欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
为了满足当今对评论性信息进行分析的需要,本文设计并实现了一个基于关键词模板的文本褒贬倾向判定系统.本系统定义了关键词类别、建立了关键词库、关键词模板库,并设计了模板匹配算法和文本褒贬倾向值算法,对测试文本进行关键词及模板匹配进而判断测试文本的褒贬倾向.本文还对文本褒贬倾向判定系统进行了测试,并在集内测试取得了较好的效果.进一步分析结果表明在语料充足的条件下,本系统可以更有效的判定文本的褒贬倾向.
英文摘要：
Proposed a recognition system of text valence based on key word template. This system defined key word classification and key word templates, built key word libraries and key word template libraries. This paper proposed template matching arithmetic, text valence value arithmetic. This system did key words matching and templates matching. Then this system calculated the valence of the testing-text. Besides this paper did a test on the recognition system. The analysis result proved this method would be more effective, given enough training corpus.
同期刊论文项目
同项目期刊论文
期刊信息
位置：