An improving gene selection for microarray data

The microarray data consists of tens of thousands of genes on a genomic scale. To avoid higher computational complexity, it needs gene selection to find the gene subsets that are able to explain the disease. In this paper, an improving gene selection for microarray data is proposed. In the proposed algorithm, scatter search is used to obtain suitable parameter settings for support vector machine and then a subset of beneficial genes is selected. These selected genes can increase the accuracy of classification for microarray data. From experimental results, it shows that the proposed algorithm can obtain a better parameter setting and reduce unnecessary genes.