Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy

Abstract

Feature selection is an important problem for pattern classification systems. We study how to select good features according to the maximal statistical dependency criterion based on mutual information. Because of the difficulty in directly implementing the maximal dependency condition, we first derive an equivalent form, called minimal-redundancy-maximal-relevance criterion (mRMR), for first-order incremental feature selection. Then, we present a two-stage feature selection algorithm by combining mRMR and other more sophisticated feature selectors (e.g., wrappers). This allows us to select a compact set of superior features at very low cost. We perform extensive experimental comparison of our algorithm and other methods using three different classifiers (naive Bayes, support vector machine, and linear discriminate analysis) and four different data sets (handwritten digits, arrhythmia, NCI cancer cell lines, and lymphoma tissues). The results confirm that mRMR leads to promising improvement on feature selection and classification accuracy.

Cite this paper

Related papers

1. Regularization and Variable Selection Via the Elastic Net2005
2. Metagenomic biomarker discovery and explanation2011
3. MEME SUITE: tools for motif discovery and searching2009
4. The STRING database in 2023: protein–protein association networks and functional enrichment analyses for any sequenced genome of interest2022
5. An “Electronic Fluorescent Pictograph” Browser for Exploring and Analyzing Large-Scale Biological Data Sets2007