A novel feature selection for gene expression data

Abstract

The feature selection process can be considered a problem of global combinatorial optimization in machine learning, which reduces the number of features, removes irrelevant, noisy and redundant data, and results in an acceptable classification accuracy. Therefore, a good feature selection method based on the number of features investigated for sample classification is needed in order to speed up the processing rate, predictive accuracy, and to avoid incomprehensibility. In this paper, particle swarm optimization (PSO) is used to implement a feature selection, and the K-nearest neighbor (K-NN) method with leave-one-out cross-validation (LOOCV) serves as an evaluator of PSO. The support vector machines (SVMs) with the one-versus-rest method serve as a classifier for the classification problem. Experimental results show that our method simplifies features effectively and obtains a higher classification accuracy compared to the other classification methods from the literature.