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

Class decomposition for GA-based classifier agents--a Pitt approach.

Sheng-Uei Guan1, Fangming Zhu

  • 1Department of Electrical and Computer Engineering, National University of Singapore, Singapore. eleguans@nus.edu.sg

IEEE Transactions on Systems, Man, and Cybernetics. Part B, Cybernetics : a Publication of the IEEE Systems, Man, and Cybernetics Society
|September 17, 2004
PubMed
Summary

This study introduces class decomposition to enhance Genetic Algorithm (GA) classifiers. This method improves classification accuracy while reducing training time for complex datasets.

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

Bidirectional regulation by BACH2 coordinates systemic CXCR5- vs. CXCR5+ CD4+ T cell differentiation.

The Journal of experimental medicine·2026
Same author

Portable multichannel immunoassay of coxsackievirus A6 using a coordination-engineered iridium-doped ZIF-8 nanozyme.

The Analyst·2026
Same author

Mechanisms of superior respiratory IgA responses against SARS-CoV-2 after mucosal vaccination.

bioRxiv : the preprint server for biology·2025
Same author

Syndiospecific polymerization of styrene with half-titanocene catalysts containing fluorinated phenoxy ligands.

RSC advances·2025
Same author

A facile method to fabricate syndiotactic polystyrene aerogels by freeze-drying.

Soft matter·2025
Same author

Synthesis and Characterization of Ultralight Syndiotactic Polystyrene Aerogels Prepared by Freeze-Drying.

Langmuir : the ACS journal of surfaces and colloids·2025

Area of Science:

  • Machine Learning
  • Artificial Intelligence
  • Computational Intelligence

Background:

  • Genetic Algorithms (GAs) are widely used for classification tasks.
  • Improving the efficiency and accuracy of GA-based classifiers remains an active research area.
  • Complex classification problems can pose challenges for standard GA approaches.

Purpose of the Study:

  • To propose and evaluate a novel class decomposition approach for Genetic Algorithm (GA)-based classifier agents.
  • To enhance the performance, specifically classification rate and training time, of GA classifiers.
  • To investigate the effectiveness of partitioning classification problems into smaller, manageable modules.

Main Methods:

  • A class decomposition strategy is employed, dividing the classification problem into distinct class modules.

Related Experiment Videos

  • Each module is trained independently and in parallel, focusing on a subset of the original problem.
  • Results from individual modules are integrated to form a final classification decision, with conflict resolution mechanisms.

    • Main Results:

    • Experimental evaluation on benchmark classification datasets demonstrates the efficacy of the proposed approach.
    • The class decomposition method leads to a significant improvement in the classification rate.
    • A notable reduction in the overall training time is achieved compared to traditional methods.

    Conclusions:

    • Class decomposition is an effective strategy for enhancing the performance of GA-based classifiers.
    • This approach offers a scalable solution for tackling complex classification problems.
    • The parallel and independent training of modules contributes to improved efficiency and accuracy.