Example queries for CLEF e-Health 2015
Task 2: User-Centred Health Information Retrieval

Last update: 5 Mar 2015
===============================================================================

Files:
------

* clef2015.training.queries-EN.txt
-------------------------------------
This file contains 5 example queries (in English). 
This year we will explore circumlocutory queries that users may pose when 
faced with signs and symptoms of a medical condition. 
To form the query set, we adopted the method recently 
investigated by [1,2].
-------------------------------------

* AR/
    clef2015.training.queries-AR.txt
    clef2015.training.queries-AR-auto.txt  
* CS/
    clef2015.training.queries-CS.txt
    clef2015.training.queries-CS-auto.txt  
* DE/
    clef2015.training.queries-DE.txt
    clef2015.training.queries-DE-auto.txt  
* ES/
    clef2015.training.queries-ES.txt
    clef2015.training.queries-ES-auto.txt  
* FA/
    clef2015.training.queries-FA.txt
    clef2015.training.queries-FA-auto.txt  
* FR/
    clef2015.training.queries-FR.txt
    clef2015.training.queries-FR-auto.txt  
* IT/
    clef2015.training.queries-IT.txt
    clef2015.training.queries-IT-auto.txt  
* PT/
    clef2015.training.queries-PT.txt
    clef2015.training.queries-PT-auto.txt  

-------------------------------------
Every query was translated from English to a target language (AR: Arabic, 
CS: Czech, DE: German, ES: Spanish, FA: Farsi, FR: French, IT: Italian, 
PT: Portuguese).

We also provide a baseline translation made by Google Translator (files containing
"auto" in their names).
-------------------------------------

* translations.xml
-------------------------------------
We provide the translations.xml file for an easier visualization 
of the automatically generated translations.
-------------------------------------

* clef2015.qrels.training.txt
-------------------------------------
A pool containing the top 10 results of different systems was made and
manually evaluated. Only the query field of each topic was used.
-------------------------------------

* clef2015.qrels.training.additional.txt
-------------------------------------
The same procedure described above was used to generate this file, but
the disease name from <narr> field was manually appended to the query.
-------------------------------------

-------------------------------------------------------------------------------

References:
-------------------------------------------------------------------------------
[1] Isabelle Stanton, Samuel Ieong, Nina Mishra. SigIR 2014.
[2] Guido Zuccon, Bevan Koopman and Joao Palotti. ECIR 2015.
-------------------------------------------------------------------------------


===============================================================================