Performance evaluation of three semantic expansions to query PubMed - Archive ouverte HAL Access content directly
Journal Articles Health Information and Libraries Journal Year : 2019

Performance evaluation of three semantic expansions to query PubMed

Abstract

Background PubMed is one of the most important basic tools to access medical literature. Semantic query expansion using synonyms can improve retrieval efficacy. Objective The objective was to evaluate the performance of three semantic query expansion strategies. Methods Queries were built for forty MeSH descriptors using three semantic expansion strategies (MeSH synonyms, UMLS mappings, and mappings created by the CISMeF team), then sent to PubMed. To evaluate expansion performances for each query, the first twenty citations were selected, and their relevance were judged by three independent evaluators based on the title and abstract. Results Queries built with the UMLS expansion provided new citations with a slightly higher mean precision (74.19%) than with the CISMeF expansion (70.28%), although the difference was not significant. Inter‐rater agreement was 0.28. Results varied greatly depending on the descriptor selected. Discussion The number of citations retrieved by the three strategies and their precision varied greatly according to the descriptor. This heterogeneity could be explained by the quality of the synonyms. Optimal use of these different expansions would be through various combinations of UMLS and CISMeF intersections or unions. Conclusion Information retrieval tools should propose different semantic expansions depending on the descriptor and the search objectives.
No file

Dates and versions

hal-02417189 , version 1 (18-12-2019)

Identifiers

Cite

Clément Massonnaud, Romain Lelong, Gaetan Kerdelhué, Émeline Lejeune, Julien Grosjean, et al.. Performance evaluation of three semantic expansions to query PubMed. Health Information and Libraries Journal, 2019, ⟨10.1111/hir.12291⟩. ⟨hal-02417189⟩
206 View
0 Download

Altmetric

Share

Gmail Facebook X LinkedIn More