用于文本分类和文本聚类的特征抽取方法的研究  

The study of feature extraction method in text classification and text clustering

作　　者：孟春艳[1,2] 

机构地区：[1]山东大学计算机科学与技术学院,山东济南250061 [2]山东电子职业技术学院计算机系,山东济南250014

出　　处：《微计算机信息》2009年第9期149-150,共2页Control & Automation

摘　　要：文本信息处理已成为一门日趋成熟、应用面日趋广泛的学科。文本分类和聚类技术是应信息检索和查询需要而出现的自然语言处理领域的重要研究课题。面对急速膨胀的各种文本信息,通过使用文本分类和聚类技术,人们能对这些信息进行高效地组织和整理,以便于实现信息的准确定位和分流,从而提高用户查询和检索的效率。本文针对文本信息处理中最重要的研究方向——文本分类和聚类技术展开了研究,分析了特征抽取法在文本分类和文本聚类中应用的重要性,以及论证了为何要对文本进行特征抽取,最后分别阐述了用于文本分类和文本聚类的特征抽取方法。Text information processing has become an increasingly mature discipline with increasingly wide range of applications. Text classification and clustering technology is an important research topic for the needs of information retrieval and query in the natural language processing areas. With the rapid expansion of the various text messages, through the use of text classification and clustering technology, people can so efficiently organize and manage the information that people can achieve an accurate positioning information and diversion, and thereby enhancing the user＇s query and retrieval efficiency. In this paper, information processing in the text of the most important research direction-text classification and clustering technology start of the study. Analysis the importance of the feature extraction method in the text classification and clustering, and demonstrated why have feature extraction on the text, respectively, for the final text on how the feature extraction method applied in text classification and clustering.

关 键 词：文本分类 文本聚类 特征抽取 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]