摘要
基于《知网》的词汇语义计算方法揭示了词汇间的语义信息。根据文本的向量空间模型描述形式,采用了基于《知网》的词汇语义计算方法来计算两篇文章向量的相关性,并用最大匹配算法来获得这两篇文章的相似度,通过该计算过程达到揭示文本所蕴涵概念的目的,并用实验对该方法的有效性进行了验证,提出了今后的改进方向。
The algorithm used to compute word similarity based on how-net opens out some relational information between words. Ac- cording to VSM, the algorithm was adopted to compute two texts pertinence. And the biggest matching algorithm was used to compute similarity. Experiments showed the validity of algorithm. Finally, development direction in future was presented.
出处
《计算机工程与设计》
CSCD
北大核心
2006年第2期241-244,共4页
Computer Engineering and Design
基金
重庆大学教师资助计划基金项目(2003A33)
