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

Recursive Feature Elimination by Sensitivity Testing.

Nicholas Sean Escanilla1, Lisa Hellerstein2, Ross Kleiman1

  • 1Department of Computer Sciences, University of Wisconsin-Madison, Madison, Wisconsin.

Proceedings of the ... International Conference on Machine Learning and Applications. International Conference on Machine Learning and Applications
|December 5, 2019
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

Rocuronium Dose and First-Attempt Intubation Success in the Critically Ill: Secondary Analysis of Two Multicenter Trials.

American journal of respiratory and critical care medicine·2026
Same author

United Global Advocacy Drives Updates to World Health Organization Essential Medicines List.

Haemophilia : the official journal of the World Federation of Hemophilia·2026
Same author

Impact of measurable residual disease on outcomes using a modified DFCI protocol for adults with BCR-ABL negative acute lymphoblastic leukemia.

Leukemia research·2026
Same author

Stratifying Risk and Treatment Benefit: A Model Predicting Overall Survival in Men with Metastatic De Novo Hormone-sensitive Prostate Cancer in Trials Investigating Docetaxel (the STOPCAP Collaboration).

European urology focus·2026
Same author

Proposed Modifications to Prognostic Classification of AML Patients Treated With Intensive Chemotherapy Based on Recent Real-World Data.

Clinical lymphoma, myeloma & leukemia·2025
Same author

Development and implementation of an entity relationship diagram for perinatal data.

JAMIA open·2025
Same journal

Diffusion Policies with Offline and Inverse Reinforcement Learning for Promoting Physical Activity in Older Adults Using Wearable Sensors.

Proceedings of the ... International Conference on Machine Learning and Applications. International Conference on Machine Learning and Applications·2026
Same journal

Automated Shared Phenotype Discovery in Undiagnosed Cohorts for Rare Disease Research.

Proceedings of the ... International Conference on Machine Learning and Applications. International Conference on Machine Learning and Applications·2025
Same journal

Clinical report classification using Natural Language Processing and Topic Modeling.

Proceedings of the ... International Conference on Machine Learning and Applications. International Conference on Machine Learning and Applications·2023
Same journal

Multiple Imputation via Generative Adversarial Network for High-dimensional Blockwise Missing Value Problems.

Proceedings of the ... International Conference on Machine Learning and Applications. International Conference on Machine Learning and Applications·2022
Same journal

A clustering-based biased Monte Carlo approach to protein titration curve prediction.

Proceedings of the ... International Conference on Machine Learning and Applications. International Conference on Machine Learning and Applications·2021
Same journal

AUTOMATIC SCORING OF A NONWORD REPETITION TEST.

Proceedings of the ... International Conference on Machine Learning and Applications. International Conference on Machine Learning and Applications·2020
See all related articles

This study introduces a novel recursive feature elimination (RFE) method for understanding complex non-linear models. The enhanced RFE technique offers improved feature relevance insights, particularly beneficial in fields like bioinformatics.

Area of Science:

  • Machine Learning
  • Bioinformatics
  • Computational Learning Theory

Background:

  • Interpreting non-linear models is challenging.
  • Feature relevance ranking is a key but difficult task for non-linear models.
  • Existing methods for feature selection often struggle with model complexity.

Purpose of the Study:

  • To introduce a novel version of recursive feature elimination (RFE) for enhanced insight into trained non-linear models.
  • To adapt sensitivity testing methods from computational learning theory for practical use with trained models.
  • To evaluate the efficacy of the proposed RFE method on both synthetic and real-world bioinformatics data.

Main Methods:

  • A novel recursive feature elimination (RFE) approach is proposed.
  • The method adapts sensitivity testing using queries to a trained non-linear model, mimicking membership queries.

Related Experiment Videos

  • Empirical validation is performed on a cancer genomics dataset and synthetic data with known ground truth.

    • Main Results:

    • The novel RFE method demonstrates theoretical and empirical benefits for feature relevance ranking in non-linear models.
    • The approach provides potential insights into complex biological data, such as cancer genomics.
    • Performance is validated on synthetic datasets where ground truth is known, confirming its utility.

    Conclusions:

    • The proposed RFE method offers a valuable tool for improving interpretability of non-linear models.
    • This technique has significant potential applications in bioinformatics and other data-intensive scientific fields.
    • The approach effectively addresses the challenge of identifying relevant features in complex machine learning models.