A Improved Feature Weighting Algorithm for Chinese Text Classification

摘要：

Based on analyzing traditional feature weighting algorithm and according to the problem that the traditional feature weighting algorithm only considers term frequency and inverse document frequency, a improved weighting algorithm is put forward, that is, synthesizing the length.position and class information of term when computing its weight. Then verify effectiveness of the classification by KNN classifier. Experimental results show that this algorithm can improve the classification accuracy and achieve great results of classification.

关键词： feature text classification feature weighting KNN classification algorithm

作者: Dong Hui Yin Siqing

作者单位: North university of China School of electronics and computer science and technology Tai Yuan, China North university of China Software school of north university of China Tai Yuan, China

会议类型: 国际会议

会议名称: The 2010 International Conference on Computer Application and System Modeling(2010计算机应用与系统建模国际会议 ICCASM 2010)

会议地点: 太原

会议语种:英文

页码: 433-436

在线出版日期: 2010-10-22（万方平台首次上网日期，不代表论文的发表时间）