QUERY EXPANSION AND MEDLINE

This paper evaluates the retrieval effectiveness of query expansion st rategies on a MEDLINE test collection using Cornell University's SMART retrieval system. Three expansion strategies are tested on their abil ity to identify appropriate MeSH terms for user queries: expansion usi ng an inter-field statistical thesaurus, expansion via retrieval feedb ack and expansion using a combined approach. These expansion strategie s do not require prior relevance decisions. The study compares retriev al effectiveness using the original unexpanded and the alternative exp anded user queries on a collection of 75 queries and 2334 MEDLINE cita tions. Retrieval effectiveness is assessed using eleven point average precision scores (11-AvgP). The combination of expansion using the the saurus followed by retrieval feedback gives the best improvement of 17 % over a baseline performance of 0.5169 11-AvgP. However this improvem ent is almost identical to that achieved by expansion via retrieval fe edback (16.4%). Query expansion using the inter-field thesaurus gives a significant but lower performance improvement (9.9%) over the same b aseline. This study recommends query expansion using retrieval feedbac k for adding MeSH search terms to a user's initial query. Copyright (C ) 1996 Elsevier Science Ltd