首页 | 本学科首页   官方微博 | 高级检索  
     

文本特征区域与文本过滤的匹配机制
引用本文:林鸿飞,战学刚,姚天顺.文本特征区域与文本过滤的匹配机制[J].计算机工程与应用,2000,36(7):7-9.
作者姓名:林鸿飞  战学刚  姚天顺
作者单位:东北大学计算机系,沈阳,110006
基金项目:国家自然科学基金!69675019,国家教委博士点基金
摘    要:为了根据用户的信息需求,在因特网上搜索相关文本,该文提出了一种文本过滤的匹配机制,其基本思想是:利用基于词典的概念扩张方法,改进用户模板。计算扩张的用户模板与文本的全局相似度,获取初步的过滤结果;在文本特征区域,进行标题、摘要段、首段和尾段等片断的局部相似度计算,以综合评价文本与用户模板的匹配情况。该方法可操作性强,效果明显。

关 键 词:文本过滤  用户模板  概念扩充  文本特征  向量空间模型

Text Feature Area and the Match Mechanism for Text Filtering
Lin Hongfei,Zhan Xuegang,Yao Tianshun.Text Feature Area and the Match Mechanism for Text Filtering[J].Computer Engineering and Applications,2000,36(7):7-9.
Authors:Lin Hongfei  Zhan Xuegang  Yao Tianshun
Abstract:Text filtering is used to seek the texts in the intemet according to the users' profiles. 'Fhe Paper presents a new match mechanism for text filtering. It improves the user profiles based on the concept expansion and pre-select the texts by global similarity which is calculated by inner product betWeen expanded user profile and the texts. It fi nally evaluates the related texts by local simildrity which is the linear combination of simildrities between user profile and title,abstract,key words,first paragraph,last paragraph and references. As a result,it can give a remarkable effect on the precision of text filtering.
Keywords:Text Filtering  User Profile  Concept Expansion  Text Feature  Vector Space Model
本文献已被 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号