Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Collocation analysis for UMLS knowledge-based word sense disambiguation.

Antonio Jimeno-Yepes1, Bridget T McInnes, Alan R Aronson

  • 1National Library of Medicine, 8600 Rockville Pike, Bethesda, MD 20894, USA. antonio.jimeno@gmail.com

BMC Bioinformatics
|June 11, 2011
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Predicting implicit concept embeddings for singular relationship discovery replication of closed literature-based discovery.

Frontiers in research metrics and analytics·2025
Same author

CACER: Clinical concept Annotations for Cancer Events and Relations.

Journal of the American Medical Informatics Association : JAMIA·2024
Same author

Exploring Representations for Singular and Multi-Concept Relations for Biomedical Named Entity Normalization.

Proceedings of the ... International World-Wide Web Conference. International WWW Conference·2023
Same author

Exploring a deep learning neural architecture for closed Literature-based discovery.

Journal of biomedical informatics·2023
Same author

An overview of biomedical entity linking throughout the years.

Journal of biomedical informatics·2022
Same author

Temporal disambiguation of relative temporal expressions in clinical texts.

Frontiers in research metrics and analytics·2022
Same journal

OpenIMC: an open-source platform for analyzing single-cell and spatial proteomics by imaging mass cytometry.

BMC bioinformatics·2026
Same journal

NAP: an open source pipeline for cross-domain microbiome profiling using Nanopore sequencing-derived amplicon data.

BMC bioinformatics·2026
Same journal

SurvGME: an R package for survival analysis with graphical and measurement error models.

BMC bioinformatics·2026
Same journal

SimMapNet: a Bayesian framework for gene regulatory network inference using gene ontology similarities as external hint.

BMC bioinformatics·2026
Same journal

Dual channel drug-drug interactions extraction based on cross attention.

BMC bioinformatics·2026
Same journal

FeSseqdb: a curated sequence-level database and interpretable machine learning framework for identifying iron-sulfur proteins.

BMC bioinformatics·2026
See all related articles

Collocations enhance knowledge-based word sense disambiguation (WSD) performance by improving context modeling. Results vary by dataset and method, with collocations benefiting both AEC and MRD approaches.

Area of Science:

  • Natural Language Processing
  • Computational Linguistics
  • Biomedical Informatics

Background:

  • Knowledge-based word sense disambiguation (WSD) effectiveness relies on reference resources, which often lack WSD optimization and may contain noise.
  • Existing resources may not adequately model context or can introduce false positives.
  • This study addresses limitations in current WSD knowledge resources.

Purpose of the Study:

  • To analyze collocation types for improving knowledge-based WSD performance.
  • To evaluate the impact of collocations on two distinct WSD methods: AEC and MRD.
  • To assess the utility of semantic group profiles and second-order features in conjunction with collocations.

Main Methods:

  • Candidate collocations were extracted from MEDLINE and assigned to ambiguous word senses using semantic group profiles or a knowledge-based method.

Related Experiment Videos

  • The study utilized two WSD test sets: the NLM WSD set (expert-curated) and the MSH WSD set (automatically generated from MeSH indexing).
  • The performance of collocations was measured within the AEC (Naïve Bayes trained on UMLS/MEDLINE examples) and MRD (UMLS-based profile comparison) WSD methods.

    • Main Results:

    • Collocations generally improved WSD performance, with variations across datasets and methods.
    • In the NLM WSD set, the MRD method with second-order features showed greater improvement; semantic group profiles benefited the AEC method.
    • The MSH WSD set showed modest improvements, with collocations and MRD performing best, while AEC saw moderate gains.

    Conclusions:

    • Collocations demonstrably enhance knowledge-based WSD, though efficacy depends on the test set and specific method.
    • The AEC method benefits significantly from a few selected terms but is sensitive to query drift.
    • The MRD method is more robust to noisy terms but requires a larger set of collocations for substantial performance gains.