摘要
本文以信息熵的理论为基础,建立了社会化媒体的本体概念抽取模型,开发了社会化媒体的本体概念抽取原型系统。实验证明该中文社会化媒体概念抽取系统具有较好的准确率和召回率,并在对本体概念的同义词抽取上具有一定的优势。
In this paper, we use information entropy theory as the basic theory, then we establishes a so- cial media ontology extraction model and develops the social media ontology extraction prototype systems. After that, we make a experiments to show that the social media concept extraction system has good accu- racy and recall rate. And there is also has certain advantages for the synonym ontology extraction.
出处
《情报科学》
CSSCI
北大核心
2014年第4期9-15,共7页
Information Science
基金
国家自然科学基金项目(71273194)
关键词
概念抽取
词性规则
互信息
左右信息熵
concept extraction
speech rules
mutual information
left and right information entropy
