|
|||||
|
|
| 一种基于语义分析的中文特征值提取方法 | |
| 引用本文： | 邹娟,周经野,邓成. 一种基于语义分析的中文特征值提取方法[J]. 计算机工程与应用, 2005, 41(36): 164-166 |
| 作者姓名： | 邹娟 周经野 邓成 |
| 作者单位： | 湘潭大学信息工程学院,湖南,湘潭,411105;湘潭大学信息工程学院,湖南,湘潭,411105;湘潭大学信息工程学院,湖南,湘潭,411105 |
| 基金项目： | 湖南省自科基金资助项目（编号：02JJY2092） |
| 摘 要： | 文章根据中文文本的特点,不仅考虑了文本中词汇概率信息,还结合了文本语义等多方面来提取文本特征值,从而提出了一种基于语义分析的中文文本特征值提取方法,并给出了具体算法。通过与传统特征值提取方法的比较试验,证明文中提出的特征值提取方法能有效提高文本分类正确率,并达到有效降低特征向量维数的目的。 |
| 关 键 词： | 文本分类 特征值提取 自然语言处理 |
| 文章编号： | 1002-8331-（2005）36-0164-03 |
| 收稿时间： | 2005-06-01 |
| 修稿时间： | 2005-06-01 |
A Characteristic Value Extractive Method for Chinese Texts Through Semantic Analysis |
|
| Zou Juan,Zhou Jingye,Deng Cheng. A Characteristic Value Extractive Method for Chinese Texts Through Semantic Analysis[J]. Computer Engineering and Applications, 2005, 41(36): 164-166 | |
| Authors: | Zou Juan Zhou Jingye Deng Cheng |
| Affiliation: | Information Engineering College of Xiangtan University, Xiangtan, Hunan 411105 |
| Abstract: | Based on the characteristics of Chinese texts,we propose a method of Chinese text characteristic value extraction through semantic analysis in this paper.The characteristic values of the text are extracted according to many aspects.In the extraction we substitute the traditional word with a kind of new synonym concept as units of characteristic value and consider not only the appearance rates of words but also the semantic information in the text. And,the model of characteristic value extraction and algorithm are provided in this paper.Finally,we present the results of the experiments comparing with traditional extraction method using appearance rates of words in the text,which illustrate that the method in this paper improve the correct rate of text categorization and reduce the dimensions effectively. |
| Keywords: | text categorization characteristic value extraction natural language procession |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |