Query Expansion Using SNOMED-CT and Weighing Schemes.

摘要

Despite all the advancements that have been made in the field of Information Retrieval there are still so many challenges. These challenges are magnified when the information that is being retrieved is in a specialized domain such as healthcare. In order to tackle these challenges and encourage research in these domains, TREC (Text RETrival Conference) has instituted a Clinical Track in 2014. This paper is the result of participation in 2014 TREC Clinical Track. It entails the approach and the results that were obtained by utilizing Ontology to expand the original topics. Ontology was used in order to improve the quality of the terms present in the queries or topics, so that the queries are better structured, and they can better target documents of interest. The value that each term brings to the result was measured by way of weighing method algorithms in the retrieval system, BM25 and InL2c1. For this research, we have used SNOMED-CT along with UMLS Methathesaurus as our ontology in medical domain to expand the queries.