Application of SVM in web page categorization

Abstract

Web page classification is an important research direction for web mining. This paper gives some methods for representation of web page, studies on several aspects of support vector machine (SVM) for structural risk minimization concepts of statistical learning theory. The web page classifier and algorithm based on SVM is proposed. Cross validation method is used to select parameters of SVM classifier. The experimental results show that SVM provides an effective method for web page categorization and has promising application in web mining.. . Index Terms—web page categorization, support vector machine, Web Page representation; Kernel function; Web Ming