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

BioCreAtIvE task1A: entity identification with a stochastic tagger.

Shuhei Kinoshita1, K Bretonnel Cohen, Philip V Ogren

  • 1Center for Computational Pharmacology, University of Colorado School of Medicine, Denver, Colorado, USA. kino@strad.ssg.fujitsu.com

BMC Bioinformatics
|June 18, 2005
PubMed
Summary

This study enhanced gene entity recognition by augmenting a part-of-speech tagger with post-processing rules. The improved system achieved high precision and recall, demonstrating its effectiveness in identifying genetic terms.

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

The future of fundamental science led by generative closed-loop artificial intelligence.

Frontiers in artificial intelligence·2026
Same author

Improving biomedical entity linking with generative relevance feedback.

Bioinformatics (Oxford, England)·2026
Same author

Desiderata for a biomedical knowledge network: opportunities, challenges and future Directions.

ArXiv·2025
Same author

Biological databases in the age of generative artificial intelligence.

Bioinformatics advances·2025
Same author

Hypothesizing mechanistic links between microbes and disease using knowledge graphs.

bioRxiv : the preprint server for biology·2023
Same author

Characterization of methods for mechanistic inference of the gut microbiome in disease.

bioRxiv : the preprint server for biology·2023

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Natural Language Processing

Background:

  • The study builds upon existing gene entity recognition systems, specifically Tanabe and Wilbur's ABGene system.
  • It frames the problem as a part-of-speech tagging task, incorporating a specific GENE tag.
  • The research utilizes the Trigrams 'n' Tags (TnT) HMM-based part-of-speech tagger.

Purpose of the Study:

  • To develop an effective system for gene entity identification.
  • To improve upon existing methods by incorporating post-processing rules for enhanced accuracy.
  • To evaluate the system's performance in both open and closed competition divisions.

Main Methods:

  • The core method involves using the TnT part-of-speech tagger, extended with a GENE tag.

Related Experiment Videos

  • A critical component is the implementation of post-processing rules derived from error analysis to refine predictions.
  • For the open division, external data from NCBI was leveraged.

    • Main Results:

    • The base system achieved an F-measure of 72.3% (68.0% precision, 77.2% recall).
    • The system with post-processing rules significantly improved performance, reaching an F-measure of 80.4% (80.3% precision, 80.5% recall).
    • An additional dictionary-based post-processing step for the open division yielded a marginal improvement (F-measure = 80.9%).

    Conclusions:

    • Augmenting part-of-speech taggers with post-processing rules is a viable strategy for building competitive entity identification systems.
    • The developed system demonstrates strong performance, achieving third place in both open and closed divisions.
    • This approach highlights the effectiveness of combining statistical tagging with rule-based refinement for biological named entity recognition.