Gene subset selection for lung cancer classification using a multi-objective strategy

Abstract

A microarray machine offers the ability to measure the expression levels of thousands of genes simultaneously. It is used to collect the infonnation from tissue and cell samples regarding gene expression differences that could be useful for cancer classification. However, the urgent problems in the use of gene expression data are the availability of a huge number of genes relative to the small number of available samples, and many of the genes are not relevant to the classification. It has been shown that selecting a small subset of genes can lead to improved classification accuracy. Hence, this paper proposes a solution to the problems by using a multi-objective strategy in genetic algorithms. This approach is experimented on one gene expression data set, namely the lung cancer. It obtains encouraging result on the data set as compared with an approach that uses single-objective strategy in genetic algorithms.