基于特征概念自动提取系统的研究  

Research of Automatic Extraction System Based on Feature Concept

作　　者：曾朝晖[1] 邓曙光[1] 陆文彦[2] 

机构地区：[1]湖南城市学院,湖南益阳413000 [2]中南大学,长沙410083

出　　处：《湖南城市学院学报（自然科学版）》2006年第1期69-71,74,共4页Journal of Hunan City University:Natural Science

基　　金：湖南省普通高校青年骨干教师资助计划(教育厅2003[165]号)

摘　　要：文本分类与文本信息特征概念的提取是当前智能信息服务研究的重点.为自动获取新的特征概念,提出了一种基于特征概念的自动提取系统.该系统包括分词、综合文本词权处理、类别归属和特征概念提取等部分,能有效地从概念上提取文本类特征,提高文本自动分类的准确性.Text categorization and feature concept extraction is the focal point of basic research in the field of intelligent information service system. An automatic extraction system based on feature concept is proposed on the ground that new feature concept needs to extract automatically. There are some key parts in this system, such as word segmentation, dealing with integrative word weight of text, ascription of category, extraction of feature concept and so on. This system can conduct efficiently text feature extraction thus to nromote the accuracy of automation text categorization.

关 键 词：特征概念 文本分类 特征提取 

分 类 号：TP391.12[自动化与计算机技术—计算机应用技术]