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

Input feature selection for classification problems.

N Kwak1, Chong-Ho Choi

  • 1Sch. of Electr. Eng., Seoul Nat. Univ.

IEEE Transactions on Neural Networks
|February 5, 2008
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

Temporal trends in treatment strategies and outcomes of newly diagnosed <i>Mycobacterium avium</i> complex pulmonary disease.

The international journal of tuberculosis and lung disease : the official journal of the International Union against Tuberculosis and Lung Disease·2026
Same author

Clinical standards for antimicrobial stewardship in TB care.

IJTLD open·2025
Same author

Lung function decline in non-tuberculous mycobacterial pulmonary disease according to disease severity.

The international journal of tuberculosis and lung disease : the official journal of the International Union against Tuberculosis and Lung Disease·2023
Same author

Multidrug-resistant tuberculosis in South Korea: a retrospective analysis of national registry data in 2011-2015.

The international journal of tuberculosis and lung disease : the official journal of the International Union against Tuberculosis and Lung Disease·2019
Same author

Multidrug-resistant tuberculosis over 20 years at a referral hospital in South Korea: trends and outcomes.

The international journal of tuberculosis and lung disease : the official journal of the International Union against Tuberculosis and Lung Disease·2019
Same author

Changes in treatment outcomes of multidrug-resistant tuberculosis.

The international journal of tuberculosis and lung disease : the official journal of the International Union against Tuberculosis and Lung Disease·2015
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

Two new feature selection algorithms are proposed to improve classification performance in large datasets. One algorithm enhances mutual information, while the other uses the Taguchi method, offering efficient attribute reduction for neural networks (NNs).

Area of Science:

  • Machine Learning
  • Data Science
  • Artificial Intelligence

Background:

  • Feature selection is crucial for efficient classification systems like neural networks (NNs).
  • Managing large datasets necessitates reducing attributes to enhance performance and lower computational load.
  • Irrelevant or redundant attributes can hinder classification accuracy and efficiency.

Purpose of the Study:

  • To propose two novel feature selection algorithms for classification tasks.
  • To address limitations of existing methods like Mutual Information Feature Selector (MIFS).
  • To develop efficient methods for identifying relevant features with minimal experimental effort.

Main Methods:

  • A modified mutual information-based feature selection algorithm is presented.

Related Experiment Videos

  • A novel feature selection algorithm employing the Taguchi method is introduced.
  • The proposed algorithms are evaluated on several classification problems.

    • Main Results:

    • The enhanced mutual information algorithm achieves performance comparable to ideal greedy selection under uniform information distribution.
    • The Taguchi method-based algorithm efficiently identifies key features with minimal experiments.
    • The combined algorithms demonstrate robust performance, complementing each other's strengths.

    Conclusions:

    • The proposed feature selection algorithms offer significant improvements for classification tasks.
    • These methods provide effective solutions for reducing computational effort and enhancing performance in large datasets.
    • The combined approach is a valuable tool for feature selection in classification problems.