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

Simultaneous Bayesian Clustering and Feature Selection Through Student's ${t}$ Mixtures Model.

Jianyong Sun, Aimin Zhou, Simeon Keates

    IEEE Transactions on Neural Networks and Learning Systems
    |April 1, 2017
    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

    Comparative study of triple intravenous chemotherapy versus dual chemotherapy combined with hepatic arterial infusion in patients with liver metastases from colorectal cancer.

    Frontiers in oncology·2026
    Same author

    Blood-Brain Barrier (BBB)-Penetrable Androgen Receptor (AR) Degrader as a Potential Therapeutic Agent for Glioblastoma.

    ACS pharmacology & translational science·2026
    Same author

    MicroRNA profiles in colostrum and transition milk of non-dairy goats fed with or without resveratrol.

    Scientific data·2026
    Same author

    Integrated transcriptomic and metabolomic analyses elucidate the molecular mechanisms underlying quality formation during Rosa acicularis fruit ripening.

    Food research international (Ottawa, Ont.)·2026
    Same author

    Lake Water Depth and Nutrient Gradients Mediate Aquatic Plant Distribution via the Strength of Trait Network Associations.

    Global change biology·2026
    Same author

    The Dianthus spiculifolius chlorophyll-binding protein DsSep2 can be used as a genetic resource to create 'golden leaf' plants.

    Journal of plant physiology·2026

    This study introduces a novel generative model for unsupervised feature selection, robust to outliers using Student's distributions. The model effectively performs clustering, feature selection, and outlier detection simultaneously.

    Area of Science:

    • Machine Learning
    • Statistical Modeling
    • Data Mining

    Background:

    • Unsupervised feature selection is crucial for dimensionality reduction in complex datasets.
    • Existing methods can be sensitive to outliers and lack simultaneous clustering and outlier detection capabilities.
    • Generative models offer a probabilistic framework for understanding data structure and feature relevance.

    Purpose of the Study:

    • To propose a generative model for feature selection in an unsupervised learning context.
    • To develop a model robust to outliers by employing a mixture of Student's distributions.
    • To enable simultaneous clustering, feature selection, and outlier detection.

    Main Methods:

    • A generative model assuming data sampled from a finite mixture of Student's distributions.

    Related Experiment Videos

  • Inclusion of latent random variables for feature salience estimation.
  • Inference performed using a tree-structured variational Bayes algorithm.
  • Full Bayesian treatment for automatic model selection.

    • Main Results:

    • The model accurately models datasets containing outliers.
    • Experimental results demonstrate superior performance compared to existing unsupervised Bayesian feature selection algorithms.
    • The algorithm successfully identified discriminating genes in leukemia gene expression data.

    Conclusions:

    • The developed generative model provides an effective approach for unsupervised feature selection, clustering, and outlier detection.
    • The model's robustness to outliers and its ability to handle real-world data, such as gene expression data, are significant advantages.
    • The Bayesian framework facilitates automatic model selection and enhances the reliability of the feature selection process.