Research on SVM-Based Automatic Classification of Chinese Web Page

This paper deals with Chinese web page classification based on support vector machine (SVM). Some methods are proposed for text extraction and Chinese word segment. And it discusses the different contribution to text classification on different locations of the page. The SVM classifier is applied on classification. The results showed that the performance of the classification has further improved, for the text without noisy blocks after extraction, high correct rate of Chinese word segment. In addition, picking the title, keywords and description out, and increasing its weighs can also improve the accuracy of classification.