	O
,	O
and	O
the	O
Harvard	O
Business	O
Review	O
have	O
launched	O
digital	B-KEY
versions	I-KEY
of	O
their	O
newspapers	B-KEY
and	O
magazines	B-KEY
to	O
augment	O
their	O
online	O
and	O
print	O
versions	O
.	O
The	O
reasons	O
are	O
as	O
varied	O
as	O
the	O
publishers	O
themselves	O
.	O
Some	O
companies	O
are	O
finding	O
that	O
readers	O
do	O
n't	O
like	O
their	O
Web-based	O
versions	O
either	O
due	O
to	O
poor	O
navigation	O
or	O
missing	O
graphics	O
and	O
images	O
.	O
Others	O
want	O
to	O
expand	O
their	O
publications	O
nationally	O
and	O
internationally	O
,	O
but	O
do	O
n't	O
want	O
the	O
added	O
cost	O
of	O
postage	O
and	O
printing	O
.	O
Still	O
others	O
are	O
looking	O
for	O
ways	O
to	O
give	O
advertisers	O
additional	O
visibility	O
and	O
boost	O
advertising	O
and	O
subscription	O
revenues	O
.	O
No	O
matter	O
what	O
the	O
reason	O
,	O
it	O
's	O
a	O
trend	O
worth	O
watching	O

Universal	B-KEY
simulation	I-KEY
of	O
Hamiltonian	B-KEY
dynamics	I-KEY
for	O
quantum	B-KEY
systems	I-KEY
with	O
finite-dimensional	O
state	O
spaces	O
What	O
interactions	O
are	O
sufficient	O
to	O
simulate	O
arbitrary	O
quantum	B-KEY
dynamics	I-KEY
in	O
a	O
composite	B-KEY
quantum	I-KEY
system	I-KEY
?	O
Dodd	O
et	O
al.	O
-LSB-	O
Phys	O
.	O
Rev.	O
A	O
65	O
,	O
040301	O
-LRB-	O
R	O
-RRB-	O
-LRB-	O
2002	O
-RRB-	O
-RSB-	O
provided	O
a	O
partial	O
solution	O
to	O
this	O
problem	O
in	O
the	O
form	O
of	O
an	O
efficient	O
algorithm	O
to	O
simulate	O
any	O
desired	O
two-body	O
Hamiltonian	O
evolution	O
using	O
any	O
fixed	O
two-body	O
entangling	O
N-qubit	O
Hamiltonian	O
,	O
and	O
local	O
unitaries	O
.	O
We	O
extend	O
this	O
result	O
to	O
the	O
case	O
where	O
the	O
component	O
systems	O
are	O
qudits	O
,	O
that	O
is	O
,	O
have	O
D	O
dimensions	O
.	O
As	O
a	O
consequence	O
we	O
explain	O
how	O
universal	B-KEY
quantum	I-KEY
computation	I-KEY
can	O
be	O
performed	O
with	O
any	O
fixed	B-KEY
two-body	I-KEY
entangling	I-KEY
N-qudit	I-KEY
Hamiltonian	I-KEY
,	O
and	O
local	B-KEY
unitaries	I-KEY

Entanglement	B-KEY
measures	I-KEY
with	O
asymptotic	B-KEY
weak-monotonicity	I-KEY
as	O
lower	O
-LRB-	O
upper	O
-RRB-	O
bound	O
for	O
the	O
entanglement	B-KEY
of	I-KEY
cost	I-KEY
-LRB-	O
distillation	B-KEY
-RRB-	O
We	O
propose	O
entanglement	B-KEY
measures	I-KEY
with	O
asymptotic	B-KEY
weak-monotonicity	I-KEY
.	O
We	O
show	O
that	O
a	O
normalized	O
form	O
of	O
entanglement	B-KEY
measures	I-KEY
with	O
the	O
asymptotic	B-KEY
weak-monotonicity	I-KEY
are	O
lower	O
-LRB-	O
upper	O
-RRB-	O
bound	O
for	O
the	O
entanglement	B-KEY
of	I-KEY
cost	I-KEY
-LRB-	O
distillation	B-KEY
-RRB-	O

Robust	B-KEY
speech	I-KEY
recognition	I-KEY
using	O
probabilistic	B-KEY
union	I-KEY
models	I-KEY
This	O
paper	O
introduces	O
a	O
new	O
statistical	O
approach	O
,	O
namely	O
the	O
probabilistic	B-KEY
union	I-KEY
model	I-KEY
,	O
for	O
speech	O
recognition	O
involving	O
partial	O
,	O
unknown	O
frequency-band	O
corruption	O
.	O
Partial	B-KEY
frequency-band	I-KEY
corruption	I-KEY
accounts	O
for	O
the	O
effect	O
of	O
a	O
family	O
of	O
real-world	O
noises	O
.	O
Previous	O
methods	O
based	O
on	O
the	O
missing	B-KEY
feature	I-KEY
theory	I-KEY
usually	O
require	O
the	O
identity	O
of	O
the	O
noisy	B-KEY
bands	I-KEY
.	O
This	O
identification	O
can	O
be	O
difficult	O
for	O
unexpected	O
noise	O
with	O
unknown	O
,	O
time-varying	B-KEY
band	I-KEY
characteristics	I-KEY
.	O
The	O
new	O
model	B-KEY
combines	O
the	O
local	B-KEY
frequency-band	I-KEY
information	I-KEY
based	O
on	O
the	O
union	O
of	O
random	O
events	O
,	O
to	O
reduce	O
the	O
dependence	O
of	O
the	O
model	B-KEY
on	O
information	O
about	O
the	O
noise	O
.	O
This	O
model	B-KEY
partially	O
accomplishes	O
the	O
target	O
:	O
offering	O
robustness	O
to	O
partial	B-KEY
frequency-band	I-KEY
corruption	I-KEY
,	O
while	O
requiring	O
no	O
information	O
about	O
the	O
noise	O
.	O
This	O
paper	O
introduces	O
the	O
theory	O
and	O
implementation	O
of	O
the	O
union	O
model	B-KEY
,	O
and	O
is	O
focused	O
on	O
several	O
important	O
advances	O
.	O
These	O
new	O
developments	O
include	O
a	O
new	O
algorithm	O
for	O
automatic	B-KEY
order	I-KEY
selection	I-KEY
,	O
a	O
generalization	O
of	O
the	O
modeling	B-KEY
principle	O
to	O
accommodate	O
partial	B-KEY
feature	I-KEY
stream	I-KEY
corruption	I-KEY
,	O
and	O
a	O
combination	O
of	O
the	O
union	O
model	B-KEY
with	O
conventional	O
noise	B-KEY
reduction	I-KEY
techniques	I-KEY
to	O
deal	O
with	O
a	O
mixture	O
of	O
stationary	B-KEY
noise	I-KEY
and	O
unknown	O
,	O
nonstationary	B-KEY
noise	I-KEY
.	O
For	O
the	O
evaluation	O
,	O
we	O
used	O
the	O
TIDIGITS	B-KEY
database	I-KEY
for	O
speaker-independent	B-KEY
connected	I-KEY
digit	I-KEY
recognition	I-KEY
.	O
The	O
utterances	O
were	O
corrupted	O
by	O
various	O
types	O
of	O
additive	B-KEY
noise	I-KEY
,	O
stationary	O
or	O
time-varying	O
,	O
assuming	O
no	O
knowledge	O
about	O
the	O
noise	B-KEY
characteristics	I-KEY
.	O
The	O
results	O
indicate	O
that	O
the	O
new	O
model	B-KEY
offers	O
significantly	O
improved	O
robustness	O
in	O
comparison	O
to	O
other	O
models	B-KEY

The	O
mutual	O
effects	O
of	O
grid	O
and	O
wind	B-KEY
turbine	I-KEY
voltage	I-KEY
stability	I-KEY
control	I-KEY
This	O
note	O
considers	O
the	O
results	O
of	O
wind	B-KEY
turbine	I-KEY
modelling	I-KEY
and	O
power	B-KEY
system	I-KEY
stability	I-KEY
investigations	O
.	O
Voltage	O
stability	O
of	O
the	O
power	B-KEY
grid	I-KEY
with	O
grid-connected	B-KEY
wind	I-KEY
turbines	I-KEY
will	O
be	O
improved	O
by	O
using	O
blade	B-KEY
angle	I-KEY
control	I-KEY
for	O
a	O
temporary	O
reduction	O
of	O
the	O
wind	O
turbine	O
power	O
during	O
and	O
shortly	O
after	O
a	O
short	B-KEY
circuit	I-KEY
fault	I-KEY
in	O
the	O
grid	O

Market	B-KEY
watch	O
-	O
air	B-KEY
conditioning	I-KEY
After	O
a	O
boom	O
period	O
in	O
the	O
late	O
nineties	O
,	O
the	O
air	B-KEY
conditioning	I-KEY
market	B-KEY
finds	O
itself	O
in	O
something	O
of	O
a	O
lull	O
at	O
present	O
,	O
but	O
manufacturers	O
are	O
n't	O
panicking	O

A	O
combinatorial	O
,	O
graph-based	O
solution	O
method	O
for	O
a	O
class	O
of	O
continuous-time	B-KEY
optimal	I-KEY
control	I-KEY
problems	I-KEY
The	O
paper	O
addresses	O
a	O
class	O
of	O
continuous-time	O
,	O
optimal	O
control	O
problems	O
whose	O
solutions	O
are	O
typically	O
characterized	O
by	O
both	O
bang-bang	O
and	O
``	O
singular	O
''	O
control	O
regimes	O
.	O
Analytical	O
study	O
and	O
numerical	B-KEY
computation	I-KEY
of	O
such	O
solutions	O
are	O
very	O
difficult	O
and	O
far	O
from	O
complete	O
when	O
only	O
techniques	O
from	O
control	O
theory	O
are	O
used	O
.	O
This	O
paper	O
solves	O
optimal	O
control	O
problems	O
by	O
reducing	O
them	O
to	O
the	O
combinatorial	B-KEY
search	I-KEY
for	O
the	O
shortest	O
path	O
in	O
a	O
specially	O
constructed	O
graph	O
.	O
Since	O
the	O
nodes	O
of	O
the	O
graph	O
are	O
weighted	O
in	O
a	O
sequence-dependent	B-KEY
manner	I-KEY
,	O
we	O
extend	O
the	O
classical	O
,	O
shortest-path	O
algorithm	O
to	O
our	O
case	O
.	O
The	O
proposed	O
solution	O
method	O
is	O
currently	O
limited	O
to	O
single-state	B-KEY
problems	I-KEY
with	O
multiple	B-KEY
control	I-KEY
functions	I-KEY
.	O
A	O
production	B-KEY
planning	I-KEY
problem	I-KEY
and	O
a	O
train	B-KEY
operation	I-KEY
problem	I-KEY
are	O
optimally	O
solved	O
to	O
illustrate	O
the	O
method	O

Separate	O
accounts	O
go	O
mainstream	O
-LSB-	O
investment	B-KEY
-RSB-	O
New	O
entrants	O
are	O
shaking	O
up	O
the	O
separate-account	B-KEY
industry	I-KEY
by	O
supplying	O
Web-based	B-KEY
platforms	I-KEY
that	O
give	O
advisers	O
the	O
tools	O
to	O
pick	O
independent	B-KEY
money	I-KEY
managers	I-KEY

Simulation	B-KEY
of	O
cardiovascular	B-KEY
physiology	I-KEY
:	O
the	O
diastolic	B-KEY
function	I-KEY
-LRB-	O
s	O
-RRB-	O
of	O
the	O
heart	B-KEY
The	O
cardiovascular	O
system	O
was	O
simulated	B-KEY
by	O
using	O
an	O
equivalent	B-KEY
electronic	I-KEY
circuit	I-KEY
.	O
Four	O
sets	O
of	O
simulations	B-KEY
were	O
performed	O
.	O
The	O
basic	O
variables	O
investigated	O
were	O
cardiac	B-KEY
output	I-KEY
and	O
stroke	B-KEY
volume	I-KEY
.	O
They	O
were	O
studied	O
as	O
functions	O
-LRB-	O
i	O
-RRB-	O
of	O
right	B-KEY
ventricular	I-KEY
capacitance	I-KEY
and	O
negative	B-KEY
intrathoracic	I-KEY
pressure	I-KEY
;	O
-LRB-	O
ii	O
-RRB-	O
of	O
left	B-KEY
ventricular	I-KEY
relaxation	I-KEY
and	O
of	O
heart	B-KEY
rate	O
;	O
and	O
-LRB-	O
iii	O
-RRB-	O
of	O
left	O
ventricle	O
failure	O
.	O
It	O
seems	O
that	O
a	O
satisfactory	O
simulation	B-KEY
of	O
systolic	O
and	O
diastolic	B-KEY
functions	I-KEY
of	O
the	O
heart	B-KEY
is	O
possible	O
.	O
Presented	O
simulations	B-KEY
improve	O
our	O
understanding	O
of	O
the	O
role	O
of	O
the	O
capacitance	O
of	O
both	O
ventricles	O
and	O
of	O
the	O
diastolic	B-KEY
relaxation	I-KEY
in	O
cardiovascular	B-KEY
physiology	I-KEY

A	O
modal	B-KEY
logic	I-KEY
for	O
indiscernibility	B-KEY
and	O
complementarity	B-KEY
in	O
information	B-KEY
systems	I-KEY
In	O
this	O
paper	O
,	O
we	O
study	O
indiscernibility	B-KEY
relations	O
and	O
complementarity	B-KEY
relations	O
in	O
information	B-KEY
systems	I-KEY
,	O
The	O
first-order	B-KEY
characterization	I-KEY
of	O
indiscernibility	B-KEY
and	O
complementarity	B-KEY
is	O
obtained	O
through	O
a	O
duality	B-KEY
result	I-KEY
between	O
information	B-KEY
systems	I-KEY
and	O
certain	O
structures	O
of	O
relational	B-KEY
type	I-KEY
characterized	O
by	O
first-order	B-KEY
conditions	I-KEY
.	O
The	O
modal	O
analysis	O
of	O
indiscernibility	B-KEY
and	O
complementarity	B-KEY
is	O
performed	O
through	O
a	O
modal	B-KEY
logic	I-KEY
which	O
modalities	O
correspond	O
to	O
indiscernibility	B-KEY
relations	O
and	O
complementarity	B-KEY
relations	O
in	O
information	B-KEY
systems	I-KEY

3D	B-KEY
reconstruction	I-KEY
from	O
uncalibrated-camera	B-KEY
optical	I-KEY
flow	I-KEY
and	O
its	O
reliability	B-KEY
evaluation	I-KEY
We	O
present	O
a	O
scheme	O
for	O
reconstructing	O
a	O
3D	O
structure	O
from	O
optical	O
flow	O
observed	O
by	O
a	O
camera	O
with	O
an	O
unknown	O
focal	O
length	O
in	O
a	O
statistically	O
optimal	O
way	O
as	O
well	O
as	O
evaluating	O
the	O
reliability	O
of	O
the	O
computed	O
shape	O
.	O
First	O
,	O
the	O
flow	B-KEY
fundamental	I-KEY
matrices	I-KEY
are	O
optimally	O
computed	O
from	O
the	O
observed	O
flow	O
.	O
They	O
are	O
then	O
decomposed	O
into	O
the	O
focal	O
length	O
,	O
its	O
rate	O
of	O
change	O
,	O
and	O
the	O
motion	B-KEY
parameters	I-KEY
.	O
Next	O
,	O
the	O
flow	O
is	O
optimally	O
corrected	O
so	O
that	O
it	O
satisfies	O
the	O
epipolar	B-KEY
equation	I-KEY
exactly	O
.	O
Finally	O
,	O
the	O
3D	O
positions	O
are	O
computed	O
,	O
and	O
their	O
covariance	B-KEY
matrices	I-KEY
are	O
evaluated	O
.	O
By	O
simulations	O
and	O
real-image	B-KEY
experiments	I-KEY
,	O
we	O
test	O
the	O
performance	O
of	O
our	O
system	O
and	O
observe	O
how	O
the	O
normalization	B-KEY
-LRB-	O
gauge	O
-RRB-	O
for	O
removing	O
indeterminacy	O
affects	O
the	O
description	O
of	O
uncertainty	O

Image	B-KEY
fusion	I-KEY
between	O
/	O
sup	O
18/FDG-PET	O
and	O
MRI/CT	B-KEY
for	O
radiotherapy	B-KEY
planning	I-KEY
of	O
oropharyngeal	O
and	O
nasopharyngeal	B-KEY
carcinomas	I-KEY
Accurate	O
diagnosis	O
of	O
tumor	O
extent	O
is	O
important	O
in	O
three-dimensional	O
conformal	O
radiotherapy	O
.	O
This	O
study	O
reports	O
the	O
use	O
of	O
image	B-KEY
fusion	I-KEY
between	O
-LRB-	O
18	O
-RRB-	O
F-fluoro-2-deoxy-D-glucose	O
positron	O
emission	O
tomography	O
-LRB-	O
/	O
sup	O
18/FDG-PET	O
-RRB-	O
and	O
magnetic	O
resonance	O
imaging/computed	O
tomography	O
-LRB-	O
MRI/CT	O
-RRB-	O
for	O
better	O
targets	O
delineation	O
in	O
radiotherapy	O
planning	O
of	O
head-and-neck	O
cancers	O
.	O
The	O
subjects	O
consisted	O
of	O
12	O
patients	O
with	O
oropharyngeal	B-KEY
carcinoma	I-KEY
and	O
9	O
patients	O
with	O
nasopharyngeal	B-KEY
carcinoma	I-KEY
-LRB-	O
NPC	O
-RRB-	O
who	O
were	O
treated	O
with	O
radical	O
radiotherapy	O
between	O
July	O
1999	O
and	O
February	O
2001	O
.	O
Image	B-KEY
fusion	I-KEY
between	O
/	O
sup	O
18/FDG-PET	O
and	O
MRI/CT	B-KEY
was	O
performed	O
using	O
an	O
automatic	O
multimodality	O
image	O
registration	O
algorithm	O
,	O
which	O
used	O
the	O
brain	O
as	O
an	O
internal	O
reference	O
for	O
registration	O
.	O
Gross	O
tumor	O
volume	O
-LRB-	O
GTV	O
-RRB-	O
was	O
determined	O
based	O
on	O
clinical	O
examination	O
and	O
/	O
sup	O
18/FDG	O
uptake	O
on	O
the	O
fusion	O
images	O
.	O
Clinical	O
target	O
volume	O
-LRB-	O
CTV	O
-RRB-	O
was	O
determined	O
following	O
the	O
usual	O
pattern	O
of	O
lymph	O
node	O
spread	O
for	O
each	O
disease	O
entity	O
along	O
with	O
the	O
clinical	O
presentation	O
of	O
each	O
patient	O
.	O
Except	O
for	O
3	O
cases	O
with	O
superficial	B-KEY
tumors	I-KEY
,	O
all	O
the	O
other	O
primary	B-KEY
tumors	I-KEY
were	O
detected	O
by	O
/	O
sup	O
18/FDG-PET	O
.	O
The	O
GTV	O
volumes	O
for	O
primary	B-KEY
tumors	I-KEY
were	O
not	O
changed	O
by	O
image	B-KEY
fusion	I-KEY
in	O
19	O
cases	O
-LRB-	O
89	O
%	O
-RRB-	O
,	O
increased	O
by	O
49	O
%	O
in	O
one	O
NPC	O
,	O
and	O
decreased	O
by	O
45	O
%	O
in	O
another	O
NPC	O
.	O
Normal	B-KEY
tissue	I-KEY
sparing	I-KEY
was	O
more	O
easily	O
performed	O
based	O
on	O
clearer	O
GTV	O
and	O
CTV	O
determination	O
on	O
the	O
fusion	O
images	O
.	O
In	O
particular	O
,	O
parotid	O
sparing	O
became	O
possible	O
in	O
15	O
patients	O
-LRB-	O
71	O
%	O
-RRB-	O
whose	O
upper	O
neck	O
areas	O
near	O
the	O
parotid	B-KEY
glands	I-KEY
were	O
tumor-free	O
by	O
/	O
sup	O
18/FDG-PET	O
.	O
Within	O
a	O
mean	O
follow-up	O
period	O
of	O
18	O
months	O
,	O
no	O
recurrence	O
occurred	O
in	O
the	O
areas	O
defined	O
as	O
CTV	O
,	O
which	O
was	O
treated	O
prophylactically	O
,	O
except	O
for	O
1	O
patient	O
who	O
experienced	O
nodal	O
recurrence	O
in	O
the	O
CTV	O
and	O
simultaneous	B-KEY
primary	I-KEY
site	I-KEY
recurrence	I-KEY
.	O
In	O
conclusion	O
,	O
this	O
preliminary	O
study	O
showed	O
that	O
image	B-KEY
fusion	I-KEY
between	O
/	O
sup	O
18/FDG-PET	O
and	O
MRI/CT	B-KEY
was	O
useful	O
in	O
GTV	O
and	O
CTV	O
determination	O
in	O
conformal	O
RT	O
,	O
thus	O
sparing	O
normal	O
tissues	O

Girls	B-KEY
,	O
boys	B-KEY
,	O
and	O
computers	O
Today	O
North	O
American	O
girls	B-KEY
,	O
boys	B-KEY
,	O
teachers	B-KEY
,	O
and	O
parents	O
frequently	O
regard	O
computer	B-KEY
science	I-KEY
and	O
programming	B-KEY
as	O
something	O
boys	B-KEY
are	O
better	O
at	O
.	O
The	O
author	O
considers	O
how	O
many	O
of	O
the	O
factors	O
that	O
contribute	O
to	O
the	O
low	O
participation	O
of	O
women	B-KEY
in	O
computing	O
occur	O
first	O
,	O
and	O
perhaps	O
most	O
forcefully	O
,	O
in	O
childhood	B-KEY
.	O
She	O
presents	O
four	O
recommendations	O
to	O
address	O
the	O
situation	O

Optimization	B-KEY
of	O
advertising	B-KEY
expenses	I-KEY
in	O
the	O
functioning	O
of	O
an	O
insurance	B-KEY
company	I-KEY
With	O
the	O
use	O
of	O
Pontryagin	O
's	O
maximum	O
principle	O
,	O
a	O
problem	O
of	O
optimal	B-KEY
time	O
distribution	O
of	O
advertising	O
expenses	O
in	O
the	O
functioning	O
of	O
an	O
insurance	O
company	O
is	O
solved	O

Optimal	O
allocation	O
of	O
runs	O
in	O
a	O
simulation	B-KEY
metamodel	I-KEY
with	O
several	O
independent	B-KEY
variables	I-KEY
Cheng	O
and	O
Kleijnen	O
-LRB-	O
1999	O
-RRB-	O
propose	O
a	O
very	O
general	B-KEY
regression	I-KEY
metamodel	I-KEY
for	O
modelling	O
the	O
output	O
of	O
a	O
queuing	B-KEY
system	I-KEY
.	O
Its	O
main	O
limitations	O
are	O
that	O
the	O
regression	B-KEY
function	I-KEY
is	O
based	O
on	O
a	O
polynomial	O
and	O
that	O
it	O
can	O
use	O
only	O
one	O
independent	B-KEY
variable	I-KEY
.	O
These	O
limitations	O
are	O
removed	O
here	O
.	O
We	O
derive	O
an	O
explicit	O
formula	O
for	O
the	O
optimal	O
way	O
of	O
assigning	O
simulation	O
runs	O
to	O
the	O
different	O
design	O
points	O

Evaluation	O
of	O
the	O
usability	O
of	O
digital	O
maintenance	O
manuals	O
developed	O
without	O
either	O
user	O
input	O
or	O
a	O
task	B-KEY
analysis	I-KEY
The	O
primary	O
objective	O
was	O
to	O
investigate	O
the	O
value	O
that	O
can	O
be	O
added	O
to	O
a	O
low-cost	O
digital	O
maintenance	O
manual	O
by	O
the	O
addition	O
of	O
a	O
navigational	B-KEY
aid	I-KEY
.	O
Two	O
versions	O
of	O
a	O
digital	O
maintenance	O
manual	O
were	O
developed	O
,	O
the	O
difference	O
between	O
them	O
being	O
the	O
number	O
of	O
design	O
heuristics	O
observed	O
when	O
designing	O
navigational	B-KEY
aids	I-KEY
.	O
Neither	O
version	O
was	O
based	O
on	O
an	O
analysis	O
of	O
the	O
tasks	O
carried	O
out	O
by	O
users	O
,	O
nor	O
were	O
users	O
involved	O
in	O
the	O
design	O
process	O
.	O
Instead	O
,	O
the	O
manuals	O
were	O
developed	O
directly	O
from	O
the	O
digital	O
information	O
used	O
to	O
produce	O
the	O
existing	O
paper	O
manual	O
.	O
Usability	B-KEY
trials	I-KEY
were	O
carried	O
out	O
to	O
test	O
both	O
versions	O
according	O
to	O
the	O
time	O
taken	O
and	O
errors	O
committed	O
by	O
users	O
during	O
typical	O
information	B-KEY
retrieval	I-KEY
tasks	O
.	O
Users	O
were	O
questioned	O
to	O
determine	O
their	O
ease	O
of	O
use	O
-LRB-	O
EOU	O
-RRB-	O
perceptions	O
for	O
each	O
manual	O
.	O
The	O
main	O
outcomes	O
were	O
that	O
the	O
navigation	B-KEY
aid	I-KEY
used	O
in	O
the	O
second	O
version	O
reduced	O
the	O
time	O
taken	O
to	O
use	O
the	O
manual	O
but	O
increased	O
the	O
number	O
of	O
errors	O
made	O
by	O
users	O
.	O
The	O
navigational	B-KEY
aid	I-KEY
also	O
seemed	O
to	O
reduce	O
the	O
perceived	O
EOU	O
compared	O
with	O
the	O
first	O
version	O
.	O
In	O
both	O
cases	O
,	O
the	O
perceived	O
EOU	O
was	O
lower	O
than	O
for	O
a	O
previous	O
digital	O
manual	O
that	O
had	O
been	O
developed	O
using	O
a	O
task	B-KEY
analysis	I-KEY
and	O
user	O
input	O
.	O
The	O
paper	O
concludes	O
by	O
recommending	O
the	O
development	O
of	O
a	O
generic	B-KEY
task	I-KEY
model	I-KEY
of	O
user	B-KEY
interaction	I-KEY
with	O
digital	O
maintenance	O
manuals	O

Mapping	O
CCF	O
to	O
MARC21	B-KEY
:	O
an	O
experimental	O
approach	O
The	O
purpose	O
of	O
this	O
article	O
is	O
to	O
raise	O
and	O
address	O
a	O
number	O
of	O
issues	O
pertaining	O
to	O
the	O
conversion	O
of	O
Common	O
Communication	O
Format	O
-LRB-	O
CCF	O
-RRB-	O
into	O
MARC21	B-KEY
.	O
In	O
this	O
era	O
of	O
global	B-KEY
resource	I-KEY
sharing	I-KEY
,	O
exchange	O
of	O
bibliographic	O
records	O
from	O
one	O
system	O
to	O
another	O
is	O
imperative	O
in	O
today	O
's	O
library	B-KEY
communities	I-KEY
.	O
Instead	O
of	O
using	O
a	O
single	O
standard	B-KEY
to	O
create	O
machine-readable	B-KEY
catalogue	I-KEY
records	I-KEY
,	O
more	O
than	O
20	O
standards	B-KEY
have	O
emerged	O
and	O
are	O
being	O
used	O
by	O
different	O
institutions	O
.	O
Because	O
of	O
these	O
variations	O
in	O
standards	B-KEY
,	O
sharing	O
of	O
resources	O
and	O
transfer	O
of	O
data	O
from	O
one	O
system	O
to	O
another	O
among	O
the	O
institutions	O
locally	O
and	O
globally	O
has	O
become	O
a	O
significant	O
problem	O
.	O
Addressing	O
this	O
problem	O
requires	O
keeping	O
in	O
mind	O
that	O
countries	O
such	O
as	O
India	B-KEY
and	O
others	O
in	O
southeast	B-KEY
Asia	I-KEY
are	O
using	O
the	O
CCF	O
as	O
a	O
standard	B-KEY
for	O
creating	O
bibliographic	O
cataloguing	O
records	O
.	O
This	O
paper	O
describes	O
a	O
way	O
to	O
map	O
the	O
bibliographic	O
catalogue	O
records	O
from	O
CCF	O
to	O
MARC21	B-KEY
,	O
although	O
100	O
%	O
mapping	O
is	O
not	O
possible	O
.	O
In	O
addition	O
,	O
the	O
paper	O
describes	O
an	O
experimental	O
approach	O
that	O
enumerates	O
problems	O
that	O
may	O
occur	O
during	O
the	O
mapping	O
of	O
records/exchanging	O
of	O
records	O
and	O
how	O
these	O
problems	O
can	O
be	O
overcome	O

Server	O
safeguards	O
tax	O
service	O
Peterborough-based	O
tax	B-KEY
consultancy	I-KEY
IE	B-KEY
Taxguard	I-KEY
wanted	O
real-time	O
failover	O
protection	O
for	O
important	O
Windows-based	O
applications	O
.	O
Its	O
solution	O
was	O
to	O
implement	O
a	O
powerful	O
failover	B-KEY
server	I-KEY
from	O
UK	O
supplier	O
Neverfail	B-KEY
in	O
order	O
to	O
provide	O
real-time	O
backup	B-KEY
for	O
three	O
core	O
production	O
servers	O

A	O
component-based	B-KEY
software	I-KEY
configuration	I-KEY
management	I-KEY
model	I-KEY
and	O
its	O
supporting	O
system	O
Software	O
configuration	O
management	O
-LRB-	O
SCM	O
-RRB-	O
is	O
an	O
important	O
key	O
technology	O
in	O
software	B-KEY
development	I-KEY
.	O
Component-based	O
software	B-KEY
development	I-KEY
-LRB-	O
CBSD	O
-RRB-	O
is	O
an	O
emerging	O
paradigm	O
in	O
software	B-KEY
development	I-KEY
.	O
However	O
,	O
to	O
apply	O
CBSD	O
effectively	O
in	O
real	O
world	O
practice	O
,	O
supporting	O
SCM	O
in	O
CBSD	O
needs	O
to	O
be	O
further	O
investigated	O
.	O
In	O
this	O
paper	O
,	O
the	O
objects	O
that	O
need	O
to	O
be	O
managed	O
in	O
CBSD	O
is	O
analyzed	O
and	O
a	O
component-based	O
SCM	O
model	O
is	O
presented	O
.	O
In	O
this	O
model	O
,	O
components	O
,	O
as	O
the	O
integral	B-KEY
logical	I-KEY
constituents	I-KEY
in	O
a	O
system	O
,	O
are	O
managed	O
as	O
the	O
basic	O
configuration	O
items	O
in	O
SCM	O
,	O
and	O
the	O
relationships	O
between/among	O
components	O
are	O
defined	O
and	O
maintained	O
.	O
Based	O
on	O
this	O
model	O
,	O
a	O
configuration	O
management	O
system	O
is	O
implemented	O

Local	B-KEY
activity	I-KEY
criteria	I-KEY
for	O
discrete-map	B-KEY
CNN	I-KEY
Discrete-time	B-KEY
CNN	I-KEY
systems	I-KEY
are	O
studied	O
in	O
this	O
paper	O
by	O
the	O
application	O
of	O
Chua	O
's	O
local	O
activity	O
principle	O
.	O
These	O
systems	O
are	O
locally	O
active	O
everywhere	O
except	O
for	O
one	O
isolated	O
parameter	O
value	O
.	O
As	O
a	O
result	O
,	O
nonhomogeneous	B-KEY
spatiotemporal	I-KEY
patterns	I-KEY
may	O
be	O
induced	O
by	O
any	O
initial	O
setting	O
of	O
the	O
CNN	O
system	O
when	O
the	O
strength	O
of	O
the	O
system	O
diffusion	O
coupling	O
exceeds	O
a	O
critical	O
threshold	O
.	O
The	O
critical	B-KEY
coupling	I-KEY
coefficient	I-KEY
can	O
be	O
derived	O
from	O
the	O
loaded	B-KEY
cell	I-KEY
impedance	I-KEY
of	O
the	O
CNN	O
system	O
.	O
Three	O
well-known	O
1D	O
map	O
CNN	O
's	O
-LRB-	O
namely	O
,	O
the	O
logistic	B-KEY
map	I-KEY
CNN	I-KEY
,	O
the	O
magnetic	B-KEY
vortex	I-KEY
pinning	I-KEY
map	I-KEY
CNN	I-KEY
,	O
and	O
the	O
spiral	B-KEY
wave	I-KEY
reproducing	I-KEY
map	I-KEY
CNN	I-KEY
-RRB-	O
are	O
introduced	O
to	O
illustrate	O
the	O
applications	O
of	O
the	O
local	O
activity	O
principle	O
.	O
In	O
addition	O
,	O
we	O
use	O
the	O
cell	O
impedance	O
to	O
demonstrate	O
the	O
period-doubling	B-KEY
scenario	O
in	O
the	O
logistic	O
and	O
the	O
magnetic	O
vortex	O
pinning	O
maps	O

End-user	B-KEY
perspectives	I-KEY
on	O
the	O
uptake	O
of	O
computer	B-KEY
supported	I-KEY
cooperative	I-KEY
working	I-KEY
Researchers	O
in	O
information	B-KEY
systems	I-KEY
have	O
produced	O
a	O
rich	O
collection	O
of	O
meta-analyses	B-KEY
and	O
models	O
to	O
further	O
understanding	O
of	O
factors	O
influencing	O
the	O
uptake	O
of	O
information	B-KEY
technologies	I-KEY
.	O
In	O
the	O
domain	O
of	O
CSCW	B-KEY
,	O
however	O
,	O
these	O
models	O
have	O
largely	O
been	O
neglected	O
,	O
and	O
while	O
there	O
are	O
many	O
case	O
studies	O
,	O
no	O
systematic	O
account	O
of	O
uptake	O
has	O
been	O
produced	O
.	O
We	O
use	O
findings	O
from	O
information	B-KEY
systems	I-KEY
research	O
to	O
structure	O
a	O
meta-analysis	O
of	O
uptake	O
issues	O
as	O
reported	O
in	O
CSCW	B-KEY
case	O
studies	O
,	O
supplemented	O
by	O
a	O
detailed	O
re-examination	O
of	O
one	O
of	O
our	O
own	O
case	O
studies	O
from	O
this	O
perspective	O
.	O
This	O
shows	O
that	O
while	O
there	O
are	O
some	O
factors	O
which	O
seem	O
to	O
be	O
largely	O
specific	O
to	O
CSCW	B-KEY
introductions	O
,	O
many	O
of	O
the	O
case	O
study	O
results	O
are	O
very	O
similar	O
to	O
standard	O
IS	O
findings	O
.	O
We	O
conclude	O
by	O
suggesting	O
how	O
the	O
two	O
communities	O
of	O
researchers	O
might	O
build	O
on	O
each	O
other	O
's	O
work	O
,	O
and	O
finally	O
propose	O
activity	B-KEY
theory	I-KEY
as	O
a	O
means	O
of	O
integrating	O
the	O
two	O
perspectives	O

Fuzzy	B-KEY
systems	I-KEY
with	O
overlapping	B-KEY
Gaussian	I-KEY
concepts	I-KEY
:	O
Approximation	O
properties	O
in	O
Sobolev	O
norms	O
In	O
this	O
paper	O
the	O
approximating	O
capabilities	O
of	O
fuzzy	B-KEY
systems	I-KEY
with	O
overlapping	B-KEY
Gaussian	I-KEY
concepts	I-KEY
are	O
considered	O
.	O
The	O
target	O
function	O
is	O
assumed	O
to	O
be	O
sampled	O
either	O
on	O
a	O
regular	O
gird	O
or	O
according	O
to	O
a	O
uniform	O
probability	O
density	O
.	O
By	O
exploiting	O
a	O
connection	O
with	O
Radial	B-KEY
Basis	I-KEY
Functions	I-KEY
approximators	O
,	O
a	O
new	O
method	O
for	O
the	O
computation	O
of	O
the	O
system	O
coefficients	O
is	O
provided	O
,	O
showing	O
that	O
it	O
guarantees	O
uniform	O
approximation	O
of	O
the	O
derivatives	O
of	O
the	O
target	O
function	O

Heuristics	O
for	O
single-pass	B-KEY
welding	I-KEY
task	I-KEY
sequencing	I-KEY
Welding	O
task	O
sequencing	O
is	O
a	O
prerequisite	O
in	O
the	O
offline	B-KEY
programming	I-KEY
of	O
robot	B-KEY
arc	I-KEY
welding	I-KEY
.	O
Single-pass	B-KEY
welding	I-KEY
task	I-KEY
sequencing	I-KEY
can	O
be	O
modelled	O
as	O
a	O
modified	B-KEY
travelling	I-KEY
salesman	I-KEY
problem	I-KEY
.	O
Owing	O
to	O
the	O
difficulty	O
of	O
the	O
resulting	O
arc-routing	O
problems	O
,	O
effective	O
local	B-KEY
search	I-KEY
heuristics	I-KEY
are	O
developed	O
.	O
Computational	B-KEY
speed	I-KEY
becomes	O
important	O
because	O
robot	B-KEY
arc	I-KEY
welding	I-KEY
is	O
often	O
part	O
of	O
an	O
automated	B-KEY
process-planning	I-KEY
procedure	I-KEY
.	O
Generating	O
a	O
reasonable	O
solution	O
in	O
an	O
acceptable	O
time	O
is	O
necessary	O
for	O
effective	O
automated	O
process	O
planning	O
.	O
Several	O
different	O
heuristics	O
are	O
proposed	O
for	O
solving	O
the	O
welding	O
task-sequencing	O
problem	O
considering	O
both	O
productivity	B-KEY
and	O
the	O
potential	O
for	O
welding	B-KEY
distortion	I-KEY
.	O
Constructive	B-KEY
heuristics	I-KEY
based	O
on	O
the	O
nearest	B-KEY
neighbour	I-KEY
concept	I-KEY
and	O
tabu	B-KEY
search	I-KEY
heuristics	I-KEY
are	O
developed	O
and	O
enhanced	O
using	O
improvement	O
procedures	O
.	O
The	O
effectiveness	O
of	O
the	O
heuristics	O
developed	O
is	O
tested	O
and	O
verified	O
on	O
actual	O
welded	B-KEY
structure	I-KEY
problems	I-KEY
and	O
random	B-KEY
problems	I-KEY

Enhancing	O
the	O
reliability	O
of	O
modular	O
medium-voltage	O
drives	O
A	O
method	O
to	O
increase	O
the	O
reliability	O
of	O
modular	B-KEY
medium-voltage	I-KEY
induction	I-KEY
motor	I-KEY
drives	I-KEY
is	O
discussed	O
,	O
by	O
providing	O
means	O
to	O
bypass	O
a	O
failed	O
module	O
.	O
The	O
impact	O
on	O
reliability	O
is	O
shown	O
.	O
A	O
control	O
,	O
which	O
maximizes	O
the	O
output	O
voltage	O
available	O
after	O
bypass	O
,	O
is	O
described	O
,	O
and	O
experimental	O
results	O
are	O
given	O

Temp	O
IT	O
chief	O
rallies	O
troops	O
-LSB-	O
Mori	B-KEY
-RSB-	O
The	O
appointment	O
of	O
a	O
highly	O
qualified	O
interim	B-KEY
IT	I-KEY
manager	I-KEY
enabled	O
market	B-KEY
research	I-KEY
company	I-KEY
Mori	B-KEY
to	O
rapidly	O
restructure	O
its	O
IT	O
department	O
.	O
Now	O
the	O
resulting	O
improvements	O
are	O
allowing	O
it	O
to	O
support	O
an	O
increasing	O
role	O
for	O
technology	O
in	O
the	O
assimilation	O
and	O
analysis	O
of	O
market	O
research	O

Time-resolved	B-KEY
contrast-enhanced	I-KEY
imaging	I-KEY
with	O
isotropic	B-KEY
resolution	I-KEY
and	O
broad	B-KEY
coverage	I-KEY
using	O
an	O
undersampled	B-KEY
3D	I-KEY
projection	I-KEY
trajectory	I-KEY
Time-resolved	O
contrast-enhanced	O
3D	O
MR	O
angiography	O
-LRB-	O
MRA	O
-RRB-	O
methods	O
have	O
gained	O
in	O
popularity	O
but	O
are	O
still	O
limited	O
by	O
the	O
tradeoff	O
between	O
spatial	O
and	O
temporal	O
resolution	O
.	O
A	O
method	O
is	O
presented	O
that	O
greatly	O
reduces	O
this	O
tradeoff	O
by	O
employing	O
undersampled	O
3D	O
projection	O
reconstruction	O
trajectories	O
.	O
The	O
variable	B-KEY
density	I-KEY
k-space	I-KEY
sampling	I-KEY
intrinsic	O
to	O
this	O
sequence	O
is	O
combined	O
with	O
temporal	B-KEY
k-space	I-KEY
interpolation	I-KEY
to	O
provide	O
time	O
frames	O
as	O
short	O
as	O
4	O
s	O
.	O
This	O
time	O
resolution	O
reduces	O
the	O
need	O
for	O
exact	O
contrast	O
timing	O
while	O
also	O
providing	O
dynamic	O
information	O
.	O
Spatial	O
resolution	O
is	O
determined	O
primarily	O
by	O
the	O
projection	O
readout	O
resolution	O
and	O
is	O
thus	O
isotropic	O
across	O
the	O
FOV	O
,	O
which	O
is	O
also	O
isotropic	O
.	O
Although	O
undersampling	O
the	O
outer	O
regions	O
of	O
k-space	O
introduces	O
aliased	O
energy	O
into	O
the	O
image	O
,	O
which	O
may	O
compromise	O
resolution	O
,	O
this	O
is	O
not	O
a	O
limiting	O
factor	O
in	O
high-contrast	O
applications	O
such	O
as	O
MRA	O
.	O
Results	O
from	O
phantom	O
and	O
volunteer	O
studies	O
are	O
presented	O
demonstrating	O
isotropic	B-KEY
resolution	I-KEY
,	O
broad	B-KEY
coverage	I-KEY
with	O
an	O
isotropic	B-KEY
field	I-KEY
of	I-KEY
view	I-KEY
-LRB-	O
FOV	O
-RRB-	O
,	O
minimal	O
projection	O
reconstruction	O
artifacts	O
,	O
and	O
temporal	O
information	O
.	O
In	O
one	O
application	O
,	O
a	O
single	O
breath-hold	O
exam	O
covering	O
the	O
entire	O
pulmonary	B-KEY
vasculature	I-KEY
generates	O
high-resolution	O
,	O
isotropic	O
imaging	O
volumes	O
depicting	O
the	O
bolus	B-KEY
passage	I-KEY

Text-independent	B-KEY
speaker	I-KEY
verification	I-KEY
using	O
utterance	B-KEY
level	I-KEY
scoring	I-KEY
and	O
covariance	B-KEY
modeling	I-KEY
This	O
paper	O
describes	O
a	O
computationally	B-KEY
simple	I-KEY
method	I-KEY
to	O
perform	O
text	O
independent	O
speaker	O
verification	O
using	O
second	B-KEY
order	I-KEY
statistics	I-KEY
.	O
The	O
suggested	O
method	O
,	O
called	O
utterance	B-KEY
level	I-KEY
scoring	I-KEY
-LRB-	O
ULS	O
-RRB-	O
,	O
allows	O
one	O
to	O
obtain	O
a	O
normalized	B-KEY
score	I-KEY
using	O
a	O
single	O
pass	O
through	O
the	O
frames	O
of	O
the	O
tested	O
utterance	O
.	O
The	O
utterance	O
sample	B-KEY
covariance	I-KEY
is	O
first	O
calculated	O
and	O
then	O
compared	O
to	O
the	O
speaker	B-KEY
covariance	I-KEY
using	O
a	O
distortion	B-KEY
measure	I-KEY
.	O
Subsequently	O
,	O
a	O
distortion	B-KEY
measure	I-KEY
between	O
the	O
utterance	O
covariance	O
and	O
the	O
sample	B-KEY
covariance	I-KEY
of	O
data	O
taken	O
from	O
different	O
speakers	O
is	O
used	O
to	O
normalize	O
the	O
score	O
.	O
Experimental	O
results	O
from	O
the	O
2000	O
NIST	B-KEY
speaker	I-KEY
recognition	I-KEY
evaluation	I-KEY
are	O
presented	O
for	O
ULS	O
,	O
used	O
with	O
different	O
distortion	B-KEY
measures	I-KEY
,	O
and	O
for	O
a	O
Gaussian	B-KEY
mixture	I-KEY
model	I-KEY
-LRB-	O
GMM	B-KEY
-RRB-	O
system	O
.	O
The	O
results	O
indicate	O
that	O
ULS	O
as	O
a	O
viable	O
alternative	O
to	O
GMM	B-KEY
whenever	O
the	O
computational	B-KEY
complexity	I-KEY
and	O
verification	B-KEY
accuracy	I-KEY
needs	O
to	O
be	O
traded	O

A	O
large	B-KEY
deviations	I-KEY
analysis	I-KEY
of	O
the	O
transient	O
of	O
a	O
queue	O
with	O
many	O
Markov	B-KEY
fluid	I-KEY
inputs	I-KEY
:	O
approximations	B-KEY
and	O
fast	O
simulation	O
This	O
article	O
analyzes	O
the	O
transient	B-KEY
buffer	I-KEY
content	I-KEY
distribution	I-KEY
of	O
a	O
queue	O
fed	O
by	O
a	O
large	O
number	O
of	O
Markov	O
fluid	O
sources	O
.	O
We	O
characterize	O
the	O
probability	O
of	O
overflow	O
at	O
time	O
t	O
,	O
given	O
the	O
current	O
buffer	O
level	O
and	O
the	O
number	O
of	O
sources	O
in	O
the	O
on-state	O
.	O
After	O
scaling	O
buffer	O
and	O
bandwidth	B-KEY
resources	I-KEY
by	O
the	O
number	O
of	O
sources	O
n	O
,	O
we	O
can	O
apply	O
large	O
deviations	O
techniques	O
.	O
The	O
transient	B-KEY
overflow	I-KEY
probability	I-KEY
decays	O
exponentially	O
in	O
n	O
.	O
In	O
the	O
case	O
of	O
exponential	O
on/off	O
sources	O
,	O
we	O
derive	O
an	O
expression	O
for	O
the	O
decay	O
rate	O
of	O
the	O
rare	O
event	O
probability	O
under	O
consideration	O
.	O
For	O
general	O
Markov	O
fluid	O
sources	O
,	O
we	O
present	O
a	O
plausible	O
conjecture	O
.	O
We	O
also	O
provide	O
the	O
``	O
most	O
likely	O
path	O
''	O
from	O
the	O
initial	O
state	O
to	O
overflow	O
-LRB-	O
at	O
time	O
t	O
-RRB-	O
.	O
Knowledge	O
of	O
the	O
decay	O
rate	O
and	O
the	O
most	O
likely	O
path	O
to	O
overflow	O
leads	O
to	O
-LRB-	O
i	O
-RRB-	O
approximations	B-KEY
of	O
the	O
transient	B-KEY
overflow	I-KEY
probability	I-KEY
and	O
-LRB-	O
ii	O
-RRB-	O
efficient	O
simulation	B-KEY
methods	I-KEY
of	O
the	O
rare	O
event	O
of	O
buffer	O
overflow	O
.	O
The	O
simulation	B-KEY
methods	I-KEY
,	O
based	O
on	O
importance	B-KEY
sampling	I-KEY
,	O
give	O
a	O
huge	O
speed-up	O
compared	O
to	O
straightforward	O
simulations	O
.	O
The	O
approximations	B-KEY
are	O
of	O
low	O
computational	B-KEY
complexity	I-KEY
and	O
are	O
accurate	O
,	O
as	O
verified	O
by	O
means	O
of	O
simulation	O
experiments	O

Women	B-KEY
in	O
computing	O
:	O
what	O
brings	O
them	O
to	O
it	O
,	O
what	O
keeps	O
them	O
in	O
it	O
?	O
Career	B-KEY
stereotyping	I-KEY
and	O
misperceptions	B-KEY
about	O
the	O
nature	O
of	O
computing	O
are	O
substantive	O
reasons	O
for	O
the	O
under	O
representation	O
of	O
women	B-KEY
in	O
professional	B-KEY
computing	I-KEY
careers	I-KEY
.	O
In	O
this	O
study	O
,	O
15	O
women	B-KEY
who	O
have	O
work	O
experience	O
in	O
several	O
aspects	O
of	O
computing	O
were	O
asked	O
about	O
their	O
reasons	O
for	O
entering	O
computing	O
,	O
what	O
they	O
liked	O
about	O
working	O
in	O
computing	O
,	O
and	O
what	O
they	O
disliked	O
.	O
While	O
there	O
are	O
many	O
common	O
threads	O
,	O
there	O
are	O
also	O
individual	O
differences	O
.	O
Common	O
reasons	O
for	O
choosing	O
computing	O
as	O
a	O
career	O
included	O
:	O
exposure	O
to	O
computing	O
in	O
a	O
setting	O
which	O
enabled	O
them	O
to	O
see	O
the	O
versatility	O
of	O
computers	O
;	O
the	O
influence	O
of	O
someone	O
close	O
to	O
them	O
;	O
personal	B-KEY
abilities	I-KEY
which	O
they	O
perceived	O
to	O
be	O
appropriate	O
for	O
a	O
career	O
in	O
computing	O
;	O
and	O
characteristics	O
of	O
such	O
careers	O
which	O
appealed	O
to	O
them	O
.	O
Generally	O
,	O
women	B-KEY
working	O
in	O
the	O
field	O
enjoy	O
the	O
work	O
they	O
are	O
doing	O
.	O
Dislikes	O
arising	O
from	O
their	O
work	O
experiences	O
are	O
more	O
likely	O
to	O
be	O
associated	O
with	O
people	O
and	O
politics	B-KEY
than	O
with	O
the	O
work	O
they	O
do-and	O
they	O
would	O
like	O
to	O
have	O
more	O
female	O
colleagues	O

Fuzzy	B-KEY
system	I-KEY
modeling	I-KEY
in	O
pharmacology	B-KEY
:	O
an	O
improved	O
algorithm	O
In	O
this	O
paper	O
,	O
we	O
propose	O
an	O
improved	O
fuzzy	B-KEY
system	I-KEY
modeling	I-KEY
algorithm	O
to	O
address	O
some	O
of	O
the	O
limitations	O
of	O
the	O
existing	O
approaches	O
identified	O
during	O
our	O
modeling	O
with	O
pharmacological	B-KEY
data	O
.	O
This	O
algorithm	O
differs	O
from	O
the	O
existing	O
ones	O
in	O
its	O
approach	O
to	O
the	O
cluster	B-KEY
validity	I-KEY
problem	I-KEY
-LRB-	O
i.e.	O
,	O
number	O
of	O
clusters	O
-RRB-	O
,	O
the	O
projection	B-KEY
schema	I-KEY
-LRB-	O
i.e.	O
,	O
input	O
membership	O
assignment	O
and	O
rule	O
determination	O
-RRB-	O
,	O
and	O
significant	B-KEY
input	I-KEY
determination	I-KEY
.	O
The	O
new	O
algorithm	O
is	O
compared	O
with	O
the	O
Bazoon-Turksen	O
model	O
,	O
which	O
is	O
based	O
on	O
the	O
well-known	O
Sugeno-Yasukawa	O
approach	O
.	O
The	O
comparison	O
was	O
made	O
in	O
terms	O
of	O
predictive	B-KEY
performance	I-KEY
using	O
two	O
different	O
data	O
sets	O
.	O
The	O
first	O
comparison	O
was	O
with	O
a	O
two	O
variable	O
nonlinear	O
function	O
prediction	O
problem	O
and	O
the	O
second	O
comparison	O
was	O
with	O
a	O
clinical	O
pharmacokinetic	B-KEY
modeling	I-KEY
problem	O
.	O
It	O
is	O
shown	O
that	O
the	O
proposed	O
algorithm	O
provides	O
more	O
precise	O
predictions	O
.	O
Determining	O
the	O
degree	O
of	O
significance	O
for	O
each	O
input	O
variable	O
,	O
allows	O
the	O
user	O
to	O
distinguish	O
their	O
relative	O
importance	O

Achieving	O
competitive	B-KEY
capabilities	I-KEY
in	O
e-services	B-KEY
What	O
implications	O
does	O
the	O
Internet	B-KEY
have	O
for	O
service	B-KEY
operations	I-KEY
strategy	I-KEY
?	O
How	O
can	O
business	B-KEY
performance	I-KEY
of	O
e-service	B-KEY
companies	O
be	O
improved	O
in	O
today	O
's	O
knowledge-based	B-KEY
economy	I-KEY
?	O
These	O
research	O
questions	O
are	O
the	O
subject	O
of	O
the	O
paper	O
.	O
We	O
propose	O
a	O
model	O
that	O
links	O
the	O
e-service	B-KEY
company	O
's	O
knowledge-based	B-KEY
competencies	I-KEY
with	O
their	O
competitive	B-KEY
capabilities	I-KEY
.	O
Drawing	O
from	O
the	O
current	O
literature	O
,	O
our	O
analysis	O
suggests	O
that	O
services	O
that	O
strategically	O
build	O
a	O
portfolio	O
of	O
knowledge-based	B-KEY
competencies	I-KEY
,	O
namely	O
human	B-KEY
capital	I-KEY
,	O
structural	B-KEY
capital	I-KEY
,	O
and	O
absorptive	B-KEY
capacity	I-KEY
have	O
more	O
operations-based	B-KEY
options	I-KEY
,	O
than	O
their	O
counterparts	O
who	O
are	O
less	O
apt	O
to	O
invest	B-KEY
.	O
We	O
assume	O
that	O
the	O
combinative	B-KEY
capabilities	I-KEY
of	O
service	B-KEY
quality	I-KEY
,	O
delivery	B-KEY
,	O
flexibility	B-KEY
,	O
and	O
cost	B-KEY
are	O
determined	O
by	O
the	O
investment	B-KEY
in	O
intellectual	B-KEY
capital	I-KEY
.	O
Arguably	O
,	O
with	O
the	O
advent	O
of	O
the	O
Internet	B-KEY
,	O
different	O
operating	O
models	O
-LRB-	O
e.g.	O
,	O
bricks-and-mortar	B-KEY
,	O
clicks-and-mortar	B-KEY
,	O
or	O
pure	O
dot-com	B-KEY
-RRB-	O
have	O
different	O
strategic	B-KEY
imperatives	I-KEY
in	O
terms	O
of	O
knowledge-based	B-KEY
competencies	I-KEY
.	O
Thus	O
,	O
the	O
new	O
e-operations	O
paradigm	O
can	O
be	O
viewed	O
as	O
a	O
configuration	O
of	O
knowledge-based	B-KEY
competencies	I-KEY
and	O
capabilities	O

A	O
new	O
method	O
of	O
regression	O
on	O
latent	B-KEY
variables	I-KEY
.	O
Application	O
to	O
spectral	B-KEY
data	I-KEY
Several	O
applications	O
are	O
based	O
on	O
the	O
assessment	O
of	O
a	O
linear	B-KEY
model	I-KEY
linking	O
a	O
set	O
of	O
variables	O
Y	O
to	O
a	O
set	O
of	O
predictors	B-KEY
X	O
.	O
In	O
the	O
presence	O
of	O
strong	B-KEY
colinearity	I-KEY
among	O
predictors	B-KEY
,	O
as	O
in	O
the	O
case	O
with	O
spectral	B-KEY
data	I-KEY
,	O
several	O
alternative	O
procedures	O
to	O
ordinary	O
least	O
squares	O
-LRB-	O
OLS	O
-RRB-	O
are	O
proposed	O
,	O
We	O
discuss	O
a	O
new	O
alternative	O
approach	O
which	O
we	O
refer	O
to	O
as	O
regression	B-KEY
models	I-KEY
through	I-KEY
constrained	I-KEY
principal	I-KEY
components	I-KEY
analysis	I-KEY
-LRB-	O
RM-CPCA	O
-RRB-	O
.	O
This	O
method	O
basically	O
shares	O
certain	O
common	O
characteristics	O
with	O
PLS	O
regression	O
as	O
the	O
dependent	B-KEY
variables	I-KEY
play	O
a	O
central	O
role	O
in	O
determining	O
the	O
latent	B-KEY
variables	I-KEY
to	O
be	O
used	O
as	O
predictors	B-KEY
.	O
Unlike	O
PLS	O
,	O
however	O
,	O
the	O
approach	O
discussed	O
leads	O
to	O
straightforward	O
models	O
.	O
This	O
method	O
also	O
bears	O
some	O
similarity	O
to	O
latent	B-KEY
root	I-KEY
regression	I-KEY
analysis	I-KEY
-LRB-	O
LRR	O
-RRB-	O
that	O
was	O
discussed	O
by	O
several	O
authors	O
.	O
Moreover	O
,	O
a	O
tuning	B-KEY
parameter	I-KEY
that	O
ranges	O
between	O
0	O
and	O
1	O
is	O
introduced	O
and	O
the	O
family	O
of	O
models	O
thus	O
formed	O
includes	O
several	O
other	O
methods	O
as	O
particular	O
cases	O

A	O
shy	O
invariant	O
of	O
graphs	O
Moving	O
from	O
a	O
well	O
known	O
result	O
of	O
P.L.	O
Hammer	O
et	O
al.	O
-LRB-	O
1982	O
-RRB-	O
,	O
we	O
introduce	O
a	O
new	O
graph	O
invariant	B-KEY
,	I-KEY
say	O
lambda	O
-LRB-	O
G	O
-RRB-	O
referring	O
to	O
any	O
graph	O
G	O
.	O
It	O
is	O
a	O
non-negative	O
integer	O
which	O
is	O
non-zero	O
whenever	O
G	O
contains	O
particular	O
induced	B-KEY
odd	I-KEY
cycles	I-KEY
or	O
,	O
equivalently	O
,	O
admits	O
a	O
particular	O
minimum	B-KEY
clique-partition	I-KEY
.	O
We	O
show	O
that	O
-RRB-	O
.	O
-LRB-	O
G	O
-RRB-	O
can	O
be	O
efficiently	O
evaluated	O
and	O
that	O
its	O
determination	O
allows	O
one	O
to	O
reduce	O
the	O
hard	O
problem	O
of	O
computing	O
a	O
minimum	B-KEY
clique-cover	I-KEY
of	O
a	O
graph	O
to	O
an	O
identical	O
problem	O
of	O
smaller	O
size	O
and	O
special	O
structure	O
.	O
Furthermore	O
,	O
one	O
has	O
alpha	O
-LRB-	O
G	O
-RRB-	O
<	O
or	O
=	O
theta	O
-LRB-	O
G	O
-RRB-	O
-	O
lambda	O
-LRB-	O
G	O
-RRB-	O
,	O
where	O
alpha	O
-LRB-	O
G	O
-RRB-	O
and	O
theta	O
-LRB-	O
G	O
-RRB-	O
respectively	O
denote	O
the	O
cardinality	B-KEY
of	O
a	O
maximum	B-KEY
stable	I-KEY
set	I-KEY
of	O
G	O
and	O
of	O
a	O
minimum	B-KEY
clique-partition	I-KEY
of	O
G	O

An	O
on-line	B-KEY
distributed	I-KEY
intelligent	I-KEY
fault	I-KEY
section	I-KEY
estimation	I-KEY
system	I-KEY
for	O
large-scale	B-KEY
power	I-KEY
networks	I-KEY
In	O
this	O
paper	O
,	O
a	O
novel	O
distributed	B-KEY
intelligent	I-KEY
system	I-KEY
is	O
suggested	O
for	O
on-line	B-KEY
fault	I-KEY
section	I-KEY
estimation	I-KEY
-LRB-	O
FSE	O
-RRB-	O
of	O
large-scale	B-KEY
power	I-KEY
networks	I-KEY
.	O
As	O
the	O
first	O
step	O
,	O
a	O
multi-way	B-KEY
graph	I-KEY
partitioning	I-KEY
method	I-KEY
based	I-KEY
on	O
weighted	B-KEY
minimum	I-KEY
degree	I-KEY
reordering	I-KEY
is	O
proposed	O
for	O
effectively	O
partitioning	O
the	O
original	O
large-scale	B-KEY
power	I-KEY
network	I-KEY
into	O
the	O
desired	O
number	O
of	O
connected	B-KEY
sub-networks	I-KEY
with	O
quasi-balanced	B-KEY
FSE	I-KEY
burdens	I-KEY
and	O
minimum	B-KEY
frontier	I-KEY
elements	I-KEY
.	O
After	O
partitioning	O
,	O
a	O
distributed	B-KEY
intelligent	I-KEY
system	I-KEY
based	O
on	O
radial	B-KEY
basis	I-KEY
function	I-KEY
neural	I-KEY
network	I-KEY
-LRB-	O
RBF	O
NN	O
-RRB-	O
and	O
companion	O
fuzzy	B-KEY
system	I-KEY
is	O
suggested	O
for	O
FSE	O
.	O
The	O
relevant	O
theoretical	O
analysis	O
and	O
procedure	O
are	O
presented	O
in	O
the	O
paper	O
.	O
The	O
proposed	O
distributed	O
intelligent	O
FSE	O
method	O
has	O
been	O
implemented	O
with	O
sparse	B-KEY
storage	I-KEY
technique	I-KEY
and	O
tested	O
on	O
the	O
IEEE	O
14	O
,	O
30	O
and	O
118-bus	O
systems	O
,	O
respectively	O
.	O
Computer	B-KEY
simulation	I-KEY
results	O
show	O
that	O
the	O
proposed	O
FSE	O
method	O
works	O
successfully	O
for	O
large-scale	B-KEY
power	I-KEY
networks	I-KEY

Verification	O
of	O
non-functional	O
properties	O
of	O
a	O
composable	B-KEY
architecture	I-KEY
with	O
Petri	B-KEY
nets	I-KEY
In	O
this	O
paper	O
,	O
we	O
introduce	O
our	O
concept	O
of	O
composability	O
and	O
present	O
the	O
MSS	B-KEY
architecture	I-KEY
as	O
an	O
example	O
for	O
a	O
composable	B-KEY
architecture	I-KEY
.	O
MSS	O
claims	O
to	O
be	O
composable	O
with	O
respect	O
to	O
timing	B-KEY
properties	I-KEY
.	O
We	O
discuss	O
,	O
how	O
to	O
model	O
and	O
prove	O
properties	O
in	O
such	O
an	O
architecture	O
with	O
time-extended	O
Petrinets	O
.	O
As	O
a	O
result	O
,	O
the	O
first	O
step	O
of	O
a	O
proof	B-KEY
of	I-KEY
composability	I-KEY
is	O
presented	O
as	O
well	O
as	O
a	O
new	O
kind	O
of	O
Petri	B-KEY
net	I-KEY
,	O
which	O
is	O
more	O
suitable	O
for	O
modeling	O
architectures	O
like	O
MSS	O

From	O
powder	O
to	O
perfect	O
parts	O
GKN	B-KEY
Sinter	I-KEY
Metals	I-KEY
has	O
increased	O
productivity	O
and	O
quality	O
by	O
automating	B-KEY
the	O
powder	B-KEY
metal	I-KEY
lines	I-KEY
that	O
produce	O
its	O
transmission	O
parts	O

Is	O
diversity	O
in	O
computing	O
a	O
moral	O
matter	O
?	O
We	O
have	O
presented	O
an	O
ethical	B-KEY
argument	I-KEY
that	O
takes	O
into	O
consideration	O
the	O
subtleties	O
of	O
the	O
issue	O
surrounding	O
under-representation	O
in	O
computing	O
.	O
We	O
should	O
emphasize	O
that	O
there	O
is	O
nothing	O
subtle	O
about	O
overt	O
,	O
unfair	B-KEY
discrimination	I-KEY
.	O
Where	O
such	O
injustice	O
occurs	O
,	O
we	O
condemn	O
it	O
.	O
Our	O
concern	O
is	O
that	O
discrimination	O
need	O
not	O
be	O
explicit	O
or	O
overt	O
.	O
It	O
need	O
not	O
be	O
individual-to-individual	O
.	O
Rather	O
,	O
it	O
can	O
be	O
subtly	O
built	O
into	O
social	B-KEY
practices	I-KEY
and	O
social	B-KEY
institutions	I-KEY
.	O
Our	O
analysis	O
raises	O
ethical	O
questions	O
about	O
aspects	O
of	O
computing	O
that	O
drive	O
women	B-KEY
away	O
,	O
aspects	O
that	O
can	O
be	O
changed	O
in	O
ways	O
that	O
improve	O
the	O
profession	O
and	O
access	O
to	O
the	O
profession	O
.	O
We	O
hope	O
that	O
computing	O
will	O
move	O
towards	O
these	O
improvements	O