摘要
中文文本主题词的提取可以浓缩一篇文章,可以提炼一个中文网页,还可以帮助实现网上广告与网页的精确匹配。提出了一种基于同义词词林的中文文本主题词提取方法,不仅考虑了传统的影响主题词语权重的因素,还考虑到了同义词、相关词以及下位词的出现对于词语权重的影响。实验表明。
Thematic words extraction from a Chinese text not only can concentrate an article,but also can extract main ideas from a Chinese Web and help to achieve precise matching between online advertisement and a webpage. The paper presents a method of thematic words extraction based on Tongyici Cilin. The method not only has taken traditional factors affecting the weight of a thematic word into account, but also has considered the factors such as the appearance of relevant words, synonymy and lower words. Experiments have confirmed that the accuracy rate of thematic word extraction from a Chinese text can reach 83.25% using this method.
出处
《广西师范大学学报(自然科学版)》
CAS
北大核心
2007年第2期145-148,共4页
Journal of Guangxi Normal University:Natural Science Edition
基金
国家自然科学基金资助项目(60272084)
北京市教育委员会科技发展计划重点项目(KZ200310772013)
北京市教委项目(KM200510772008
KM200610772008)
关键词
主题词提取
同义词词林
权值
同义词
thematic words extraction
tongyici cilin
weight
synonymy