W3C home > Mailing lists > Public > public-semweb-lifesci@w3.org > October 2012

Request of collaborating for making a benchmark for semantic search

From: Saeedeh <shekarpour@informatik.uni-leipzig.de>
Date: Tue, 9 Oct 2012 18:38:26 +0200
To: <public-semweb-lifesci@w3.org>
Cc: <ngonga@informatik.uni-leipzig.de>, "'Sören Auer'" <auer@informatik.uni-leipzig.de>
Message-ID: <000901cda63c$84c712e0$8e5538a0$@uni-leipzig.de>
Dear all,

I am carrying some research work on federated information retrieval based on keyword queries. One of the most challenging tasks in this respect lies in creating a benchmark for evaluating the precision and recall of the SPARQL queries that are derived from the user input (which consists of a sequence of keywords or of a natural language question). This is where I need your help as bio-medical experts. I would like to create a freely available benchmark for Semantic Web researchers based on the data contained in Sider, Drugbank, DBpedia and Diseasome. Could you please provide me with queries that can be answered by combining these datasets? I created a sheet which you can find at
https://docs.google.com/spreadsheet/ccc?key=0ArCk_xV46hT5dGZnTktIakNUbEVTcGF2b0dSX2VTLWc

 

 

to collect your queries. Therein, you can find two examples of queries and corresponding SPARQL queries (see third column).



The involving datasets have been loaded on this endpoint: <http://139.18.2.96:8910/sparql> http://139.18.2.96:8910/sparql 

 

with the graph names: 

 

 < <http://www4.wiwiss.fu-berlin.de/drugbank> http://www4.wiwiss.fu-berlin.de/drugbank>

 < <http://www4.wiwiss.fu-berlin.de/diseasome> http://www4.wiwiss.fu-berlin.de/diseasome>

 < <http://www4.wiwiss.fu-berlin.de/sider> http://www4.wiwiss.fu-berlin.de/sider>

 < <http://dbpedia.org/> http://dbpedia.org>

 

 

\Note,  if writing SPARQL queries is difficult, so we can skip this part and if you provide me just some natural language or keyword-based queries over the involving data sets would be great. 

 

Probably you have come across at least one single query about drugs/side effects of them and diseases...

I would like to pick up common queries, that is why I am asking crowd.

 


Many thanks in advance

 


-- 



Saeedeh Shekarpour

Phd student

Department of Computer Science, University of Leipzig

Research Group:  <http://aksw.org/> http://aksw.org

 

 

هر آنکسی که دراین حلقه نیست زنده به عشق 
بر او نمرده به فتوی من نماز کنید

 
Received on Wednesday, 10 October 2012 15:47:51 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:21:17 UTC