sfla based gene selection approach for improving cancer classification accuracy

In this paper, we propose a new gene selection algorithm based on shuffled frog leaping algorithm that is called sfla-fs. the proposed algorithm is used for improving cancer classification accuracy. most of the biological datasets such as cancer datasets have a large number of genes and few samples. however, most of these genes are not usable in some tasks for example in cancer classification. therefore, selection of the appropriate genes is important in bioinformatics and machine learning. the proposed method combines the advantage of wrapper and filter methods for gene subset selection. sfla-fs consists of two phases. in the first phase a filter method is used for gene ranking from high dimensional microarray data and in the second phase, sfla is applied to gene selection. the performance of sfla-fs evaluated for cancer classification using seven standard microarray cancer datasets. experimental results are compared with those of obtained from several existing well-known gene selection algorithm. the experimental results show that sfla-fs has a remarkable ability to generate reduced size of genes while yielding significant classification accuracy in cancer classification.