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

Resolving "orphaned" non-specific structures using machine learning and natural language processing methods.

Dongfang Xu1, Steven S Chong1,2, Thomas Rodenhausen1

  • 1University of Arizona, Tucson, United States of America University of Arizona Tucson United States of America.

Biodiversity Data Journal
|November 6, 2018
PubMed
Summary

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

Recombinant human IGF-1 alleviates dyslipidemia induced by lactational maternal dietary restriction.

Frontiers in pediatrics·2026
Same author

Subplate neurons: potential targets for dysfunction after hypoxic-ischemic brain injury.

Experimental neurology·2026
Same author

PSD3 links autophagic flux to MHC-I-associated immune modulation in esophageal squamous cell carcinoma.

Apoptosis : an international journal on programmed cell death·2026
Same author

Integrated learning-assisted design of metal-nitrogen-carbon single-atom catalysts: electronegativity regulates the coupling rules of interfacial valence electrons.

Chemical science·2026
Same author

iNOS deficiency aggravates DNFB-induced atopic dermatitis in mice.

International immunopharmacology·2026
Same author

Active Components and Potential Mechanism of Redujing Oral Liquid for Treating EBV Infection: A UHPLC-MS and Network Pharmacology- Based Approach.

Combinatorial chemistry & high throughput screening·2026
Same journal

A scalable exemplar-based method for aligning biological taxonomies.

Biodiversity data journal·2026
Same journal

First record of <i>Onchidium reevesii</i> from Korean tidal flats (Mollusca, Gastropoda, Onchidiidae).

Biodiversity data journal·2026
Same journal

Camera trapping survey of vertebrates in Lipis Geopark, Pahang, Malaysia with notes on cave utilisation by mainland clouded leopard (<i>Neofelis nebulosa</i>).

Biodiversity data journal·2026
Same journal

A seasonal bird occurrence dataset from Karataş Lake (Burdur, Turkey) with associated NDVI and climate data, 2021 to 2022.

Biodiversity data journal·2026
Same journal

A new record of the southern erebid moth species, <i>Hypena nakajimai</i> Kishida, 2010 (Lepidoptera, Erebidae), from Korea with molecular analyses.

Biodiversity data journal·2026
Same journal

Population-level transcriptomic datasets from two benthic invertebrates exposed to long-term experimental warming and acidification.

Biodiversity data journal·2026
See all related articles
This summary is machine-generated.

Syntactic methods achieved 92.1% F1 score in resolving part-of relations in biodiversity literature, outperforming SVM methods. Combining approaches may further reduce errors in computational biology research.

Area of Science:

  • Computational Biology
  • Bioinformatics
  • Biodiversity Informatics

Background:

  • Scholarly biodiversity literature contains rich, human-readable morphological descriptions.
  • Extracting this information computationally is challenging due to linguistic complexities.
  • Understanding anatomical part-of relationships is crucial for biodiversity analysis.

Purpose of the Study:

  • To develop and compare machine-based methods for resolving meronym (part-of) relations in biodiversity literature.
  • To evaluate the effectiveness of syntactic and machine learning approaches using domain ontologies.

Main Methods:

  • A syntactic rule-based approach was employed.
  • A Support Vector Machine (SVM)-based method was utilized.
  • Both methods leveraged domain ontologies for knowledge integration.
Keywords:
Anaphora ResolutionBiodiversity LiteratureInformation ExtractionMachine LearningMorphological DescriptionsOntology ApplicationPerformance Evaluation

Related Experiment Videos

Main Results:

  • The syntactic rule-based method achieved a 92.1% F1 score.
  • The SVM-based method achieved an 80.7% F1 score.
  • Part-of ontologies proved valuable knowledge sources, and method errors showed little overlap.

Conclusions:

  • Syntactic methods demonstrate superior performance for extracting part-of relations compared to SVM methods in this domain.
  • The complementary nature of different approaches offers potential for further error reduction.
  • Accurate extraction is vital, as even small error rates can lead to significant inaccuracies in large-scale biodiversity data analysis.