一种基于支持向量机的专业中文网页分类器

文中提出了一种基于支持向量机的专业中文网页分类算法，利用支持向量机对网页进行二类分类，找出所需专业的中文网页；然后利用向量空间模型，对分类好的专业网页进行多类分类。在构造支持向量机的过程中，为了提高分类的召回率，采用了一种偏移因子。该算法只需要计算二类SVM分类器，实验表明，它不仅具有较高的训练效率，同时能得到很高的分类精确率和召回率。

A Topic-specific Chinese Web Page Classifier Based on Support Vector Machine

This paper presents a topic-specific Chinese Web page classification algorithm based on support vector machine(SVM). The algorithm classifies Chinese Web pages by 2-kind SVM and finds out the Web pages in the topic-specific class. Then the topic-specific Web pages are classified in several child classes of the topic-specific class with vector space model(VSM). To improve the recall,an excursion factor is used in the procedure in which SVM is constructed. The algorithm only computes 2-kind SVM classifier,and experiment shows that it not only improves training efficiency,but also has good precision and recall.