A survey on hybrid feature selection methods in microarray gene expression data for cancer classification

Abstract

The emergence of DNA Microarray technology has enabled researchers to analyze the expression level of thousands of genes simultaneously. The Microarray data analysis is the process of finding the most informative genes as well as remove redundant and irrelevant genes. One of the most important applications of the Microarray data analysis is cancer classification. However, the curse of dimensionality and the curse of sparsity make classifying gene expression profiles a challenging task. One of the most effective methods to overcome these challenges is feature (gene) selection. In this paper, we aim to review and compare the most recent hybrid approaches that employ bio-inspired evolutionary methods as the wrapper method.