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

Feature selection in MLPs and SVMs based on maximum output information.

Vikas Sindhwani1, Subrata Rakshit, Dipti Deodhare

  • 1Department of Computer Science, the University of Chicago, IL 60637, USA.

IEEE Transactions on Neural Networks
|October 6, 2004
PubMed
Summary

This study introduces Maximum Output Information (MOI) algorithms for effective feature selection in Multilayer Perceptrons (MLPs) and Support Vector Machines (SVMs). These methods efficiently identify optimal feature subsets for improved classifier performance.

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

Human-AI Cooperation in Healthcare and Rehabilitation.

Delaware journal of public health·2026
Same author

Temporal point process modeling of aggressive behavior onset in psychiatric inpatient youths with autism.

Scientific reports·2026
Same author

Training sparse convolutional deep predictive coding networks with attention.

Neural networks : the official journal of the International Neural Network Society·2026
Same author

Ultrathin boron nanoribbons: stability, conductivity, and edge magnetism.

Nanotechnology·2026
Same author

Corticomorphic Hybrid CNN-SNN Architecture for EEG-Based Low-Footprint Low-Latency Auditory Attention Detection.

Annals of biomedical engineering·2026
Same author

Deep Learning-Based Prediction of Cardiopulmonary Disease in Retinal Images of Premature Infants.

JAMA ophthalmology·2026

Area of Science:

  • Machine Learning
  • Data Science
  • Computational Intelligence

Background:

  • Feature selection is crucial for optimizing classifier performance and reducing computational complexity.
  • Existing methods may struggle with computational cost or assumptions about data distributions.
  • Multilayer Perceptrons (MLPs) and multiclass Support Vector Machines (SVMs) are widely used classification models.

Purpose of the Study:

  • To develop novel feature selection algorithms for MLPs and SVMs.
  • To utilize mutual information between class labels and classifier outputs as an objective function.
  • To enhance classifier performance and efficiency through optimized feature subsets.

Main Methods:

  • Development of Maximum Output Information (MOI) algorithms for feature selection.

Related Experiment Videos

  • Employing mutual information on discrete variables as an objective function.
  • Utilizing greedy elimination and directed search strategies for subset selection.

    • Main Results:

    • MOI algorithms provide an efficient and robust objective function for feature selection.
    • The algorithms generate a user-defined feature subset and a trained classifier.
    • Demonstrated favorable performance compared to other methods on diverse datasets.

    Conclusions:

    • The proposed MOI algorithms offer a computationally inexpensive and effective approach to feature selection.
    • These algorithms are robust to prior class probabilities and provide error bounds.
    • The methods are applicable to both artificial and real-world datasets for MLPs and SVMs.