论文本分类中特征选择方法  

A Survey on the Method of Feature Selection in Text Categorization

作　　者：张小艳[1] 宋丽平[1] 

机构地区：[1]西安科技大学计算机系,陕西西安710054

出　　处：《现代情报》2009年第3期131-133,138,共4页Journal of Modern Information

基　　金：陕西省教育厅专项课题(06JK248)资助

摘　　要：文本分类技术在信息过滤和信息检索中有着重要应用。文本表示技术是文本分类中的首要任务,特征选择技术又是文本表示中的核心技术,对分类效果起着至关重要的作用。本文介绍了文本表示和特征选择技术的发展,并在详细分析目前各种文本表示和特征选择的方法和技术特点基础上,比较了各种方法的适用性和优缺点,最后总结出了文本表示和特征选择技术研究的方向和目标。The text categorization technology has an important application in the information filtration and retrieval. The text representation technology is the firstline assignment in the text categorization. What＇s more, the feature selection technology is the key in text representation. Both of them play an important role in classification effect. This thesis introduced the development of the text representation and feature selection technology, compared the applicability, advantages and disadvantages of various methods, which is on the basis of seriously analyzing the methods, technological characteristics of text representation, and feature selection at present. Eventually, it summarized the researching direction and target of text representation, as well as feature selection technology.

关 键 词：文本分类 文本表示 特征选择 语义特征 

分 类 号：G20[文化科学—传播学]