Text Classifier Based on an Improved SVM Decision Tree

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 SVM-DT is proposed. Experimental results demonstrate the effectiveness and efficiency of the improved SVM decision tree. (C) 2012 Published by Elsevier B.V. Selection and/or peer review under responsibility of ICMPBE International Committee.