Combining thesauri-based methods for biomedical retrieval

Edgar Meij, Leonie Ijzereef, Leif Azzopardi, Jaap Kamps, Maarten De Rijke

Research output: Contribution to journalConference article

Abstract

This paper describes our participation in the TREC 2005 Genomics track. We took part in the ad hoc retrieval task and aimed at integrating thesauri in the retrieval model. We developed three thesauri-based methods, two of which made use of the existing MeSH thesaurus. One method uses blind relevance feedback on MeSH terms, the second uses an index of the MeSH thesaurus for query expansion. The third method makes use of a dynamically generated lookup list, by which acronyms and synonyms could be inferred. We show that, despite the relatively minor improvements in retrieval performance of individually applied methods, a combination works best and is able to deliver significant improvements over the baseline.

Fingerprint

Thesauri
Feedback

Keywords

  • thesauri
  • semantic indexing
  • information retrieval
  • ad-hoc retrieval tasks
  • lookups
  • query expansion
  • retrieval models
  • biomedical retrieval

Cite this

Meij, Edgar ; Ijzereef, Leonie ; Azzopardi, Leif ; Kamps, Jaap ; De Rijke, Maarten. / Combining thesauri-based methods for biomedical retrieval. In: NIST Special Publication. 2005.
@article{e5f63f6712934c249179371b0f3f3664,
title = "Combining thesauri-based methods for biomedical retrieval",
abstract = "This paper describes our participation in the TREC 2005 Genomics track. We took part in the ad hoc retrieval task and aimed at integrating thesauri in the retrieval model. We developed three thesauri-based methods, two of which made use of the existing MeSH thesaurus. One method uses blind relevance feedback on MeSH terms, the second uses an index of the MeSH thesaurus for query expansion. The third method makes use of a dynamically generated lookup list, by which acronyms and synonyms could be inferred. We show that, despite the relatively minor improvements in retrieval performance of individually applied methods, a combination works best and is able to deliver significant improvements over the baseline.",
keywords = "thesauri, semantic indexing, information retrieval, ad-hoc retrieval tasks, lookups, query expansion, retrieval models, biomedical retrieval",
author = "Edgar Meij and Leonie Ijzereef and Leif Azzopardi and Jaap Kamps and {De Rijke}, Maarten",
year = "2005",
month = "12",
day = "1",
language = "English",
journal = "NIST Special Publication",
issn = "1048-776X",

}

Combining thesauri-based methods for biomedical retrieval. / Meij, Edgar; Ijzereef, Leonie; Azzopardi, Leif; Kamps, Jaap; De Rijke, Maarten.

In: NIST Special Publication, 01.12.2005.

Research output: Contribution to journalConference article

TY - JOUR

T1 - Combining thesauri-based methods for biomedical retrieval

AU - Meij, Edgar

AU - Ijzereef, Leonie

AU - Azzopardi, Leif

AU - Kamps, Jaap

AU - De Rijke, Maarten

PY - 2005/12/1

Y1 - 2005/12/1

N2 - This paper describes our participation in the TREC 2005 Genomics track. We took part in the ad hoc retrieval task and aimed at integrating thesauri in the retrieval model. We developed three thesauri-based methods, two of which made use of the existing MeSH thesaurus. One method uses blind relevance feedback on MeSH terms, the second uses an index of the MeSH thesaurus for query expansion. The third method makes use of a dynamically generated lookup list, by which acronyms and synonyms could be inferred. We show that, despite the relatively minor improvements in retrieval performance of individually applied methods, a combination works best and is able to deliver significant improvements over the baseline.

AB - This paper describes our participation in the TREC 2005 Genomics track. We took part in the ad hoc retrieval task and aimed at integrating thesauri in the retrieval model. We developed three thesauri-based methods, two of which made use of the existing MeSH thesaurus. One method uses blind relevance feedback on MeSH terms, the second uses an index of the MeSH thesaurus for query expansion. The third method makes use of a dynamically generated lookup list, by which acronyms and synonyms could be inferred. We show that, despite the relatively minor improvements in retrieval performance of individually applied methods, a combination works best and is able to deliver significant improvements over the baseline.

KW - thesauri

KW - semantic indexing

KW - information retrieval

KW - ad-hoc retrieval tasks

KW - lookups

KW - query expansion

KW - retrieval models

KW - biomedical retrieval

UR - http://www.scopus.com/inward/record.url?scp=84873531117&partnerID=8YFLogxK

M3 - Conference article

JO - NIST Special Publication

T2 - NIST Special Publication

JF - NIST Special Publication

SN - 1048-776X

ER -