A method of Chinese text categorization based on proximal support vector machine

摘要

A Chinese text categorization method based on proximal support vector machine and similarity of words is studied in the paper. Firstly feature vectors are extracted, and then the text feature subset based on similarity of words is obtained, finally the text is categorized based on proximal support vector machine. The tests on the large-scale text show that the recall is comparatively low and the precision is comparatively high.