【作者】 陈庆轩； 郑德权； 郑博文； 赵铁军； 李生；

【Author】 CHEN Qing-xuan,ZHENG De-quan,ZHEN Bo-wen,ZHAO Tie-jun,LI Sheng (MOE-MS Key Laboratory of Natural Language Processing and Speech,Harbin Institute of Technology,Harbin 150001,China)

【机构】 哈尔滨工业大学教育部微软语言语音重点实验室；

【摘要】 随着互联网信息及电子资源的急剧膨胀,文本分类技术成为信息组织与管理的有效手段。基于Fisher判别分析理论,通过分析词条类间分布的不均匀性与类内分布的相对均匀性,提出了基于文档频度分布的文本特征选择方法,应用于文本分类中,实验表明,该方法能够充分地表达文本的特征信息,获得了良好的效果。更多还原

【Abstract】 With the development of internet and the rapid expansion of electronic resource,text classification technology has became an effective organization and management tool to deal with information.Based on Fisher discriminant analysis theory,a text feature selection method based on document frequency distribution was presented in this paper,which analyzed the uneven term distribution among categories and the even term distribution within categories. Experiments showed that this method improved the performance of text classification.更多还原