Classification of gene expression autism data based on adaptive penalized logistic regression

Abstract

The common issues of high-dimensional gene expression data are that many of genes may not be relevant to their diseases. Gene selection has been proved to be an effective way to improve the result of many classification methods. In this paper, an adaptive penalized logistic regression is proposed, with the aim of identification relevant genes and provides high classification accuracy of autism data, by combining the logistic regression with the weighted L₁-norm. Experimental results show that the proposed method significantly outperforms two competitor methods in terms of classification accuracy, G-mean, and area under the curve. Thus, the proposed method can be useful for other cancer classification using DNA gene expression data in the real clinical practice.