The Relevance Sample-Feature Machine: A Sparse Bayesian Learning Approach to Joint Feature-Sample Selection

    Authors

    Y. Mohsenzadeh; H. Sheikhzadeh; A. M. Reza; N. Bathaee;M. M. Kalayeh

    Comments

    Authors: contact us about adding a copy of your work at STARS@ucf.edu

    Abbreviated Journal Title

    IEEE T. Cybern.

    Keywords

    Joint feature selection and classifier design; relevance features; relevance sample feature machine (RSFM); relevance samples; sparse; Bayesian learning; VECTOR MACHINES; GAUSSIAN-PROCESSES; CLASSIFICATION; REGRESSION; TRACKING; Computer Science, Artificial Intelligence; Computer Science, Cybernetics

    Abstract

    This paper introduces a novel sparse Bayesian machine-learning algorithm for embedded feature selection in classification tasks. Our proposed algorithm, called the relevance sample feature machine (RSFM), is able to simultaneously choose the relevance samples and also the relevance features for regression or classification problems. We propose a separable model in feature and sample domains. Adopting a Bayesian approach and using Gaussian priors, the learned model by RSFM is sparse in both sample and feature domains. The proposed algorithm is an extension of the standard RVM algorithm, which only opts for sparsity in the sample domain. Experimental comparisons on synthetic as well as benchmark data sets show that RSFM is successful in both feature selection (eliminating the irrelevant features) and accurate classification. The main advantages of our proposed algorithm are: less system complexity, better generalization and avoiding overfitting, and less computational cost during the testing stage.

    Journal Title

    Ieee Transactions on Cybernetics

    Volume

    43

    Issue/Number

    6

    Publication Date

    1-1-2013

    Document Type

    Article

    Language

    English

    First Page

    2241

    Last Page

    2254

    WOS Identifier

    WOS:000327647500060

    ISSN

    2168-2267

    Share

    COinS