作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2008, Vol. 34 ›› Issue (20): 13-14. doi: 10.3969/j.issn.1000-3428.2008.20.005

• 博士论文 • 上一篇    下一篇

基于文章标题信息的汉语自动文本分类

缪建明1,张 全1,赵金仿2   

  1. (1. 中国科学院声学研究所,北京 100080;2. 江苏科技大学电子信息学院,镇江 212003)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2008-10-20 发布日期:2008-10-20

Chinese Automatic Text Categorization Based on Article Title Information

MIAO Jian-ming1, ZHANG Quan1, ZHAO Jin-fang2   

  1. (1. Institute of Acoustics, Chinese Academy of Sciences, Beijing 100080;2. School of Electronics and Information, Jiangsu University of Science and Technology, Zhenjiang 212003)
  • Received:1900-01-01 Revised:1900-01-01 Online:2008-10-20 Published:2008-10-20

摘要: 文本分类是文本挖掘的一个重要组成部分,是信息搜索领域的一项重要研究课题。该文提出一种基于文章标题信息的汉语自动文本分类方法,在HNC理论的领域概念框架下,通过标题信息所蕴涵的领域信息词语激活对应的HNC领域,实现文本的自动分类。实验证明,该方法与采用SVM算法进行文本分类的方法比较,测试速度和分类平均准确率明显提高。

关键词: 文本分类, HNC理论, 领域

Abstract: The text categorization is an important part of the text excavation, and it becomes the research topic in the information searching field. The paper proposes a method of the automatic text categorization which uses the article title information. The method based on the domain concept frame of the Hierarchical Network of Concepts(HNC) theory uses the domain concept of the title information to activate the corresponding HNC domain, and realizes the text automatic categorization. The experiment proves that the method may effectively enhance the efficiency and the accurate rate of the text automatic categorization.

Key words: text categorization, Hierarchical Network of Concepts(HNC) theory, domain

中图分类号: