Dataset Viewer
Auto-converted to Parquet
A hybrid of genetic algorithm and particle swarm optimization for recurrent network design
An evolutionary recurrent network which automates the design of recurrent neural/fuzzy networks using a new evolutionary learning algorithm is proposed in this paper. This new evolutionary learning algorithm is based on a hybrid of genetic algorithm (GA) and particle swarm optimization (PSO), and is thus called HGAPSO. In HGAPSO, individuals in a new generation are created, not only by crossover and mutation operation as in GA, but also by PSO. The concept of elite strategy is adopted in HGAPSO, where the upper-half of the best-performing individuals in a population are regarded as elites. However, instead of being reproduced directly to the next generation, these elites are first enhanced. The group constituted by the elites is regarded as a swarm, and each elite corresponds to a particle within it. In this regard, the elites are enhanced by PSO, an operation which mimics the maturing phenomenon in nature. These enhanced elites constitute half of the population in the new generation, whereas the other half is generated by performing crossover and mutation operation on these enhanced elites. HGAPSO is applied to recurrent neural/fuzzy network design as follows. For recurrent neural network, a fully connected recurrent neural network is designed and applied to a temporal sequence production problem. For recurrent fuzzy network design, a Takagi-Sugeno-Kang-type recurrent fuzzy network is designed and applied to dynamic plant control. The performance of HGAPSO is compared to both GA and PSO in these recurrent networks design problems, demonstrating its superiority.
A Hybrid EP and SQP for Dynamic Economic Dispatch with Nonsmooth Fuel Cost Function
Dynamic economic dispatch (DED) is one of the main functions of power generation operation and control. It determines the optimal settings of generator units with predicted load demand over a certain period of time. The objective is to operate an electric power system most economically while the system is operating within its security limits. This paper proposes a new hybrid methodology for solving DED. The proposed method is developed in such a way that a simple evolutionary programming (EP) is applied as a based level search, which can give a good direction to the optimal global region, and a local search sequential quadratic programming (SQP) is used as a fine tuning to determine the optimal solution at the final. Ten units test system with nonsmooth fuel cost function is used to illustrate the effectiveness of the proposed method compared with those obtained from EP and SQP alone.
Genetic Fuzzy Systems - Evolutionary Tuning and Learning of Fuzzy Knowledge Bases
It's not surprisingly when entering this site to get the book. One of the popular books now is the genetic fuzzy systems evolutionary tuning and learning of fuzzy knowledge bases. You may be confused because you can't find the book in the book store around your city. Commonly, the popular book will be sold quickly. And when you have found the store to buy the book, it will be so hurt when you run out of it. This is why, searching for this popular book in this website will give you benefit. You will not run out of this book.
A modified particle swarm optimizer
In this paper, we introduce a new parameter, called inertia weight, into the original particle swarm optimizer. Simulations have been done to illustrate the signilicant and effective impact of this new parameter on the particle swarm optimizer.
Identification and control of dynamic systems using recurrent fuzzy neural networks
This paper proposes a recurrent fuzzy neural network (RFNN) structure for identifying and controlling nonlinear dynamic systems. The RFNN is inherently a recurrent multilayered connectionist network for realizing fuzzy inference using dynamic fuzzy rules. Temporal relations are embedded in the network by adding feedback connections in the second layer of the fuzzy neural network (FNN). The RFNN expands the basic ability of the FNN to cope with temporal problems. In addition, results for the FNNfuzzy inference engine, universal approximation, and convergence analysis are extended to the RFNN. For the control problem, we present the direct and indirect adaptive control approaches using the RFNN. Based on the Lyapunov stability approach, rigorous proofs are presented to guarantee the convergence of the RFNN by choosing appropriate learning rates. Finally, the RFNN is applied in several simulations (time series prediction, identification, and control of nonlinear systems). The results confirm the effectiveness of the RFNN.
Separate face and body selectivity on the fusiform gyrus.
Recent reports of a high response to bodies in the fusiform face area (FFA) challenge the idea that the FFA is exclusively selective for face stimuli. We examined this claim by conducting a functional magnetic resonance imaging experiment at both standard (3.125 x 3.125 x 4.0 mm) and high resolution (1.4 x 1.4 x 2.0 mm). In both experiments, regions of interest (ROIs) were defined using data from blocked localizer runs. Within each ROI, we measured the mean peak response to a variety of stimulus types in independent data from a subsequent event-related experiment. Our localizer scans identified a fusiform body area (FBA), a body-selective region reported recently by Peelen and Downing (2005) that is anatomically distinct from the extrastriate body area. The FBA overlapped with and was adjacent to the FFA in all but two participants. Selectivity of the FFA to faces and FBA to bodies was stronger for the high-resolution scans, as expected from the reduction in partial volume effects. When new ROIs were constructed for the high-resolution experiment by omitting the voxels showing overlapping selectivity for both bodies and faces in the localizer scans, the resulting FFA* ROI showed no response above control objects for body stimuli, and the FBA* ROI showed no response above control objects for face stimuli. These results demonstrate strong selectivities in distinct but adjacent regions in the fusiform gyrus for only faces in one region (the FFA*) and only bodies in the other (the FBA*).
Scheduling for Reduced CPU Energy
The energy usage of computer systems is becoming more important, especially for battery operated systems. Displays, disks, and cpus, in that order, use the most energy. Reducing the energy used by displays and disks has been studied elsewhere; this paper considers a new method for reducing the energy used by the cpu. We introduce a new metric for cpu energy performance, millions-of-instructions-per-joule (MIPJ). We examine a class of methods to reduce MIPJ that are characterized by dynamic control of system clock speed by the operating system scheduler. Reducing clock speed alone does not reduce MIPJ, since to do the same work the system must run longer. However, a number of methods are available for reducing energy with reduced clock-speed, such as reducing the voltage [Chandrakasan et al 1992][Horowitz 1993] or using reversible [Younis and Knight 1993] or adiabatic logic [Athas et al 1994]. What are the right scheduling algorithms for taking advantage of reduced clock-speed, especially in the presence of applications demanding ever more instructions-per-second? We consider several methods for varying the clock speed dynamically under control of the operating system, and examine the performance of these methods against workstation traces. The primary result is that by adjusting the clock speed at a fine grain, substantial CPU energy can be saved with a limited impact on performance.
A data mining approach for location prediction in mobile environments
Mobility prediction is one of the most essential issues that need to be explored for mobility management in mobile computing systems. In this paper, we propose a new algorithm for predicting the next inter-cell movement of a mobile user in a Personal Communication Systems network. In the first phase of our threephase algorithm, user mobility patterns are mined from the history of mobile user trajectories. In the second phase, mobility rules are extracted from these patterns, and in the last phase, mobility predictions are accomplished by using these rules. The performance of the proposed algorithm is evaluated through simulation as compared to two other prediction methods. The performance results obtained in terms of Precision and Recall indicate that our method can make more accurate predictions than the other methods. 2004 Elsevier B.V. All rights reserved.
$\mathsf {pSCAN}$ : Fast and Exact Structural Graph Clustering
We study the problem of structural graph clustering, a fundamental problem in managing and analyzing graph data. Given an undirected unweighted graph, structural graph clustering is to assign vertices to clusters, and to identify the sets of hub vertices and outlier vertices as well, such that vertices in the same cluster are densely connected to each other while vertices in different clusters are loosely connected. In this paper, we develop a new two-step paradigm for scalable structural graph clustering based on our three observations. Then, we present a <inline-formula> <tex-math notation="LaTeX">$\mathsf {pSCAN}$</tex-math><alternatives> <inline-graphic xlink:href="chang-ieq2-2618795.gif"/></alternatives></inline-formula> approach, within the paradigm, aiming to reduce the number of structural similarity computations, and propose optimization techniques to speed up checking whether two vertices are structure-similar. <inline-formula><tex-math notation="LaTeX">$\mathsf {pSCAN}$ </tex-math><alternatives><inline-graphic xlink:href="chang-ieq3-2618795.gif"/></alternatives></inline-formula> outputs exactly the same clusters as the existing approaches <inline-formula><tex-math notation="LaTeX">$\mathsf {SCAN}$ </tex-math><alternatives><inline-graphic xlink:href="chang-ieq4-2618795.gif"/></alternatives></inline-formula> and <inline-formula><tex-math notation="LaTeX">$\mathsf {SCAN\text{++}}$</tex-math><alternatives> <inline-graphic xlink:href="chang-ieq5-2618795.gif"/></alternatives></inline-formula>, and we prove that <inline-formula><tex-math notation="LaTeX">$\mathsf {pSCAN}$</tex-math><alternatives> <inline-graphic xlink:href="chang-ieq6-2618795.gif"/></alternatives></inline-formula> is worst-case optimal. Moreover, we propose efficient techniques for updating the clusters when the input graph dynamically changes, and we also extend our techniques to other similarity measures, e.g., Jaccard similarity. Performance studies on large real and synthetic graphs demonstrate the efficiency of our new approach and our dynamic cluster maintenance techniques. Noticeably, for the twitter graph with 1 billion edges, our approach takes 25 minutes while the state-of-the-art approach cannot finish even after 24 hours.
Synthesis, properties, and applications of iron nanoparticles.
Iron, the most ubiquitous of the transition metals and the fourth most plentiful element in the Earth's crust, is the structural backbone of our modern infrastructure. It is therefore ironic that as a nanoparticle, iron has been somewhat neglected in favor of its own oxides, as well as other metals such as cobalt, nickel, gold, and platinum. This is unfortunate, but understandable. Iron's reactivity is important in macroscopic applications (particularly rusting), but is a dominant concern at the nanoscale. Finely divided iron has long been known to be pyrophoric, which is a major reason that iron nanoparticles have not been more fully studied to date. This extreme reactivity has traditionally made iron nanoparticles difficult to study and inconvenient for practical applications. Iron however has a great deal to offer at the nanoscale, including very potent magnetic and catalytic properties. Recent work has begun to take advantage of iron's potential, and work in this field appears to be blossoming.
Automatic Machine Translation Evaluation: A Qualitative Approach
ADVERTIMENT. La consulta d鈥檃questa tesi queda condicionada a l鈥檃cceptaci贸 de les seg眉ents condicions d'煤s: La difusi贸 d鈥檃questa tesi per mitj脿 del servei TDX ( i a trav茅s del Dip貌sit Digital de la UB ( ha estat autoritzada pels titulars dels drets de propietat intel路lectual 煤nicament per a usos privats emmarcats en activitats d鈥檌nvestigaci贸 i doc猫ncia. No s鈥檃utoritza la seva reproducci贸 amb finalitats de lucre ni la seva difusi贸 i posada a disposici贸 des d鈥檜n lloc ali猫 al servei TDX ni al Dip貌sit Digital de la UB. No s鈥檃utoritza la presentaci贸 del seu contingut en una finestra o marc ali猫 a TDX o al Dip貌sit Digital de la UB (framing). Aquesta reserva de drets afecta tant al resum de presentaci贸 de la tesi com als seus continguts. En la utilitzaci贸 o cita de parts de la tesi 茅s obligat indicar el nom de la persona autora.
Gualzru's Path to the Advertisement World
This paper describes the genesis of Gualzru, a robot commissioned by a large Spanish technological company to provide advertisement services in open public spaces. Gualzru has to stand by at an interactive panel observing the people passing by and, at some point, select a promising candidate and approach her to initiate a conversation. After a small verbal interaction, the robot is supposed to convince the passerby to walk back to the panel, leaving the rest of the selling task to an interactive software embedded in it. The whole design and building process took less than three years of team composed of five groups at different geographical locations. We describe here the lessons learned during this period of time, from different points of view including the hardware, software, architectural decisions and team collaboration issues.
Treatment-Response Models for Counterfactual Reasoning with Continuous-time, Continuous-valued Interventions
Treatment effects can be estimated from observational data as the difference in potential outcomes. In this paper, we address the challenge of estimating the potential outcome when treatment-dose levels can vary continuously over time. Further, the outcome variable may not be measured at a regular frequency. Our proposed solution represents the treatment response curves using linear time-invariant dynamical systems鈥攖his provides a flexible means for modeling response over time to highly variable dose curves. Moreover, for multivariate data, the proposed method: uncovers shared structure in treatment response and the baseline across multiple markers; and, flexibly models challenging correlation structure both across and within signals over time. For this, we build upon the framework of multiple-output Gaussian Processes. On simulated and a challenging clinical dataset, we show significant gains in accuracy over stateof-the-art models.
Planar Helical Antenna of Circular Polarization
A planar helical antenna is presented for achieving wideband end-fire radiation of circular polarization while maintaining a very low profile. The helix is formed using printed strips with straight-edge connections implemented by plated viaholes. The currents flowing on the strips and along via-holes of the helix contribute to the horizontal and vertical polarizations, respectively. Besides, the current on the ground plane is utilized to weaken the strong amplitude of the horizontal electric field generated by the one on the strips. Thus, a good circular polarization can be achieved. Furthermore, a tapered helix and conducting side-walls are employed to broaden the axial ratio (AR) bandwidth as well as to improve the end-fire radiation pattern. The designed antenna operates at the center frequency of 10 GHz. Simulated results show that the planar helical antenna achieves wide-impedance bandwidth (|S11| <; -10 dB) from 7.4 to 12.8 GHz (54%) and 3-dB AR bandwidth from 8.2 to 11.6 GHz (34%), while retaining a thickness of only 0.11位0 at the center frequency. A prototype of the proposed antenna is fabricated and tested. Measured results are in good agreement with simulated ones.
Body Composition Changes After Very-Low-Calorie Ketogenic Diet in Obesity Evaluated by 3 Standardized Methods.
Context Common concerns when using low-calorie diets as a treatment for obesity are the reduction in fat-free mass, mostly muscular mass, that occurs together with the fat mass (FM) loss, and determining the best methodologies to evaluate body composition changes. Objective This study aimed to evaluate the very-low-calorie ketogenic (VLCK) diet-induced changes in body composition of obese patients and to compare 3 different methodologies used to evaluate those changes. Design Twenty obese patients followed a VLCK diet for 4 months. Body composition assessment was performed by dual-energy X-ray absorptiometry (DXA), multifrequency bioelectrical impedance (MF-BIA), and air displacement plethysmography (ADP) techniques. Muscular strength was also assessed. Measurements were performed at 4 points matched with the ketotic phases (basal, maximum ketosis, ketosis declining, and out of ketosis). Results After 4 months the VLCK diet induced a -20.2 卤 4.5 kg weight loss, at expenses of reductions in fat mass (FM) of -16.5 卤 5.1 kg (DXA), -18.2 卤 5.8 kg (MF-BIA), and -17.7 卤 9.9 kg (ADP). A substantial decrease was also observed in the visceral FM. The mild but marked reduction in fat-free mass occurred at maximum ketosis, primarily as a result of changes in total body water, and was recovered thereafter. No changes in muscle strength were observed. A strong correlation was evidenced between the 3 methods of assessing body composition. Conclusion The VLCK diet-induced weight loss was mainly at the expense of FM and visceral mass; muscle mass and strength were preserved. Of the 3 body composition techniques used, the MF-BIA method seems more convenient in the clinical setting.
Measuring pictorial balance perception at first glance using Japanese calligraphy
According to art theory, pictorial balance acts to unify picture elements into a cohesive composition. For asymmetrical compositions, balancing elements is thought to be similar to balancing mechanical weights in a framework of symmetry axes. Assessment of preference for balance (APB), based on the symmetry-axes framework suggested in Arnheim R, 1974 Art and Visual Perception: A Psychology of the Creative Eye (Berkeley, CA: University of California Press), successfully matched subject balance ratings of images of geometrical shapes over unlimited viewing time. We now examine pictorial balance perception of Japanese calligraphy during first fixation, isolated from later cognitive processes, comparing APB measures with results from balance-rating and comparison tasks. Results show high between-task correlation, but low correlation with APB. We repeated the rating task, expanding the image set to include five rotations of each image, comparing balance perception of artist and novice participant groups. Rotation has no effect on APB balance computation but dramatically affects balance rating, especially for art experts. We analyze the variety of rotation effects and suggest that, rather than depending on element size and position relative to symmetry axes, first fixation balance processing derives from global processes such as grouping of lines and shapes, object recognition, preference for horizontal and vertical elements, closure, and completion, enhanced by vertical symmetry.
Smoking and cervical cancer: pooled analysis of the IARC multi-centric case鈥揷ontrol study
Background: Smoking has long been suspected to be a risk factor for cervical cancer. However, not all previous studies have properly controlled for the effect of human papillomavirus (HPV) infection, which has now been established as a virtually necessary cause of cervical cancer. To evaluate the role of smoking as a cofactor of progression from HPV infection to cancer, we performed a pooled analysis of 10 previously published case鈥揷ontrol studies. This analysis is part of a series of analyses of cofactors of HPV in the aetiology of cervical cancer. Methods: Data were pooled from eight case鈥揷ontrol studies of invasive cervical carcinoma (ICC) and two of carcinoma in situ (CIS) from four continents. All studies used a similar protocol and questionnaires and included a PCR-based evaluation of HPV DNA in cytological smears or biopsy specimens. Only subjects positive for HPV DNA were included in the analysis. A total of 1463 squamous cell ICC cases were analyzed, along with 211 CIS cases, 124 adeno- or adeno-squamous ICC cases and 254 control women. Pooled odds ratios (OR) and 95% confidence intervals (CI) were estimated using logistic regression models controlling for sexual and non-sexual confounding factors. Results: There was an excess risk for ever smoking among HPV positive women (OR 2.17 95%CI 1.46鈥3.22). When results were analyzed by histological type, an excess risk was observed among cases of squamous cell carcinoma for current smokers (OR 2.30, 95%CI 1.31鈥4.04) and ex-smokers (OR 1.80, 95%CI 0.95鈥3.44). No clear pattern of association with risk was detected for adenocarcinomas, although the number of cases with this histologic type was limited. Conclusions: Smoking increases the risk of cervical cancer among HPV positive women. The results of our study are consistent with the few previously conducted studies of smoking and cervical cancer that have adequately controlled for HPV infection. Recent increasing trends of smoking among young women could have a serious impact on cervical cancer incidence in the coming years.
Breathing Detection: Towards a Miniaturized, Wearable, Battery-Operated Monitoring System
This paper analyzes the main challenges associated with noninvasive, continuous, wearable, and long-term breathing monitoring. The characteristics of an acoustic breathing signal from a miniature sensor are studied in the presence of sources of noise and interference artifacts that affect the signal. Based on these results, an algorithm has been devised to detect breathing. It is possible to implement the algorithm on a single integrated circuit, making it suitable for a miniature sensor device. The algorithm is tested in the presence of noise sources on five subjects and shows an average success rate of 91.3% (combined true positives and true negatives).
Speech-driven 3 D Facial Animation with Implicit Emotional Awareness : A Deep Learning Approach
We introduce a long short-term memory recurrent neural network (LSTM-RNN) approach for real-time facial animation, which automatically estimates head rotation and facial action unit activations of a speaker from just her speech. Specifically, the time-varying contextual non-linear mapping between audio stream and visual facial movements is realized by training a LSTM neural network on a large audio-visual data corpus. In this work, we extract a set of acoustic features from input audio, including Mel-scaled spectrogram, Mel frequency cepstral coefficients and chromagram that can effectively represent both contextual progression and emotional intensity of the speech. Output facial movements are characterized by 3D rotation and blending expression weights of a blendshape model, which can be used directly for animation. Thus, even though our model does not explicitly predict the affective states of the target speaker, her emotional manifestation is recreated via expression weights of the face model. Experiments on an evaluation dataset of different speakers across a wide range of affective states demonstrate promising results of our approach in real-time speech-driven facial animation.
Compressed Nonparametric Language Modelling
Hierarchical Pitman-Yor Process priors are compelling methods for learning language models, outperforming point-estimate based methods. However, these models remain unpopular due to computational and statistical inference issues, such as memory and time usage, as well as poor mixing of sampler. In this work we propose a novel framework which represents the HPYP model compactly using compressed suffix trees. Then, we develop an efficient approximate inference scheme in this framework that has a much lower memory footprint compared to full HPYP and is fast in the inference time. The experimental results illustrate that our model can be built on significantly larger datasets compared to previous HPYP models, while being several orders of magnitudes smaller, fast for training and inference, and outperforming the perplexity of the state-of-the-art Modified Kneser-Ney countbased LM smoothing by up to 15%.
Dissecting and Reassembling Color Correction Algorithms for Image Stitching
This paper introduces a new compositional framework for classifying color correction methods according to their two main computational units. The framework was used to dissect fifteen among the best color correction algorithms and the computational units so derived, with the addition of four new units specifically designed for this work, were then reassembled in a combinatorial way to originate about one hundred distinct color correction methods, most of which never considered before. The above color correction methods were tested on three different existing datasets, including both real and artificial color transformations, plus a novel dataset of real image pairs categorized according to the kind of color alterations induced by specific acquisition setups. Differently from previous evaluations, special emphasis was given to effectiveness in real world applications, such as image mosaicing and stitching, where robustness with respect to strong image misalignments and light scattering effects is required. Experimental evidence is provided for the first time in terms of the most recent perceptual image quality metrics, which are known to be the closest to human judgment. Comparative results show that combinations of the new computational units are the most effective for real stitching scenarios, regardless of the specific source of color alteration. On the other hand, in the case of accurate image alignment and artificial color alterations, the best performing methods either use one of the new computational units, or are made up of fresh combinations of existing units.
An analysis on the significance of ticket analytics and defect analysis from software quality perspective
Software even though intangible should undergo evolution to fit into the ever changing real world scenarios. Each issue faced by the development and service team directly reflects in the quality of the software product. According to the related work, very few research is going on in the field of ticket and its related incident; a part of corrective maintenance. In depth research on incident tickets should be viewed as critical since, it provides information related to the kind of maintenance activities that is performed in any timestamp. Therefore classifying and analyzing tickets becomes a critical task in managing the operations of the service since each incident will be having a service level agreement associated with it. Further, incident analysis is essential to identify the patterns associated. Due to the existence of huge population of software in each organization and millions of incidents get reported per software product every year, it is practically impossible to manually analyze all the tickets. This paper focuses on projecting the importance of ticket to maintain the quality of software products and also distinguish it from the defect associated with a software system. This paper projects the importance of identifying defects in software as well as handling the incident related tickets and resolving it when viewed from the perspective of quality. It also gives an overview of the scope defect analysis and ticket analytics provide to the researchers.
Clickbait Convolutional Neural Network
With the development of online advertisements, clickbait spread wider and wider. Clickbait dissatisfies users because the article content does not match their expectation. Thus, clickbait detection has attracted more and more attention recently. Traditional clickbait-detection methods rely on heavy feature engineering and fail to distinguish clickbait from normal headlines precisely because of the limited information in headlines. A convolutional neural network is useful for clickbait detection, since it utilizes pretrained Word2Vec to understand the headlines semantically, and employs different kernels to find various characteristics of the headlines. However, different types of articles tend to use different ways to draw users鈥 attention, and a pretrained Word2Vec model cannot distinguish these different ways. To address this issue, we propose a clickbait convolutional neural network (CBCNN) to consider not only the overall characteristics but also specific characteristics from different article types. Our experimental results show that our method outperforms traditional clickbait-detection algorithms and the TextCNN model in terms of precision, recall and accuracy.
A Database of Narrative Schemas
This paper describes a new language resource of events and semantic roles that characterize real-world situations. Narrative schemas contain sets of related events (edit and publish), a temporal ordering of the events (edit before publish), and the semantic roles of the participants (authors publish books). This type of world knowledge was central to early research in natural language understanding. Scripts were one of the main formalisms, representing common sequences of events that occur in the world. Unfortunately, most of this knowledge was hand-coded and time consuming to create. Current machine learning techniques, as well as a new approach to learning through coreference chains, has allowed us to automatically extract rich event structure from open domain text in the form of narrative schemas. The narrative schema resource described in this paper contains approximately 5000 unique events combined into schemas of varying sizes. We describe the resource, how it is learned, and a new evaluation of the coverage of these schemas over unseen documents.
User-Oriented Context Suggestion
Recommender systems have been used in many domains to assist users' decision making by providing item recommendations and thereby reducing information overload. Context-aware recommender systems go further, incorporating the variability of users' preferences across contexts, and suggesting items that are appropriate in different contexts. In this paper, we present a novel recommendation task, "Context Suggestion", whereby the system recommends contexts in which items may be selected. We introduce the motivations behind the notion of context suggestion and discuss several potential solutions. In particular, we focus specifically on user-oriented context suggestion which involves recommending appropriate contexts based on a user's profile. We propose extensions of well-known context-aware recommendation algorithms such as tensor factorization and deviation-based contextual modeling and adapt them as methods to recommend contexts instead of items. In our empirical evaluation, we compare the proposed solutions to several baseline algorithms using four real-world data sets.
Brand followers' retweeting behavior on Twitter: How brand relationships influence brand electronic word-of-mouth
Twitter, the popular microblogging site, has received increasing attention as a unique communication tool that facilitates electronic word-of-mouth (eWOM). To gain greater insight into this potential, this study investigates how consumers鈥 relationships with brands influence their engagement in retweeting brand messages on Twitter. Data from a survey of 315 Korean consumers who currently follow brands on Twitter show that those who retweet brand messages outscore those who do not on brand identification, brand trust, community commitment, community membership intention, Twitter usage frequency, and total number of postings. 2014 Elsevier Ltd. All rights reserved.
Memory, navigation and theta rhythm in the hippocampal-entorhinal system
Theories on the functions of the hippocampal system are based largely on two fundamental discoveries: the amnestic consequences of removing the hippocampus and associated structures in the famous patient H.M. and the observation that spiking activity of hippocampal neurons is associated with the spatial position of the rat. In the footsteps of these discoveries, many attempts were made to reconcile these seemingly disparate functions. Here we propose that mechanisms of memory and planning have evolved from mechanisms of navigation in the physical world and hypothesize that the neuronal algorithms underlying navigation in real and mental space are fundamentally the same. We review experimental data in support of this hypothesis and discuss how specific firing patterns and oscillatory dynamics in the entorhinal cortex and hippocampus can support both navigation and memory.
Predictive Control of Power Converters: Designs With Guaranteed Performance
In this work, a cost function design based on Lyapunov stability concepts for finite control set model predictive control is proposed. This predictive controller design allows one to characterize the performance of the controlled converter, while providing sufficient conditions for local stability for a class of power converters. Simulation and experimental results on a buck dc-dc converter and a two-level dc-ac inverter are conducted to validate the effectiveness of our proposal.
Design and Implementation of a Fast Dynamic Packet Filter
This paper presents Swift, a packet filter for high-performance packet capture on commercial off-the-shelf hardware. The key features of the Swift include: 1) extremely lowfilter update latency for dynamic packet filtering, and 2) gigabits-per-second high-speed packet processing. Based on complex instruction set computer (CISC) instruction set architecture (ISA), Swift achieves the former with an instruction set design that avoids the need for compilation and security checking, and the latter by mainly utilizing single instruction, multiple data (SIMD). We implement Swift in the Linux 2.6 kernel for both i386 and 脳86-64 architectures and extensively evaluate its dynamic and static filtering performance on multiple machines with different hardware setups. We compare Swift to BPF (the BSD packet filter)--the de facto standard for packet filtering in modern operating systems--and hand-coded optimized C filters that are used for demonstrating possible performance gains. For dynamic filtering tasks, Swift is at least three orders of magnitude faster than BPF in terms of filter update latency. For static filtering tasks, Swift outperforms BPF up to three times in terms of packet processing speed and achieves much closer performance to the optimized C filters. We also show that Swift can harness the processing power of hardware SIMD instructions by virtue of its SIMD-capable instruction set.
Learning Scalable Deep Kernels with Recurrent Structure
Many applications in speech, robotics, finance, and biology deal with sequential data, where ordering matters and recurrent structures are common. However, this structure cannot be easily captured by standard kernel functions. To model such structure, we propose expressive closed-form kernel functions for Gaussian processes. The resulting model, GP-LSTM, fully encapsulates the inductive biases of long short-term memory (LSTM) recurrent networks, while retaining the non-parametric probabilistic advantages of Gaussian processes. We learn the properties of the proposed kernels by optimizing the Gaussian process marginal likelihood using a new provably convergent semi-stochastic gradient procedure, and exploit the structure of these kernels for scalable training and prediction. This approach provides a practical representation for Bayesian LSTMs. We demonstrate state-of-the-art performance on several benchmarks, and thoroughly investigate a consequential autonomous driving application, where the predictive uncertainties provided by GP-LSTM are uniquely valuable.
SemEval-2015 Task 11: Sentiment Analysis of Figurative Language in Twitter
This report summarizes the objectives and evaluation of the SemEval 2015 task on the sentiment analysis of figurative language on Twitter (Task 11). This is the first sentiment analysis task wholly dedicated to analyzing figurative language on Twitter. Specifically, three broad classes of figurative language are considered: irony, sarcasm and metaphor. Gold standard sets of 8000 training tweets and 4000 test tweets were annotated using workers on the crowdsourcing platform CrowdFlower. Participating systems were required to provide a fine-grained sentiment score on an 11-point scale (-5 to +5, including 0 for neutral intent) for each tweet, and systems were evaluated against the gold standard using both a Cosinesimilarity and a Mean-Squared-Error measure.
Text mining of news-headlines for FOREX market prediction: A Multi-layer Dimension Reduction Algorithm with semantics and sentiment
In this paper a novel approach is proposed to predict intraday directional-movements of a currency-pair in the foreign exchange market based on the text of breaking financial news-headlines. The motivation behind this work is twofold: First, although market-prediction through text-mining is shown to be a promising area of work in the literature, the text-mining approaches utilized in it at this stage are not much beyond basic ones as it is still an emerging field. This work is an effort to put more emphasis on the text-mining methods and tackle some specific aspects thereof that are weak in previous works, namely: the problem of high dimensionality as well as the problem of ignoring sentiment and semantics in dealing with textual language. This research assumes that addressing these aspects of text-mining have an impact on the quality of the achieved results. The proposed system proves this assumption to be right. The second part of the motivation is to research a specific market, namely, the foreign exchange market, which seems not to have been researched in the previous works based on predictive text-mining. Therefore, results of this work also successfully demonstrate a predictive relationship between this specific market-type and the textual data of news. Besides the above two main components of the motivation, there are other specific aspects that make the setup of the proposed system and the conducted experiment unique, for example, the use of news article-headlines only and not news article-bodies, which enables usage of short pieces of text rather than long ones; or the use of general financial breaking news without any further filtration. In order to accomplish the above, this work produces a multi-layer algorithm that tackles each of the mentioned aspects of the text-mining problem at a designated layer. The first layer is termed the Semantic Abstraction Layer and addresses the problem of co-reference in text mining that is contributing to sparsity. Co-reference occurs when two or more words in a text corpus refer to the same concept. This work produces a custom approach by the name of Heuristic-Hypernyms Feature-Selection which creates a way to recognize words with the same parent-word to be regarded as one entity. As a result, prediction accuracy increases significantly at this layer which is attributed to appropriate noise-reduction from the feature-space. The second layer is termed Sentiment Integration Layer, which integrates sentiment analysis capability into the algorithm by proposing a sentiment weight by the name of SumScore that reflects investors鈥 sentiment. Additionally, this layer reduces the dimensions by eliminating those that are of zero value in terms of sentiment and thereby improves prediction accuracy. The third layer encompasses a dynamic model creation algorithm, termed Synchronous Targeted Feature Reduction (STFR). It is suitable for the challenge at hand whereby the mining of a stream of text is concerned. It updates the models with the most recent information available and, more importantly, it ensures that the dimensions are reduced to the absolute minimum. The algorithm and each of its layers are extensively evaluated using real market data and news content across multiple years and have proven to be solid and superior to any other comparable solution. The proposed techniques implemented in the system, result in significantly high directional-accuracies of up to 83.33%. On top of a well-rounded multifaceted algorithm, this work contributes a much needed research framework for this context with a test-bed of data that must make future research endeavors more convenient. The produced algorithm is scalable and its modular design allows improvement in each of its layers in future research. This paper provides ample details to reproduce the entire system and the conducted experiments. 2014 Elsevier Ltd. All rights reserved. A. Khadjeh Nassirtoussi et al. / Expert Systems with Applications 42 (2015) 306鈥324 307
BabelNet: The automatic construction, evaluation and application of a wide-coverage multilingual semantic network
a r t i c l e i n f o a b s t r a c t We present an automatic approach to the construction of BabelNet, a very large, wide-coverage multilingual semantic network. Key to our approach is the integration of lexicographic and encyclopedic knowledge from WordNet and Wikipedia. In addition, Machine Translation is applied to enrich the resource with lexical information for all languages. We first conduct in vitro experiments on new and existing gold-standard datasets to show the high quality and coverage of BabelNet. We then show that our lexical resource can be used successfully to perform both monolingual and cross-lingual Word Sense Disambiguation: thanks to its wide lexical coverage and novel semantic relations, we are able to achieve state-of the-art results on three different SemEval evaluation tasks.
Thumbs Up or Thumbs Down? Semantic Orientation Applied to Unsupervised Classification of Reviews
This paper presents a simple unsupervised learning algorithm for classifying reviews as recommended (thumbs up) or not recommended (thumbs down). The classification of a review is predicted by the average semantic orientation of the phrases in the review that contain adjectives or adverbs. A phrase has a positive semantic orientation when it has good associations (e.g., 鈥渟ubtle nuances鈥) and a negative semantic orientation when it has bad associations (e.g., 鈥渧ery cavalier鈥). In this paper, the semantic orientation of a phrase is calculated as the mutual information between the given phrase and the word 鈥渆xcellent鈥 minus the mutual information between the given phrase and the word 鈥減oor鈥. A review is classified as recommended if the average semantic orientation of its phrases is positive. The algorithm achieves an average accuracy of 74% when evaluated on 410 reviews from Epinions, sampled from four different domains (reviews of automobiles, banks, movies, and travel destinations). The accuracy ranges from 84% for automobile reviews to 66% for movie reviews.
Opinion observer: analyzing and comparing opinions on the Web
The Web has become an excellent source for gathering consumer opinions. There are now numerous Web sites containing such opinions, e.g., customer reviews of products, forums, discussion groups, and blogs. This paper focuses on online customer reviews of products. It makes two contributions. First, it proposes a novel framework for analyzing and comparing consumer opinions of competing products. A prototype system called Opinion Observer is also implemented. The system is such that with a single glance of its visualization, the user is able to clearly see the strengths and weaknesses of each product in the minds of consumers in terms of various product features. This comparison is useful to both potential customers and product manufacturers. For a potential customer, he/she can see a visual side-by-side and feature-by-feature comparison of consumer opinions on these products, which helps him/her to decide which product to buy. For a product manufacturer, the comparison enables it to easily gather marketing intelligence and product benchmarking information. Second, a new technique based on language pattern mining is proposed to extract product features from Pros and Cons in a particular type of reviews. Such features form the basis for the above comparison. Experimental results show that the technique is highly effective and outperform existing methods significantly.
A survey of software testing practices in alberta
Software organizations have typically de-emphasized the importance of software testing. In this paper, the results of a regional survey of software testing and software quality assurance techniques are described. Researchers conducted the study during the summer and fall of 2002 by surveying software organizations in the Province of Alberta. Results indicate that Alberta-based organizations tend to test less than their counterparts in the United States. The results also indicate that Alberta software organizations tend to train fewer personnel on testing-related topics. This practice has the potential for a two-fold impact: first, the ability to detect trends that lead to reduced quality and to identify the root causes of reductions in product quality may suffer from the lack of testing. This consequence is serious enough to warrant consideration, since overall quality may suffer from the reduced ability to detect and eliminate process or product defects. Second, the organization may have a more difficult time adopting methodologies such as extreme programming. This is significant because other industry studies have concluded that many software organizations have tried or will in the next few years try some form of agile method. Newer approaches to software development like extreme programming increase the extent to which teams rely on testing skills. Organizations should consider their testing skill level as a key indication of their readiness for adopting software development techniques such as test-driven development, extreme programming, agile modelling, or other agile methods.
Analysis and Design of Average Current Mode Control Using a Describing-Function-Based Equivalent Circuit Model
This paper proposes a small-signal model for average current mode control based on an equivalent circuit. The model uses a three-terminal equivalent circuit model based on a linearized describing function method to include the feedback effect of the sideband frequency components of the inductor current. The model extends the results obtained in peak current mode control to average current mode control. The proposed small-signal model is accurate up to half switching frequency, predicting the subharmonic instability. The proposed model is verified using SIMPLIS simulation and hardware experiments, which show good agreement with the measurement results. Based on the proposed model, new feedback design guidelines are presented. The proposed design guidelines are compared with several conventional, widely used design criteria. By designing the external ramp following the proposed design guidelines, the quality factor of the double poles at half of the switching frequency in the control-to-output transfer function can be precisely controlled. This helps the feedback loop design to achieve wide control bandwidth and proper damping.
Using deep learning for short text understanding
Classifying short texts to one category or clustering semantically related texts is challenging, and the importance of both is growing due to the rise of microblogging platforms, digital news feeds, and the like. We can accomplish this classifying and clustering with the help of a deep neural network which produces compact binary representations of a short text, and can assign the same category to texts that have similar binary representations. But problems arise when there is little contextual information on the short texts, which makes it difficult for the deep neural network to produce similar binary codes for semantically related texts. We propose to address this issue using semantic enrichment. This is accomplished by taking the nouns, and verbs used in the short texts and generating the concepts and co-occurring words with the help of those terms. The nouns are used to generate concepts within the given short text, whereas the verbs are used to prune the ambiguous context (if any) present in the text. The enriched text then goes through a deep neural network to produce a prediction label for that short text representing it鈥檚 category.
MVOR: A Multi-view RGB-D Operating Room Dataset for 2D and 3D Human Pose Estimation
Person detection and pose estimation is a key requirement to develop intelligent context-aware assistance systems. To foster the development of human pose estimation methods and their applications in the Operating Room (OR), we release the Multi-View Operating Room (MVOR) dataset, the first public dataset recorded during real clinical interventions. It consists of 732 synchronized multi-view frames recorded by three RGB-D cameras in a hybrid OR. It also includes the visual challenges present in such environments, such as occlusions and clutter. We provide camera calibration parameters, color and depth frames, human bounding boxes, and 2D/3D pose annotations. In this paper, we present the dataset, its annotations, as well as baseline results from several recent person detection and 2D/3D pose estimation methods. Since we need to blur some parts of the images to hide identity and nudity in the released dataset, we also present a comparative study of how the baselines have been impacted by the blurring. Results show a large margin for improvement and suggest that the MVOR dataset can be useful to compare the performance of the different methods.
Beyond engagement analytics: which online mixed-data factors predict student learning outcomes?
This mixed-method study focuses on online learning analytics, a research area of importance. Several important student attributes and their online activities are examined to identify what seems to work best to predict higher grades. The purpose is to explore the relationships between student grade and key learning engagement factors using a large sample from an online undergraduate business course at an accredited American university (n聽=聽228). Recent studies have discounted the ability to predict student learning outcomes from big data analytics but a few significant indicators have been found by some researchers. Current studies tend to use quantitative factors in learning analytics to forecast outcomes. This study extends that work by testing the common quantitative predictors of learning outcome, but qualitative data is also examined to triangulate the evidence. Pre and post testing of information technology understanding is done at the beginning of the course. First quantitative data is collected, and depending on the hypothesis test results, qualitative data is collected and analyzed with text analytics to uncover patterns. Moodle engagement analytics indicators are tested as predictors in the model. Data is also taken from the Moodle system logs. Qualitative data is collected from student reflection essays. The result was a significant General Linear Model with four online interaction predictors that captured 77.5聽% of grade variance in an undergraduate business course.
Which Hotel attributes Matter ? A review of previous and a framework for future research
A lot of effort has been made in the last decades to reveal, which hotel attributes guest care about. Due to the high costs that are typically involved with investments in the hotel industry, it makes a lot of sense to study, which product components the travellers appreciate. This study reveals that hotel attribute research turns out to be a wide and extremely heterogeneous field of research. The authors review empirical studies investigating the importance of hotel attributes, provide attribute rankings and suggest a framework for past and future research projects in the field, based on the dimensions 鈥渇ocus of research鈥, 鈥漴isk versus utility鈥 and 鈥渢rade-off versus no trade-off questioning situation鈥.
On Power Quality of Variable-Speed Constant-Frequency Aircraft Electric Power Systems
In this paper, a comprehensive model of the variable-speed constant-frequency aircraft electric power system is developed to study the performance characteristics of the system and, in particular, the system power quality over a frequency range of operation of 400 Hz to 800 Hz. A fully controlled active power filter is designed to regulate the load terminal voltage, eliminate harmonics, correct supply power factor, and minimize the effect of unbalanced loads. The control algorithm for the active power filter (APF) is based on the perfect harmonic cancellation method which provides a three-phase reference supply current in phase with its positive-sequence fundamental voltage. The proposed APF is integrated into the model of a 90-kVA advanced aircraft electric power system under VSCF operation. The performance characteristics of the system are studied with the frequency of the generator's output voltage varied from 400 Hz to 800 Hz under different loading conditions. Several case studies are presented including dc loads as well as passive and dynamic ac loads. The power quality characteristics of the studied aircraft electric power system with the proposed active filter are shown to be in compliance with the most recent military aircraft electrical standards MIL-STD-704F as well as with the IEEE Std. 519.
Shape-aware Instance Segmentation
We address the problem of instance-level semantic segmentation, which aims at jointly detecting, segmenting and classifying every individual object in an image. In this context, existing methods typically propose candidate objects, usually as bounding boxes, and directly predict a binary mask within each such proposal. As a consequence, they cannot recover from errors in the object candidate generation process, such as too small or shifted boxes. In this paper, we introduce a novel object segment representation based on the distance transform of the object masks. We then design an object mask network (OMN) with a new residual-deconvolution architecture that infers such a representation and decodes it into the final binary object mask. This allows us to predict masks that go beyond the scope of the bounding boxes and are thus robust to inaccurate object candidates. We integrate our OMN into a Multitask Network Cascade framework, and learn the resulting shape-aware instance segmentation (SAIS) network in an end-to-end manner. Our experiments on the PASCAL VOC 2012 and the CityScapes datasets demonstrate the benefits of our approach, which outperforms the state-of-the-art in both object proposal generation and instance segmentation.
RT-Mover: a rough terrain mobile robot with a simple leg-wheel hybrid mechanism
There is a strong demand in many fields for practical robots, such as a porter robot and a personal mobility robot, that can move over rough terrain while carrying a load horizontally. We have developed a robot, called RT-Mover, which shows adequate mobility performance on targeted types of rough terrain. It has four drivable wheels and two leg-like axles but only five active shafts. A strength of this robot is that it realizes both a leg mode and a wheel mode in a simple mechanism. In this paper, the mechanical design concept is discussed. With an emphasis on minimizing the number of drive shafts, a mechanism is designed for a four-wheeled mobile body that is widely used in practical locomotive machinery. Also, strategies for moving on rough terrain are proposed. The kinematics, stability, and control of RT-Mover are also described in detail. Some typical cases of rough terrain for wheel mode and leg mode are selected, and the robot鈥檚 ability of locomotion is assessed through simulations and experiments. In each case, the robot is able to move over rough terrain while maintaining the horizontal orientation of its platform.
Privacy attacks in social media using photo tagging networks: a case study with Facebook
Social-networking users unknowingly reveal certain kinds of personal information that malicious attackers could profit from to perpetrate significant privacy breaches. This paper quantitatively demonstrates how the simple act of tagging pictures on the social-networking site of Facebook could reveal private user attributes that are extremely sensitive. Our results suggest that photo tags can be used to help predicting some, but not all, of the analyzed attributes. We believe our analysis make users aware of significant breaches of their privacy and could inform the design of new privacy-preserving ways of tagging pictures on social-networking sites.
A framework based on RSA and AES encryption algorithms for cloud computing services
Cloud computing is an emerging computing model in which resources of the computing communications are provided as services over the Internet. Privacy and security of cloud storage services are very important and become a challenge in cloud computing due to loss of control over data and its dependence on the cloud computing provider. While there is a huge amount of transferring data in cloud system, the risk of accessing data by attackers raises. Considering the problem of building a secure cloud storage service, current scheme is proposed which is based on combination of RSA and AES encryption methods to share the data among users in a secure cloud system. The proposed method allows providing difficulty for attackers as well as reducing the time of information transmission between user and cloud data storage.
BCSAT : A Benchmark Corpus for Sentiment Analysis in Telugu Using Word-level Annotations
The presented work aims at generating a systematically annotated corpus that can support the enhancement of sentiment analysis tasks in Telugu using wordlevel sentiment annotations. From OntoSenseNet, we extracted 11,000 adjectives, 253 adverbs, 8483 verbs and sentiment annotation is being done by language experts. We discuss the methodology followed for the polarity annotations and validate the developed resource. This work aims at developing a benchmark corpus, as an extension to SentiWordNet, and baseline accuracy for a model where lexeme annotations are applied for sentiment predictions. The fundamental aim of this paper is to validate and study the possibility of utilizing machine learning algorithms, word-level sentiment annotations in the task of automated sentiment identification. Furthermore, accuracy is improved by annotating the bi-grams extracted from the target corpus.
Effects of quiz-style information presentation on user understanding
This paper proposes quiz-style information presentation for interactive systems as a means to improve user understanding in educational tasks. Since the nature of quizzes can highly motivate users to stay voluntarily engaged in the interaction and keep their attention on receiving information, it is expected that information presented as quizzes can be better understood by users. To verify the effectiveness of the approach, we implemented read-out and quiz systems and performed comparison experiments using human subjects. In the task of memorizing biographical facts, the results showed that user understanding for the quiz system was significantly better than that for the read-out system, and that the subjects were more willing to use the quiz system despite the long duration of the quizzes. This indicates that quiz-style information presentation promotes engagement in the interaction with the system, leading to the improved user understanding.
Research on continuous auditing: A bibliometric analysis
This paper presents the results of a bibliometric study about the evolution of research on Continuous Auditing. This study has as main motivation to find reasons for the very slow evolvement of research on this topic. In addition, Continuous Auditing is one of the features of the emerging concept of Continuous Assurance. Thus, considering that Continuous Assurance represents numerous advantages for the organizational performance, this study also intends to understand if there is a relation between the evolution of research on Continuous Auditing and the still very low maturity levels of continuous assurance solutions. This study shows that the number of publications is considerably reduced and that the slow growth of research on Continuous Auditing may be contributing to the lack of maturity of Continuous Assurance.
Dispositional Factors in Internet Use: Personality Versus Cognitive Style
This study directly tests the effect of personality and cognitive style on three measures of Internet use. The results support the use of personality鈥攂ut not cognitive style鈥攁s an antecedent variable. After controlling for computer anxiety, selfefficacy, and gender, including the 鈥淏ig Five鈥 personality factors in the analysis significantly adds to the predictive capabilities of the dependent variables. Including cognitive style does not. The results are discussed in terms of the role of personality and cognitive style in models of technology adoption and use.
Voice quality evaluation of recent open source codecs
鈥 Averaged frequency responses at different 16, and 24 kHz. External sampling rate does not tell the internal sampling rate. 鈥 Supported signal bandwidth depends on bitrate, but no documentation exists bandwidths were found out expirementally 鈥 We tested 32kHz sampling with 16ms frame length. There is also 8 ms lookahead. 鈥 The results show that bitrates below 32 kbit/s are not useable for voice applications.The voice quality is much worse than with SILK or bitrates shown in steady state
Chain Replication for Supporting High Throughput and Availability
Chain replication is a new approach to coordinating clusters of fail-stop storage servers. The approach is intended for supporting large-scale storage services that exhibit high throughput and availability without sacrificing strong consistency guarantees. Besides outlining the chain replication protocols themselves, simulation experiments explore the performance characteristics of a prototype implementation. Throughput, availability, and several objectplacement strategies (including schemes based on distributed hash table routing) are discussed.
Deep Voice 2 : Multi-Speaker Neural Text-to-Speech
We introduce a technique for augmenting neural text-to-speech (TTS) with lowdimensional trainable speaker embeddings to generate different voices from a single model. As a starting point, we show improvements over the two state-ofthe-art approaches for single-speaker neural TTS: Deep Voice 1 and Tacotron. We introduce Deep Voice 2, which is based on a similar pipeline with Deep Voice 1, but constructed with higher performance building blocks and demonstrates a significant audio quality improvement over Deep Voice 1. We improve Tacotron by introducing a post-processing neural vocoder, and demonstrate a significant audio quality improvement. We then demonstrate our technique for multi-speaker speech synthesis for both Deep Voice 2 and Tacotron on two multi-speaker TTS datasets. We show that a single neural TTS system can learn hundreds of unique voices from less than half an hour of data per speaker, while achieving high audio quality synthesis and preserving the speaker identities almost perfectly.
A Literature Review on Kidney Disease Prediction using Data Mining Classification Technique
-The huge amounts of data generated by healthcare transactions are too complex and voluminous to be processed and analyzed by traditional methods. Data mining provides the methodology and technology to transform these mounds of data into useful information for decision making. The Healthcare industry is generally 鈥渋nformation rich鈥, which is not feasible to handle manually. These large amounts of data are very important in the field of data mining to extract useful information and generate relationships amongst the attributes. Kidney disease is a complex task which requires much experience and knowledge. Kidney disease is a silent killer in developed countries and one of the main contributors to disease burden in developing countries. In the health care industry the data mining is mainly used for predicting the diseases from the datasets. The Data mining classification techniques, namely Decision trees, ANN, Naive Bayes are analyzed on Kidney disease data set. Keywords--Data Mining, Kidney Disease, Decision tree, Naive Bayes, ANN, K-NN, SVM, Rough Set, Logistic Regression, Genetic Algorithms (GAs) / Evolutionary Programming (EP), Clustering
Real-Time 3D Reconstruction and 6-DoF Tracking with an Event Camera
We propose a method which can perform real-time 3D reconstruction from a single hand-held event camera with no additional sensing, and works in unstructured scenes of which it has no prior knowledge. It is based on three decoupled probabilistic filters, each estimating 6-DoF camera motion, scene logarithmic (log) intensity gradient and scene inverse depth relative to a keyframe, and we build a real-time graph of these to track and model over an extended local workspace. We also upgrade the gradient estimate for each keyframe into an intensity image, allowing us to recover a real-time video-like intensity sequence with spatial and temporal super-resolution from the low bit-rate input event stream. To the best of our knowledge, this is the first algorithm provably able to track a general 6D motion along with reconstruction of arbitrary structure including its intensity and the reconstruction of grayscale video that exclusively relies on event camera data.
Multiple ramp schemes
A (t; k; n; S) ramp scheme is a protocol to distribute a secret s chosen inS among a setP of n participants in such a way that: 1) sets of participants of cardinality greater than or equal to k can reconstruct the secrets; 2) sets of participants of cardinality less than or equal tot have no information on s, whereas 3) sets of participants of cardinality greater than t and less thank might have 鈥渟ome鈥 information on s. In this correspondence we analyze multiple ramp schemes, which are protocols to share many secrets among a set P of participants, using different ramp schemes. In particular, we prove a tight lower bound on the size of the shares held by each participant and on the dealer鈥檚 randomness in multiple ramp schemes.
Protein function in precision medicine: deep understanding with machine learning.
Precision medicine and personalized health efforts propose leveraging complex molecular, medical and family history, along with other types of personal data toward better life. We argue that this ambitious objective will require advanced and specialized machine learning solutions. Simply skimming some low-hanging results off the data wealth might have limited potential. Instead, we need to better understand all parts of the system to define medically relevant causes and effects: how do particular sequence variants affect particular proteins and pathways? How do these effects, in turn, cause the health or disease-related phenotype? Toward this end, deeper understanding will not simply diffuse from deeper machine learning, but from more explicit focus on understanding protein function, context-specific protein interaction networks, and impact of variation on both.
The impact of design management and process management on quality : an empirical investigation
沤 . Design management and process management are two important elements of total quality management TQM implementation. They are drastically different in their targets of improvement, visibility, and techniques. In this paper, we establish a framework for identifying the synergistic linkages of design and process management to the operational quality 沤 . 沤 . outcomes during the manufacturing process internal quality and upon the field usage of the products external quality . Through a study of quality practices in 418 manufacturing plants from multiple industries, we empirically demonstrate that both design and process management efforts have an equal positive impact on internal quality outcomes such as scrap, rework, defects, performance, and external quality outcomes such as complaints, warranty, litigation, market share. A detailed contingency analysis shows that the proposed model of synergies between design and process management holds true for large and small firms; for firms with different levels of TQM experience; and in different industries with varying levels of competition, logistical complexity of production, or production process characteristics. Finally, the results also suggest that organizational learning enables mature TQM firms to implement both design and process efforts more rigorously and their synergy helps these firms to attain better quality outcomes. These findings indicate that, to attain superior quality outcomes, firms need to balance their design and process management efforts and persevere with long-term 沤 . implementation of these efforts. Because the study spans all of the manufacturing sectors SIC 20 through 39 , these conclusions should help firms in any industry revisit their priorities in terms of the relative efforts in design management and process management. q 2000 Elsevier Science B.V. All rights reserved.
A Web Service Discovery Approach Based on Common Topic Groups Extraction
Web services have attracted much attention from distributed application designers and developers because of their roles in abstraction and interoperability among heterogeneous software systems, and a growing number of distributed software applications have been published as Web services on the Internet. Faced with the increasing numbers of Web services and service users, researchers in the services computing field have attempted to address a challenging issue, i.e., how to quickly find the suitable ones according to user queries. Many previous studies have been reported towards this direction. In this paper, a novel Web service discovery approach based on topic models is presented. The proposed approach mines common topic groups from the service-topic distribution matrix generated by topic modeling, and the extracted common topic groups can then be leveraged to match user queries to relevant Web services, so as to make a better trade-off between the accuracy of service discovery and the number of candidate Web services. Experiment results conducted on two publicly-available data sets demonstrate that, compared with several widely used approaches, the proposed approach can maintain the performance of service discovery at an elevated level by greatly decreasing the number of candidate Web services, thus leading to faster response time.
Computational Capabilities of Graph Neural Networks
In this paper, we will consider the approximation properties of a recently introduced neural network model called graph neural network (GNN), which can be used to process-structured data inputs, e.g., acyclic graphs, cyclic graphs, and directed or undirected graphs. This class of neural networks implements a function tau(G, n) isin R m that maps a graph G and one of its nodes n onto an m-dimensional Euclidean space. We characterize the functions that can be approximated by GNNs, in probability, up to any prescribed degree of precision. This set contains the maps that satisfy a property called preservation of the unfolding equivalence, and includes most of the practically useful functions on graphs; the only known exception is when the input graph contains particular patterns of symmetries when unfolding equivalence may not be preserved. The result can be considered an extension of the universal approximation property established for the classic feedforward neural networks (FNNs). Some experimental examples are used to show the computational capabilities of the proposed model.
Hipikat: a project memory for software development
Sociological and technical difficulties, such as a lack of informal encounters, can make it difficult for new members of noncollocated software development teams to learn from their more experienced colleagues. To address this situation, we have developed a tool, named Hipikat that provides developers with efficient and effective access to the group memory for a software development project that is implicitly formed by all of the artifacts produced during the development. This project memory is built automatically with little or no change to existing work practices. After describing the Hipikat tool, we present two studies investigating Hipikat's usefulness in software modification tasks. One study evaluated the usefulness of Hipikat's recommendations on a sample of 20 modification tasks performed on the Eclipse Java IDE during the development of release 2.1 of the Eclipse software. We describe the study, present quantitative measures of Hipikat's performance, and describe in detail three cases that illustrate a range of issues that we have identified in the results. In the other study, we evaluated whether software developers who are new to a project can benefit from the artifacts that Hipikat recommends from the project memory. We describe the study, present qualitative observations, and suggest implications of using project memory as a learning aid for project newcomers.
A Metrics Suite for Object Oriented Design
Given the central role that software development plays in the delivery and application of information technology, managers are increasingly focusing on process improvement in the software development area. This demand has spurred the provision of a number of new and/or improved approaches to software development, with perhaps the most prominent being object-orientation (OO). In addition, the focus on process improvement has increased the demand for software measures, or metrics with which to manage the process. The need for such metrics is particularly acute when an organization is adopting a new technology for which established practices have yet to be developed. This research addresses these needs through the development and implementation of a new suite of metrics for OO design. Metrics developed in previous research, while contributing to the field鈥檚 understanding of software development processes, have generally been subject to serious criticisms, including the lack of a theoretical base. Following Wand and Weber, the theoretical base chosen for the metrics was the ontology of Bunge. Six design metrics are developed, and then analytically evaluated against Weyuker鈥檚 proposed set of measurement principles. An automated data collection tool was then developed and implemented to collect an empirical sample of these metrics at two field sites in order to demonstrate their feasibility and suggest ways in which managers may use these metrics for process improvement. 鈥淎 Metrics Suite For Object Oriented Design鈥 Shyam R. Chidamber Chris F. Kemerer Index Terms CR
Two case studies of open source software development: Apache and Mozilla
According to its proponents, open source style software development has the capacity to compete successfully, and perhaps in many cases displace, traditional commercial development methods. In order to begin investigating such claims, we examine data from two major open source projects, the Apache web server and the Mozilla browser. By using email archives of source code change history and problem reports we quantify aspects of developer participation, core team size, code ownership, productivity, defect density, and problem resolution intervals for these OSS projects. We develop several hypotheses by comparing the Apache project with several commercial projects. We then test and refine several of these hypotheses, based on an analysis of Mozilla data. We conclude with thoughts about the prospects for high-performance commercial/open source process hybrids.
A File Comparison Program
A file comparison program produces a list of differences between two files. These differences can be couched in terms of lines, e.g. by telling which lines must be inserted, deleted or moved to convert the first file to the second. Alternatively, the list of differences can identify individual bytes. Byte-oriented comparisons are useful with non-text files, such as compiled programs, that are not divided into lines. The approach adopted here is to generate only instructions to insert or delete entire lines. Since lines are treated as indivisible objects, files can be treated as containing lines consisting of a single symbol. In other words, an n-line file is modelled by a string of n symbols. In more formal terms, the file comparison problem can be rephrased as follows. The edit distance between two strings of symbols is the length of a shortest sequence of insertions and deletions that will convert the first string to the second. T h e goal, then, is to write a program that computes the edit distance between two arbitrary strings of symbols. In addition, the program must explicitly produce a shortest possible edit script (i.e. sequence of edit commands) for the given strings. Other approaches have been tried. For example, Tichy ' discusses a file-comparison tool that determines how one file can be constructed from another by copying blocks of lines and appending lines. However, the ability to economically generate shortestpossible edit scripts depends critically on the repertoire of instructions that are allowed in the scripts.2 File comparison algorithms have a number of potential uses beside merely producing a set of edit commands to be read by someone trying to understand the evolution of a program or document. For example, the edit scripts might be text editor instructions that are saved to avoid the expense of storing nearly identical files. Rather than storing
Object Detection Featuring 3D Audio Localization for Microsoft HoloLens - A Deep Learning based Sensor Substitution Approach for the Blind
Finding basic objects on a daily basis is a difficult but common task for blind people. This paper demonstrates the implementation of a wearable, deep learning backed, object detection approach in the context of visual impairment or blindness. The prototype aims to substitute the impaired eye of the user and replace it with technical sensors. By scanning its surroundings, the prototype provides a situational overview of objects around the device. Object detection has been implemented using a near real-time, deep learning model named YOLOv2. The model supports the detection of 9000 objects. The prototype can display and read out the name of augmented objects which can be selected by voice commands and used as directional guides for the user, using 3D audio feedback. A distance announcement of a selected object is derived from the HoloLens鈥檚 spatial model. The wearable solution offers the opportunity to efficiently locate objects to support orientation without extensive training of the user. Preliminary evaluation covered the detection rate of speech recognition and the response times of the server.
English as a Formal Specification Language
PENG is a computer-processable controlled natural language designed for writing unambiguous and precise specifications. PENG covers a strict subset of standard English and is precisely defined by a controlled grammar and a controlled lexicon. In contrast to other controlled languages, the author does not need to know the grammatical restrictions explicitly. ECOLE, a look-ahead text editor, indicates the restrictions while the specification is written. The controlled lexicon contains domain-specific content words that can be defined by the author on the fly and predefined function words. Specifications written in PENG can be deterministically translated into discourse representations structures to cope with anaphora and presuppositions and also into first-order predicate logic. To test the formal properties of PENG, we reformulated Schubert鈥檚 steamroller puzzle in PENG, translated the resulting specification via discourse representation structures into first-order predicate logic with equality, and proved the steamroller鈥檚 conclusion with OTTER, a standard theorem prover.
Automatic Retraction and Full-Cycle Operation for a Class of Airborne Wind Energy Generators
Airborne wind energy systems aim to harvest the power of winds blowing at altitudes higher than what conventional wind turbines reach. They employ a tethered flying structure, usually a wing, and exploit the aerodynamic lift to produce electrical power. In the case of ground-based systems, where the traction force on the tether is used to drive a generator on the ground, a two-phase power cycle is carried out: one phase to produce power, where the tether is reeled out under high traction force, and a second phase where the tether is recoiled under lower load. The problem of controlling a tethered wing in this second phase, the retraction phase, is addressed here, by proposing two possible control strategies. Theoretical analyses, numerical simulations, and experimental results are presented to show the performance of the two approaches. Finally, the experimental results of complete autonomous power generation cycles are reported and compared with those in first-principle models.
Simulation of object and human skin formations in a grasping task
This paper addresses the problem of simulating deformations between objects and the hand of a synthetic character during a grasping process. A numerical method based on finite element theory allows us to take into account the active forces of the fingers on the object and the reactive forces of the object on the fingers. The method improves control of synthetic human behavior in a task level animation system because it provides information about the environment of a synthetic human and so can be compared to the sense of touch. Finite element theory currently used in engineering seems one of the best approaches for modeling both elastic and plastic deformation of objects, as well as shocks with or without penetration between deformable objects. We show that intrinsic properties of the method based on composition/decomposition of elements have an impact in computer animation. We also state that the use of the same method for modeling both objects and human bodies improves the modeling both objects and human bodies improves the modeling of the contacts between them. Moreover, it allows a realistic envelope deformation of the human fingers comparable to existing methods. To show what we can expect from the method, we apply it to the grasping and pressing of a ball. Our solution to the grasping problem is based on displacement commands instead of force commands used in robotics and human behavior.
Electromigration and its impact on physical design in future technologies
Electromigration (EM) is one of the key concerns going forward for interconnect reliability in integrated circuit (IC) design. Although analog designers have been aware of the EM problem for some time, digital circuits are also being affected now. This talk addresses basic design issues and their effects on electromigration during interconnect physical design. The intention is to increase current density limits in the interconnect by adopting electromigration-inhibiting measures, such as short-length and reservoir effects. Exploitation of these effects at the layout stage can provide partial relief of EM concerns in IC design flows in future.
Optimizing Space-Air-Ground Integrated Networks by Artificial Intelligence
It is widely acknowledged that the development of traditional terrestrial communication technologies cannot provide all users with fair and high quality services due to the scarce network resource and limited coverage areas. To complement the terrestrial connection, especially for users in rural, disasterstricken, or other difficult-to-serve areas, satellites, unmanned aerial vehicles (UAVs), and balloons have been utilized to relay the communication signals. On the basis, Space-Air-Ground Integrated Networks (SAGINs) have been proposed to improve the users鈥 Quality of Experience (QoE). However, compared with existing networks such as ad hoc networks and cellular networks, the SAGINs are much more complex due to the various characteristics of three network segments. To improve the performance of SAGINs, researchers are facing many unprecedented challenges. In this paper, we propose the Artificial Intelligence (AI) technique to optimize the SAGINs, as the AI technique has shown its predominant advantages in many applications. We first analyze several main challenges of SAGINs and explain how these problems can be solved by AI. Then, we consider the satellite traffic balance as an example and propose a deep learning based method to improve the traffic control performance. Simulation results evaluate that the deep learning technique can be an efficient tool to improve the performance of SAGINs.
We propose a virtual try-on method based on generative adversarial networks (GANs). By considering clothing regions, this method enables us to reflect the pattern of clothes better than Conditional Analogy GAN (CAGAN), an existing virtual try-on method based on GANs. Our method first obtains the clothing region on a person by using a human parsing model learned with a large-scale dataset. Next, using the acquired region, the clothing part is removed from a human image. A desired clothing image is added to the blank area. The network learns how to apply new clothing to the area of people鈥檚 clothing. Results demonstrate the possibility of reflecting a clothing pattern. Furthermore, an image of the clothes that the person is originally wearing becomes unnecessary during testing. In experiments, we generate images using images gathered from Zaland (a fashion E-commerce site).
Stereotypes of individuals with learning disabilities: views of college students with and without learning disabilities.
To explore possible reasons for low self-identification rates among undergraduates with learning disabilities (LD), we asked students (38 with LD, 100 without LD) attending two large, public, research-intensive universities to respond to a questionnaire designed to assess stereotypes about individuals with LD and conceptions of ability. Responses were coded into six categories of stereotypes about LD (low intelligence, compensation possible, process deficit, nonspecific insurmountable condition, working the system, and other), and into three categories of conceptions of intelligence (entity, incremental, neither). Consistent with past findings, the most frequent metastereotype reported by individuals in both groups related to generally low ability. In addition, students with LD were more likely to espouse views of intelligence as a fixed trait. As a whole, the study's findings have implications for our understanding of factors that influence self-identification and self-advocacy at the postsecondary level.
Concurrency control methods in distributed database: A review and comparison
In the last years, remarkable improvements have been made in the ability of distributed database systems performance. A distributed database is composed of some sites which are connected to each other through network connections. In this system, if good harmonization isn't made between different transactions, it may result in database incoherence. Nowadays, because of the complexity of many sites and their connection methods, it is difficult to extend different models in distributed database serially. The principle goal of concurrency control in distributed database is to ensure not interfering in accessibility of common database by different sites. Different concurrency control algorithms have been suggested to use in distributed database systems. In this paper, some available methods have been introduced and compared for concurrency control in distributed database.
Health App Use Among US Mobile Phone Users: Analysis of Trends by Chronic Disease Status
BACKGROUND Mobile apps hold promise for serving as a lifestyle intervention in public health to promote wellness and attenuate chronic conditions, yet little is known about how individuals with chronic illness use or perceive mobile apps. OBJECTIVE The objective of this study was to explore behaviors and perceptions about mobile phone-based apps for health among individuals with chronic conditions. METHODS Data were collected from a national cross-sectional survey of 1604 mobile phone users in the United States that assessed mHealth use, beliefs, and preferences. This study examined health app use, reason for download, and perceived efficacy by chronic condition. RESULTS Among participants, having between 1 and 5 apps was reported by 38.9% (314/807) of respondents without a condition and by 6.6% (24/364) of respondents with hypertension. Use of health apps was reported 2 times or more per day by 21.3% (172/807) of respondents without a condition, 2.7% (10/364) with hypertension, 13.1% (26/198) with obesity, 12.3% (20/163) with diabetes, 12.0% (32/267) with depression, and 16.6% (53/319) with high cholesterol. Results of the logistic regression did not indicate a significant difference in health app download between individuals with and without chronic conditions (P>.05). Compared with individuals with poor health, health app download was more likely among those with self-reported very good health (odds ratio [OR] 3.80, 95% CI 2.38-6.09, P<.001) and excellent health (OR 4.77, 95% CI 2.70-8.42, P<.001). Similarly, compared with individuals who report never or rarely engaging in physical activity, health app download was more likely among those who report exercise 1 day per week (OR 2.47, 95% CI 1.6-3.83, P<.001), 2 days per week (OR 4.77, 95% CI 3.27-6.94, P<.001), 3 to 4 days per week (OR 5.00, 95% CI 3.52-7.10, P<.001), and 5 to 7 days per week (OR 4.64, 95% CI 3.11-6.92, P<.001). All logistic regression results controlled for age, sex, and race or ethnicity. CONCLUSIONS Results from this study suggest that individuals with poor self-reported health and low rates of physical activity, arguably those who stand to benefit most from health apps, were least likely to report download and use these health tools.
Hadoop+Aparapi: Making heterogenous MapReduce programming easier
Lately, programmers have started to take advantage of GPU capabilities of cloud-based machines. Using the GPUs can decrease the number of nodes required to perform the computation by increasing the productivity per node. We combine Hadoop, a widely-used MapReduce framework, with Aparapi, a new Java-to-OpenCL conversion tool from AMD. We propose an easy-to-use API which allows easy implementation of MapReduce algorithms that make use of the GPU. Our API improves upon Hadoop by further hiding the complexity of GPU programming, thus allowing the programmer to concentrate on her algorithm. We also propose an accompanying refactoring that allows the programmer to specify the GPU part of their map computation by using very lightweight annotation.
A Review on Internet of Things (IoT)
Internet, a revolutionary invention, is always transforming into some new kind of hardware and software making it unavoidable for anyone. The form of communication that we see now is either human-human or human-device, but the Internet of Things (IoT) promises a great future for the internet where the type of communication is machine-machine (M2M). This paper aims to provide a comprehensive overview of the IoT scenario and reviews its enabling technologies and the sensor networks. Also, it describes a six-layered architecture of IoT and points out the related key challenges.
of Analytical Chemistry Wearable and Implantable Sensors for Biomedical Applications Hatice
Mobile health technologies offer great promise for reducing healthcare costs and improving patient care. Wearable and implantable technologies are contributing to a transformation in the mobile health era in terms of improving healthcare and health outcomes and providing real-time guidance on improved health management and tracking. In this article, we review the biomedical applications of wearable and implantable medical devices and sensors, ranging from monitoring to prevention of diseases, as well as the materials used in the fabrication of these devices and the standards for wireless medical devices and mobile applications. We conclude by discussing some of the technical challenges in wearable and implantable technology and possible solutions for overcoming these difficulties.
Effective application of process improvement patterns to business processes
Improving the operational effectiveness and efficiency of processes is a fundamental task of business process management (BPM). There exist many proposals of process improvement patterns (PIPs) as practices that aim at supporting this goal. Selecting and implementing relevant PIPs are therefore an important prerequisite for establishing process-aware information systems in enterprises. Nevertheless, there is still a gap regarding the validation of PIPs with respect to their actual business value for a specific application scenario before implementation investments are incurred. Based on empirical research as well as experiences from BPM projects, this paper proposes a method to tackle this challenge. Our approach toward the assessment of process improvement patterns considers real-world constraints such as the role of senior stakeholders or the cost of adapting available IT systems. In addition, it outlines process improvement potentials that arise from the information technology infrastructure available to organizations, particularly regarding the combination of enterprise resource planning with business process intelligence. Our approach is illustrated along a real-world business process from human resource management. The latter covers a transactional volume of about 29,000 process instances over a period of 1聽year. Overall, our approach enables both practitioners and researchers to reasonably assess PIPs before taking any process implementation decision.
A Study of Birth Order , Academic Performance , and Personality
This study aimed to investigate birth order effect on personality and academic performance amongst 120 Malaysians. Besides, it also aimed to examine the relationship between personality and academic achievement. Thirty firstborns, 30 middle children, 30 lastborns, and 30 only children, who shared the mean age of 20.0 years (SD= 1.85), were recruited into this study. Participants鈥 Sijil Pelajaran Malaysia (SPM) results were recorded and their personality was assessed by Ten Item Personality Inventory (TIPI). Results indicated that participants of different birth positions did not differ significantly in terms of personality and academic performance. However, Pearson鈥檚 correlation showed that extraversion correlated positively with academic performance. Keywordsbirth order; personality; academic achievement
A compact printed log-periodic antenna with loaded stub
A compact printed log-periodic dipole antenna (LPDA) with distributed inductive load has been presented in this paper. By adding a stub on the top of each element, the dimension of the LPAD can be reduced by 60%. The antenna has obtained an impedance bandwidth of 10GHz (8GHz-18GHz). According to the simulation results, the designed structure achieves stable radiation patterns throughout the operating frequency band.
LSDA: Large Scale Detection Through Adaptation
A major challenge in scaling object detection is the difficulty of obtaining labeled images for large numbers of categories. Recently, deep convolutional neural networks (CNNs) have emerged as clear winners on object classification benchmarks, in part due to training with 1.2M+ labeled classification images. Unfortunately, only a small fraction of those labels are available for the detection task. It is much cheaper and easier to collect large quantities of image-level labels from search engines than it is to collect detection data and label it with precise bounding boxes. In this paper, we propose Large Scale Detection through Adaptation (LSDA), an algorithm which learns the difference between the two tasks and transfers this knowledge to classifiers for categories without bounding box annotated data, turning them into detectors. Our method has the potential to enable detection for the tens of thousands of categories that lack bounding box annotations, yet have plenty of classification data. Evaluation on the ImageNet LSVRC-2013 detection challenge demonstrates the efficacy of our approach. This algorithm enables us to produce a >7.6K detector by using available classification data from leaf nodes in the ImageNet tree. We additionally demonstrate how to modify our architecture to produce a fast detector (running at 2fps for the 7.6K detector). Models and software are available at
The Parable of Google Flu: Traps in Big Data Analysis
In February 2013, Google Flu Trends (GFT) made headlines but not for a reason that Google executives or the creators of the flu tracking system would have hoped. Nature reported that GFT was predicting more than double the proportion of doctor visits for influenza-like illness (ILI) than the Centers for Disease Control and Prevention (CDC), which bases its estimates on surveillance reports from laboratories across the United States (1, 2). This happened despite the fact that GFT was built to predict CDC reports. Given that GFT is often held up as an exemplary use of big data (3, 4), what lessons can we draw from this error?
Human Mobility in a Continuum Approach
Human mobility is investigated using a continuum approach that allows to calculate the probability to observe a trip to any arbitrary region, and the fluxes between any two regions. The considered description offers a general and unified framework, in which previously proposed mobility models like the gravity model, the intervening opportunities model, and the recently introduced radiation model are naturally resulting as special cases. A new form of radiation model is derived and its validity is investigated using observational data offered by commuting trips obtained from the United States census data set, and the mobility fluxes extracted from mobile phone data collected in a western European country. The new modeling paradigm offered by this description suggests that the complex topological features observed in large mobility and transportation networks may be the result of a simple stochastic process taking place on an inhomogeneous landscape.
Automated Diagnosis of Glaucoma Using Digital Fundus Images
Glaucoma is a disease of the optic nerve caused by the increase in the intraocular pressure of the eye. Glaucoma mainly affects the optic disc by increasing the cup size. It can lead to the blindness if it is not detected and treated in proper time. The detection of glaucoma through Optical Coherence Tomography (OCT) and Heidelberg Retinal Tomography (HRT) is very expensive. This paper presents a novel method for glaucoma detection using digital fundus images. Digital image processing techniques, such as preprocessing, morphological operations and thresholding, are widely used for the automatic detection of optic disc, blood vessels and computation of the features. We have extracted features such as cup to disc (c/d) ratio, ratio of the distance between optic disc center and optic nerve head to diameter of the optic disc, and the ratio of blood vessels area in inferior-superior side to area of blood vessel in the nasal-temporal side. These features are validated by classifying the normal and glaucoma images using neural network classifier. The results presented in this paper indicate that the features are clinically significant in the detection of glaucoma. Our system is able to classify the glaucoma automatically with a sensitivity and specificity of 100% and 80% respectively.
Conditional Generative Adversarial Networks for Speech Enhancement and Noise-Robust Speaker Verification
Improving speech system performance in noisy environments remains a challenging task, and speech enhancement (SE) is one of the effective techniques to solve the problem. Motivated by the promising results of generative adversarial networks (GANs) in a variety of image processing tasks, we explore the potential of conditional GANs (cGANs) for SE, and in particular, we make use of the image processing framework proposed by Isola et al. [1] to learn a mapping from the spectrogram of noisy speech to an enhanced counterpart. The SE cGAN consists of two networks, trained in an adversarial manner: a generator that tries to enhance the input noisy spectrogram, and a discriminator that tries to distinguish between enhanced spectrograms provided by the generator and clean ones from the database using the noisy spectrogram as a condition. We evaluate the performance of the cGAN method in terms of perceptual evaluation of speech quality (PESQ), short-time objective intelligibility (STOI), and equal error rate (EER) of speaker verification (an example application). Experimental results show that the cGAN method overall outperforms the classical short-time spectral amplitude minimum mean square error (STSA-MMSE) SE algorithm, and is comparable to a deep neural network-based SE approach (DNN-SE).
Direct and Indirect Discrimination Prevention Methods
Along with privacy, discrimination is a very import ant issue when considering the legal and ethical aspects of data mini ng. It is more than obvious that most people do not want to be discriminated because of their gender, religion, nationality, age and so on, especially when those att ribu es are used for making decisions about them like giving them a job, loan, insu rance, etc. Discovering such potential biases and eliminating them from the traini ng data without harming their decision-making utility is therefore highly desirab le. For this reason, antidiscrimination techniques including discrimination discovery and prevention have been introduced in data mining. Discrimination prev ention consists of inducing patterns that do not lead to discriminatory decisio n even if the original training datasets are inherently biased. In this chapter, by focusing on the discrimination prevention, we present a taxonomy for classifying a d examining discrimination prevention methods. Then, we introduce a group of p re-processing discrimination prevention methods and specify the different featur es of each approach and how these approaches deal with direct or indirect discr imination. A presentation of metrics used to evaluate the performance of those app ro ches is also given. Finally, we conclude our study by enumerating interesting fu ture directions in this research body.
Markowitz Revisited: Mean-Variance Models in Financial Portfolio Analysis
Mean-variance portfolio analysis provided the first quantitative treatment of the tradeoff between profit and risk. We describe in detail the interplay between objective and constraints in a number of single-period variants, including semivariance models. Particular emphasis is laid on avoiding the penalization of overperformance. The results are then used as building blocks in the development and theoretical analysis of multiperiod models based on scenario trees. A key property is the possibility of removing surplus money in future decisions, yielding approximate downside risk minimization.
Federated Learning for Keyword Spotting
We propose a practical approach based on federated learning to solve out-of-domain issues with continuously running embedded speech-based models such as wake word detectors. We conduct an extensive empirical study of the federated averaging algorithm for the 鈥淗ey Snips鈥 wake word based on a crowdsourced dataset that mimics a federation of wake word users. We empirically demonstrate that using an adaptive averaging strategy inspired from Adam in place of standard weighted model averaging highly reduces the number of communication rounds required to reach our target performance. The associated upstream communication costs per user are estimated at 8 MB, which is a reasonable in the context of smart home voice assistants. Additionally, the dataset used for these experiments is being open sourced with the aim of fostering further transparent research in the application of federated learning to speech data.
Sample-Based Tree Search with Fixed and Adaptive State Abstractions
Sample-based tree search (SBTS) is an approach to solving Markov decision problems based on constructing a lookahead search tree using random samples from a generative model of the MDP. It encompasses Monte Carlo tree search (MCTS) algorithms like UCT as well as algorithms such as sparse sampling. SBTS is well-suited to solving MDPs with large state spaces due to the relative insensitivity of SBTS algorithms to the size of the state space. The limiting factor in the performance of SBTS tends to be the exponential dependence of sample complexity on the depth of the search tree. The number of samples required to build a search tree is O((|A|B)), where |A| is the number of available actions, B is the number of possible random outcomes of taking an action, and d is the depth of the tree. State abstraction can be used to reduce B by aggregating random outcomes together into abstract states. Recent work has shown that abstract tree search often performs substantially better than tree search conducted in the ground state space. This paper presents a theoretical and empirical evaluation of tree search with both fixed and adaptive state abstractions. We derive a bound on regret due to state abstraction in tree search that decomposes abstraction error into three components arising from properties of the abstraction and the search algorithm. We describe versions of popular SBTS algorithms that use fixed state abstractions, and we introduce the Progressive Abstraction Refinement in Sparse Sampling (PARSS) algorithm, which adapts its abstraction during search. We evaluate PARSS as well as sparse sampling with fixed abstractions on 12 experimental problems, and find that PARSS outperforms search with a fixed abstraction and that search with even highly inaccurate fixed abstractions outperforms search without abstraction. These results establish progressive abstraction refinement as a promising basis for new tree search algorithms, and we propose directions for future work within the progressive refinement framework.
Query by Committee
We propose an algorithm called query by commitee, in which a committee of students is trained on the same data set. The next query is chosen according to the principle of maximal disagreement. The algorithm is studied for two toy models: the high-low game and perceptron learning of another perceptron. As the number of queries goes to infinity, the committee algorithm yields asymptotically finite information gain. This leads to generalization error that decreases exponentially with the number of examples. This in marked contrast to learning from randomly chosen inputs, for which the information gain approaches zero and the generalization error decreases with a relatively slow inverse power law. We suggest that asymptotically finite information gain may be an important characteristic of good query algorithms.
Adaptive Manifold Learning
Manifold learning algorithms seek to find a low-dimensional parameterization of high-dimensional data. They heavily rely on the notion of what can be considered as local, how accurately the manifold can be approximated locally, and, last but not least, how the local structures can be patched together to produce the global parameterization. In this paper, we develop algorithms that address two key issues in manifold learning: 1) the adaptive selection of the local neighborhood sizes when imposing a connectivity structure on the given set of high-dimensional data points and 2) the adaptive bias reduction in the local low-dimensional embedding by accounting for the variations in the curvature of the manifold as well as its interplay with the sampling density of the data set. We demonstrate the effectiveness of our methods for improving the performance of manifold learning algorithms using both synthetic and real-world data sets.
Manifold-Learning-Based Feature Extraction for Classification of Hyperspectral Data: A Review of Advances in Manifold Learning
Advances in hyperspectral sensing provide new capability for characterizing spectral signatures in a wide range of physical and biological systems, while inspiring new methods for extracting information from these data. HSI data often lie on sparse, nonlinear manifolds whose geometric and topological structures can be exploited via manifold-learning techniques. In this article, we focused on demonstrating the opportunities provided by manifold learning for classification of remotely sensed data. However, limitations and opportunities remain both for research and applications. Although these methods have been demonstrated to mitigate the impact of physical effects that affect electromagnetic energy traversing the atmosphere and reflecting from a target, nonlinearities are not always exhibited in the data, particularly at lower spatial resolutions, so users should always evaluate the inherent nonlinearity in the data. Manifold learning is data driven, and as such, results are strongly dependent on the characteristics of the data, and one method will not consistently provide the best results. Nonlinear manifold-learning methods require parameter tuning, although experimental results are typically stable over a range of values, and have higher computational overhead than linear methods, which is particularly relevant for large-scale remote sensing data sets. Opportunities for advancing manifold learning also exist for analysis of hyperspectral and multisource remotely sensed data. Manifolds are assumed to be inherently smooth, an assumption that some data sets may violate, and data often contain classes whose spectra are distinctly different, resulting in multiple manifolds or submanifolds that cannot be readily integrated with a single manifold representation. Developing appropriate characterizations that exploit the unique characteristics of these submanifolds for a particular data set is an open research problem for which hierarchical manifold structures appear to have merit. To date, most work in manifold learning has focused on feature extraction from single images, assuming stationarity across the scene. Research is also needed in joint exploitation of global and local embedding methods in dynamic, multitemporal environments and integration with semisupervised and active learning.
Stochastic Neighbor Embedding
We describe a probabilistic approach to the task of placing objects, described by high-dimensional vectors or by pairwise dissimilarities, in a low-dimensional space in a way that preserves neighbor identities. A Gaussian is centered on each object in the high-dimensional space and the densities under this Gaussian (or the given dissimilarities) are used to define a probability distribution over all the potential neighbors of the object. The aim of the embedding is to approximate this distribution as well as possible when the same operation is performed on the low-dimensional 鈥渋mages鈥 of the objects. A natural cost function is a sum of Kullback-Leibler divergences, one per object, which leads to a simple gradient for adjusting the positions of the low-dimensional images. Unlike other dimensionality reduction methods, this probabilistic framework makes it easy to represent each object by a mixture of widely separated low-dimensional images. This allows ambiguous objects, like the document count vector for the word 鈥渂ank鈥, to have versions close to the images of both 鈥渞iver鈥 and 鈥渇inance鈥 without forcing the images of outdoor concepts to be located close to those of corporate concepts.
Nonlinear Component Analysis as a Kernel Eigenvalue Problem
A new method for performing a nonlinear form of principal component analysis is proposed. By the use of integral operator kernel functions, one can efficiently compute principal components in high-dimensional feature spaces, related to input space by some nonlinear mapfor instance, the space of all possible five-pixel products in 16 16 images. We give the derivation of the method and present experimental results on polynomial feature extraction for pattern recognition.
Leveraging graph topology and semantic context for pharmacovigilance through twitter-streams
Adverse drug events (ADEs) constitute one of the leading causes of post-therapeutic death and their identification constitutes an important challenge of modern precision medicine. Unfortunately, the onset and effects of ADEs are often underreported complicating timely intervention. At over 500 million posts per day, Twitter is a commonly used social media platform. The ubiquity of day-to-day personal information exchange on Twitter makes it a promising target for data mining for ADE identification and intervention. Three technical challenges are central to this problem: (1) identification of salient medical keywords in (noisy) tweets, (2) mapping drug-effect relationships, and (3) classification of such relationships as adverse or non-adverse. We use a bipartite graph-theoretic representation called a drug-effect graph (DEG) for modeling drug and side effect relationships by representing the drugs and side effects as vertices. We construct individual DEGs on two data sources. The first DEG is constructed from the drug-effect relationships found in FDA package inserts as recorded in the SIDER database. The second DEG is constructed by mining the history of Twitter users. We use dictionary-based information extraction to identify medically-relevant concepts in tweets. Drugs, along with co-occurring symptoms are connected with edges weighted by temporal distance and frequency. Finally, information from the SIDER DEG is integrate with the Twitter DEG and edges are classified as either adverse or non-adverse using supervised machine learning. We examine both graph-theoretic and semantic features for the classification task. The proposed approach can identify adverse drug effects with high accuracy with precision exceeding 85聽% and F1 exceeding 81聽%. When compared with leading methods at the state-of-the-art, which employ un-enriched graph-theoretic analysis alone, our method leads to improvements ranging between 5 and 8聽% in terms of the aforementioned measures. Additionally, we employ our method to discover several ADEs which, though present in medical literature and Twitter-streams, are not represented in the SIDER databases. We present a DEG integration model as a powerful formalism for the analysis of drug-effect relationships that is general enough to accommodate diverse data sources, yet rigorous enough to provide a strong mechanism for ADE identification.
We present a method for automatic object localization and recognition in 3D point clouds representing outdoor urban scenes. The method is based on the implicit shape models (ISM) framework, which recognizes objects by voting for their center locations. It requires only few training examples per class, which is an important property for practical use. We also introduce and evaluate an improved version of the spin image descriptor, more robust to point density variation and uncertainty in normal direction estimation. Our experiments reveal a significant impact of these modifications on the recognition performance. We compare our results against the state-of-the-art method and get significant improvement in both precision and recall on the Ohio dataset, consisting of combined aerial and terrestrial LiDAR scans of 150,000 m of urban area in total.
Multi-task Learning for Maritime Traffic Surveillance from AIS Data Streams
In a world of global trading, maritime safety, security and efficiency are crucial issues. We propose a multi-task deep learning framework for vessel monitoring using Automatic Identification System (AIS) data streams. We combine recurrent neural networks with latent variable modeling and an embedding of AIS messages to a new representation space to jointly address key issues to be dealt with when considering AIS data streams: massive amount of streaming data, noisy data and irregular timesampling. We demonstrate the relevance of the proposed deep learning framework on real AIS datasets for a three-task setting, namely trajectory reconstruction, anomaly detection and vessel type identification.
The Dysphagia Outcome and Severity Scale
The Dysphagia Outcome and Severity Scale (DOSS) is a simple, easy-to-use, 7-point scale developed to systematically rate the functional severity of dysphagia based on objective assessment and make recommendations for diet level, independence level, and type of nutrition. Intra- and interjudge reliabilities of the DOSS was established by four clinicians on 135 consecutive patients who underwent a modified barium swallow procedure at a large teaching hospital. Patients were assigned a severity level, independence level, and nutritional level based on three areas most associated with final recommendations: oral stage bolus transfer, pharyngeal stage retention, and airway protection. Results indicate high interrater (90%) and intrarater (93%) agreement with this scale. Implications are suggested for use of the DOSS in documenting functional outcomes of swallowing and diet status based on objective assessment.
Depth Map Super-Resolution Considering View Synthesis Quality
Accurate and high-quality depth maps are required in lots of 3D applications, such as multi-view rendering, 3D reconstruction and 3DTV. However, the resolution of captured depth image is much lower than that of its corresponding color image, which affects its application performance. In this paper, we propose a novel depth map super-resolution (SR) method by taking view synthesis quality into account. The proposed approach mainly includes two technical contributions. First, since the captured low-resolution (LR) depth map may be corrupted by noise and occlusion, we propose a credibility based multi-view depth maps fusion strategy, which considers the view synthesis quality and interview correlation, to refine the LR depth map. Second, we propose a view synthesis quality based trilateral depth-map up-sampling method, which considers depth smoothness, texture similarity and view synthesis quality in the up-sampling filter. Experimental results demonstrate that the proposed method outperforms state-of-the-art depth SR methods for both super-resolved depth maps and synthesized views. Furthermore, the proposed method is robust to noise and achieves promising results under noise-corruption conditions.
A New Approach to Linear Filtering and Prediction Problems
AN IMPORTANT class of theoretical and practical problems in communication and control is of a statistical nature. Such problems are: (i) Prediction of random signals; (ii) separation of random signals from random noise; (iii) detection of signals of known form (pulses, sinusoids) in the presence of random noise. In his pioneering work, Wiener [1]3 showed that problems (i) and (ii) lead to the so-called Wiener-Hopf integral equation; he also gave a method (spectral factorization) for the solution of this integral equation in the practically important special case of stationary statistics and rational spectra. Many extensions and generalizations followed Wiener鈥檚 basic work. Zadeh and Ragazzini solved the finite-memory case [2]. Concurrently and independently of Bode and Shannon [3], they also gave a simplified method [2) of solution. Booton discussed the nonstationary Wiener-Hopf equation [4]. These results are now in standard texts [5-6]. A somewhat different approach along these main lines has been given recently by Darlington [7]. For extensions to sampled signals, see, e.g., Franklin [8], Lees [9]. Another approach based on the eigenfunctions of the WienerHopf equation (which applies also to nonstationary problems whereas the preceding methods in general don鈥檛), has been pioneered by Davis [10] and applied by many others, e.g., Shinbrot [11], Blum [12], Pugachev [13], Solodovnikov [14]. In all these works, the objective is to obtain the specification of a linear dynamic system (Wiener filter) which accomplishes the prediction, separation, or detection of a random signal.4 鈥斺斺 1 This research was supported in part by the U. S. Air Force Office of Scientific Research under Contract AF 49 (638)-382. 2 7212 Bellona Ave. 3 Numbers in brackets designate References at end of paper. 4 Of course, in general these tasks may be done better by nonlinear filters. At present, however, little or nothing is known about how to obtain (both theoretically and practically) these nonlinear filters. Contributed by the Instruments and Regulators Division and presented at the Instruments and Regulators Conference, March 29鈥 Apri1 2, 1959, of THE AMERICAN SOCIETY OF MECHANICAL ENGINEERS. NOTE: Statements and opinions advanced in papers are to be understood as individual expressions of their authors and not those of the Society. Manuscript received at ASME Headquarters, February 24, 1959. Paper No. 59-IRD鈥11. A New Approach to Linear Filtering and Prediction Problems