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

Estimating optimal feature subsets using efficient estimation of high-dimensional mutual information.

Tommy W S Chow1, D Huang

  • 1City University of Hong Kong, Hong Kong. eetchow@cityu.edu.hk

IEEE Transactions on Neural Networks
|March 1, 2005
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 author

Search for Signatures of Dark Matter Annihilation in the Galactic Center with HAWC.

Physical review letters·2026
Same author

Network Pharmacological Prediction and Experimental Analyses Reveal That Naringin Alleviates Osteoarthritis Progression by Targeting MMP13.

Physiological research·2026
Same author

Novel Uses of Optical Coherence Tomography in Periodontology.

Journal of dental research·2026
Same author

B-Cell Landscape in Inflamed Dental Pulp Reveals Potential Autoimmune Involvement.

Journal of dental research·2026
Same author

Early Results of Acellular Dermal Matrix as an Interpositional Allograft Following Excision Arthroplasty of a Rheumatoid Forefoot: A Case Report.

Malaysian orthopaedic journal·2026
Same author

FDG PET/CT reveals solitary hepatic metastasis from urachal adenocarcinoma mimicking a hepatic pseudolesion on contrast-enhanced CT.

Revista espanola de medicina nuclear e imagen molecular·2025
Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013
Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011
Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011
Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011
Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011
Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011
See all related articles

This study introduces a new feature selection method using mutual information (MI) estimation. The approach effectively identifies key features for classification, even with high-dimensional data.

Area of Science:

  • Machine Learning
  • Data Science
  • Pattern Recognition

Background:

  • Feature selection is critical for effective classification.
  • Estimating high-dimensional mutual information (MI) is a significant challenge in MI-based feature selection.
  • Existing methods may lack efficiency or effectiveness in complex datasets.

Purpose of the Study:

  • To propose a novel and efficient method for estimating high-dimensional mutual information (MI).
  • To develop a robust feature selection technique based on the improved MI estimation.
  • To reliably identify optimal feature subsets for classification tasks.

Main Methods:

  • A pruned Parzen window estimator was combined with quadratic mutual information (QMI) for MI estimation.
  • A novel feature selection algorithm was developed, identifying salient features sequentially.

Related Experiment Videos

  • The methodology was validated across four diverse classification applications.

    • Main Results:

    • The proposed method effectively and efficiently estimates mutual information in high-dimensional settings.
    • The feature selection technique successfully identifies relevant features.
    • Appropriate feature subsets for classification were reliably determined.
    • Promising results were observed across datasets with varying feature numbers (10 to over 15,000).

    Conclusions:

    • The combined pruned Parzen window estimator and QMI offer an effective solution for high-dimensional MI estimation.
    • The novel feature selection method demonstrates strong performance and reliability in classification.
    • This work contributes a valuable tool for data scientists and machine learning practitioners dealing with large feature sets.