On the potential search effectiveness of MeSH (medical subject headings) terms

  • Authors:
  • Ying-Hsang Liu

  • Affiliations:
  • Charles Sturt University, Wagga Wagga, Australia

  • Venue:
  • Proceedings of the third symposium on Information interaction in context
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

The creation of innovative techniques of document representation is critical to the development of effective information retrieval (IR) systems. In this paper, we report on the impact of state-of-the-art human indexing techniques, exemplified by MeSH (Medical Subject Headings) terms, in the document representation. We studied queries formulated by four different kinds of information seekers interactively using an experimental IR system: (1) search novices; (2) domain experts; (3) search experts and (4) medical librarians. The 3,442,321 documents came from the TREC 2004 Genomics Track document set. Effectiveness of retrieval was measured using the relevance judgments provided by TREC. Inclusion of MeSH terms in the document representation did not affect the effectiveness of queries with respect to precision and recall. Adding MeSH terms to the index did not have a positive impact on the effectiveness of queries formulated by different kinds of users. These findings contribute to our understanding of the associations between the users' cognitive space and the information objects in light of the principle of polyrepresentation.