Paper: Towards Retrieving Relevant Information for Answering Clinical Comparison Questions

ACL ID W09-1320
Title Towards Retrieving Relevant Information for Answering Clinical Comparison Questions
Venue Workshop on Biomedical Natural Language Processing
Session
Year 2009
Authors

This paper introduces the task of automatical- ly answering clinical comparison questions using MEDLINE® abstracts. In the beginning, clinical comparison questions and the main challenges in recognising and extracting their components are described. Then, different strategies for retrieving MEDLINE® abstracts are shown. Finally, the results of an initial ex- periment judging the relevance of MEDLINE® abstracts retrieved by searching for the com- ponents of twelve comparison questions will be shown and discussed.