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

DTranNER: biomedical named entity recognition with deep learning-based label-label transition model.

S K Hong1, Jae-Gil Lee2,3

  • 1Graduate School of Knowledge Service Engineering, KAIST, 291 Daehak-ro, Yuseong-gu, Daejeon, 34141, South Korea.

BMC Bioinformatics
|February 13, 2020
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 journal

Covariance decomposition for distance based species tree estimation.

BMC bioinformatics·2026
Same journal

SNPio: a Python interface for population genomic data processing.

BMC bioinformatics·2026
Same journal

SpaHNR: a spatial domain identification method via sparse attention-based hierarchical node representation and multi-view contrastive learning.

BMC bioinformatics·2026
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
See all related articles

This study introduces DTranNER, a novel framework that enhances biomedical named-entity recognition (BioNER) by dynamically modeling label transitions. DTranNER improves accuracy by using deep learning to adaptively capture contextual relationships between labels in biomedical texts.

Area of Science:

  • Computational biology
  • Bioinformatics
  • Natural Language Processing

Background:

  • Biomedical Named-Entity Recognition (BioNER) is crucial for information extraction from biomedical literature.
  • Traditional Conditional Random Field (CRF) models treat BioNER as a sequence labeling problem, but static label transitions limit performance.
  • Deep learning models combined with CRF have improved BioNER, yet struggle to dynamically adapt label transitions to context.

Purpose of the Study:

  • To introduce DTranNER, a novel CRF-based framework that incorporates a deep learning-based label-label transition model for BioNER.
  • To address the limitations of static transition models in capturing contextual information for accurate entity segmentation.

Main Methods:

  • DTranNER utilizes two distinct deep learning networks: a Unary-Network for individual label prediction and a Pairwise-Network for modeling label-label transitions.
Keywords:
BioinformaticsData miningNamed entity recognitionNeural network

Related Experiment Videos

  • The Pairwise-Network dynamically explores contextual relations between adjacent labels based on input sentence context.

    • Main Results:

    • DTranNER achieved state-of-the-art F1-scores on multiple benchmark BioNER corpora, including BC2GM (84.56%), BC4CHEMD (91.99%), chemical NER (94.16%), and BC5CDR disease NER (87.22%).
    • The framework demonstrated superior performance compared to existing state-of-the-art methods across various biomedical entity recognition tasks.
    • A near-best F1-score of 88.62% was achieved on the NCBI-Disease corpus.

    Conclusions:

    • The integration of a deep learning-based dynamic label-label transition model significantly enhances BioNER performance over static models.
    • DTranNER adaptively captures contextual label relationships, offering a fine-grained approach to improving entity recognition.
    • This framework is expected to advance biomedical literature mining and related research areas.