Semantic Graph Based Document-Indexing Strategy for MEDLINE

Abstract

The indexing vocabulary is an important determinant of success in text retrieval. Researchers have compared the effectiveness of indexing strategies using free-text and controlled vocabularies in a variety of text contexts. This paper introduces a new approach in creating indices for medical literature using the Semantic Graph (SG) derived from both UMLS Metathesaurus and WordNet. Our performance studies indicate that using the Semantic Graph based strategy results in an increase in 11pt-average precision over traditional free-text and MeSH indexed strategies.