Gene Expression Classification: Decision Trees vs. SVMs

Xiaojing Yuan, Xiaohui Yuan, Fan Yang, Jing Peng, and Bill P. Buckles

In this article, we compare decision trees (DT) and support vector machines (SVM) in classifying gene expressions. With the explosion of genome research, tremendous amount of data have been made available and a deep insight study becomes demanding. Among various kinds of gene analysis approaches being developed, sequence based gene expression classification shows the importance due to its ability to identify existence of some specific gene pieces. In this article, we focus on two major categories of classification methods, namely decision trees and support vector machines. By comparing various versions of decision tree algorithms, SVMs, and a particular SVM that integrates structural information of the gene sequence, it is shown that the structural information does help in achieving better performance with respect to the classification accuracy.


This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.