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

A shallow parser based on closed-class words to capture relations in biomedical text.

Gondy Leroy1, Hsinchun Chen, Jesse D Martinez

  • 1Management Information Systems, The University of Arizona, McClelland Hall, Room 430, 1130 E. Helen St., Tucson, AZ 85721, USA. gondy.leroy@cgu.edu

Journal of Biomedical Informatics
|November 15, 2003
PubMed
Summary

This study introduces a novel shallow parser for biomedical text, improving relation extraction accuracy. The system achieved 90% precision, identifying key biological relationships automatically.

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

Pediatric Autism Diagnosis Accuracy and Confidence: A Comparison of Experienced and Inexperienced Clinicians Making Decisions with and without AI Decision Support.

Research square·2026
Same author

Enhancing Text Datasets With Scaling and Targeting Data Augmentation to Improve BERT-Based Machine Learners.

Expert systems with applications·2026
Same author

Are LLM-generated plain language summaries truly understandable? A large-scale crowdsourced evaluation.

Journal of biomedical informatics·2026
Same author

Comparative Evaluation of Text and Audio Simplification: A Methodological Replication Study.

Communications of the Association for Information Systems·2026
Same author

Influence of Audio Speech Rate and Source Text Difficulty on Health Information Comprehension and Retention.

Proceedings of the ... Annual Hawaii International Conference on System Sciences. Annual Hawaii International Conference on System Sciences·2026
Same author

Deep learning for autism detection using clinical notes: A comparison of transfer learning for a transparent and black-box approach.

Artificial intelligence in medicine·2025

Area of Science:

  • Computational Biology
  • Bioinformatics
  • Natural Language Processing

Background:

  • Current biomedical natural language processing (NLP) primarily focuses on pre-specified entities (e.g., proteins) and relations (e.g., inhibition).
  • Existing methods often lack the ability to automatically discover and structure relationships from free text.
  • Negation and coordinating conjunctions are frequently overlooked in relation extraction.

Purpose of the Study:

  • To develop and evaluate a shallow parser for automatic extraction of relations between noun phrases in biomedical text.
  • To create a system capable of identifying generic relations beyond pre-specified entity-relation pairs.
  • To incorporate the recognition of coordinating conjunctions and negation into the relation extraction process.

Main Methods:

Related Experiment Videos

  • Utilized heuristics and a noun phrase chunker to identify entities of interest within the text.
  • Employed cascaded finite state automata, based on closed-class English words, to model and structure generic relations.
  • Integrated recognition of coordinating conjunctions and negation to enhance relation capture.

Main Results:

  • Evaluated on 26 cancer-related abstracts, the parser extracted 330 relations.
  • Achieved a precision of 90%, with 296 relations correctly identified.
  • The system averaged 11 correct relations per abstract, demonstrating significant utility.

Conclusions:

  • The developed shallow parser effectively extracts relations between noun phrases from biomedical text with high precision.
  • The system's ability to model generic relations and handle negation offers an advancement over traditional entity-relation extraction methods.
  • This approach shows promise for accelerating knowledge discovery in biomedical research by automating relation identification.