科研之友
微信
新浪微博
Facebook
分享链接
摘要
The keywords of multiple documents declare the main information of these documents. Keyword extraction is an important basis of text mining. In this paper, we present an algorithm of keyword extraction from multilingual corpus which consists of multiple documents. The algorithm uses bilingual knowledge base to compute relationship and similarity between concepts in cross-language manner. And thus perform semantic disambiguation and abstract keywords from documents. At last, we evaluated the algorithm on a Chinese-English bilingual corpus. The result proves the effectiveness of our algorithm.
- 出版日期2008
- 单位北京理工大学
