Text Classifier Based on an Improved SVM Decision Tree
Zhenqiang Xu, Pengwei Li, Yunxia Wang · Physics Procedia · 2012
The classifier of SVM decision tree (SVM-DT) takes advantage of both the efficient computation of the tree architecture and the high classification accuracy of SVMs. The paper proposes a new effective approach to optimize the SVM -DT classifier while presents the research on text categorization using SVM-DT classifier. In this approach, a novel separability measure is defined base on Support vector domain description (SVDD), and an improved SVMDT is proposed. Experimental results demonstrate the effectiveness and efficiency of the improved SVM decision tree.