Web Page Classification Based on SVM

This paper studies several key aspects of support vector machine (SVM) for Web page classification. Developed from statistical learning theory, SVM is widely investigated and used for text categorization because of its high generalization performance and tolerant ability of processing high dimension classification. Firstly, some methods for Web page presentation are studied. Secondly the Web page classification based on SVM is implementation on data set, and NB classifier is used for study the performance of the SVM classifier processing high dimension space. Finally, the comparison on the polynomial kernel function and the radius basis function (RBF) kernel function is studied. It is proved that if a kernel has a perfect alignment with the classification task, the SVM classifier has better performances