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

A comparative study of feature-salience ranking techniques.

W Wang, P Jones, D Partridge

    Neural Computation
    |July 7, 2001
    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

    A patient sink tap facilitating carbapenemase-producing enterobacterales transmission.

    The Journal of hospital infection·2019
    Same author

    Food additives: Assessing the impact of exposure to permitted emulsifiers on bowel and metabolic health - introducing the FADiets study.

    Nutrition bulletin·2019
    Same author

    Seeing a cause for blindness in severe pancreatitis.

    Gut·2015
    Same author

    Reliability and validity of a psycho-social problem inventory for childhood epilepsy.

    The Nigerian postgraduate medical journal·2001
    Same author

    Interaction of stilbene disulphonates with cloned K(ATP) channels.

    British journal of pharmacology·2001
    Same author

    Democratic prioritization of maternity care: a rational basis for planning a clinical effectiveness programme.

    Journal of public health medicine·2001

    This study compares methods for identifying important features for predicting outcomes. Simple techniques like neural network weight clamping and decision tree feature ranking effectively identify key features in complex data.

    Area of Science:

    • Artificial Intelligence
    • Machine Learning
    • Computational Neuroscience

    Background:

    • Determining feature salience is crucial for accurate prediction and optimal system design.
    • Existing methods include neural network approaches and statistical techniques.
    • Identifying salient features is challenging in noisy, real-world datasets.

    Purpose of the Study:

    • To evaluate and compare various techniques for assessing feature set salience.
    • To identify efficient and effective methods for selecting high-salience feature subsets.
    • To provide a basis for designing optimal computational systems.

    Main Methods:

    • Assessed neural-network-based techniques (e.g., weight clamping).
    • Included a standard statistical technique.

    Related Experiment Videos

  • Introduced a technique based on inductively generated decision trees.
  • Evaluated feature ranking using decision trees.

    • Main Results:

    • Weight clamping (neural network) and decision tree feature ranking provided consistent feature ordering.
    • Linear correlation was also found to be effective.
    • These simple approaches performed well in identifying salient features.

    Conclusions:

    • Neural network weight clamping and decision tree feature ranking are effective for determining feature salience.
    • These methods offer efficient and reliable ways to identify important features in complex data.
    • The findings support the development of more optimal computational systems.