基于向量空间模型的文本分类系统的研究与实现  

Research and Implementation of Text Classification System Based on VSP

作　　者：陈治纲[1] 何丕廉[1] 孙越恒[1] 郑小慎[1] 

机构地区：[1]天津大学电子信息工程学院,天津300072

出　　处：《中文信息学报》2005年第1期36-41,共6页Journal of Chinese Information Processing

基　　金：天津市科技发展计划项目 (0 2 310 0 5 11)

摘　　要：文本分类是信息处理的一个重要的研究课题 ,它可以有效的解决信息杂乱的现象并有助于定位所需的信息。本文综合考虑了频度、分散度和集中度等几项测试指标 ,提出了一种新的特征抽取算法 ,克服了传统的从单一或片面的测试指标进行特征抽取所造成的特征“过度拟合”问题 ,并基于此实现了二级分类模式的文本分类系统。和类中心分类法相比 ,实验结果表明二级分类模式具有较高的精度和召回率。Text classification is an important research task of natural language processing, which can efficiently resolve the issue of information chaos and help to locate the required information. The traditional approaches of text classification commonly extract feature terms from a single test criterion, which will lead to the problem of “over fitting'. This paper comprehensively takes test criterions such as frequency, distribution and concentration into account and proposes a new arithmetic of feature extraction and implements text classification system with two-level mode. The experimental results show that two-level classification mode has higher classification precision and recall compared with center classification method.

关 键 词：计算机应用 中文信息处理 文本分类 测试指标 特征抽取 二级分类模式 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]