Computer Engineering and Applications ›› 2008, Vol. 44 ›› Issue (22): 152-154.DOI: 10.3778/j.issn.1002-8331.2008.22.045

• 数据库、信号与信息处理 • Previous Articles     Next Articles

Research of meta-search engine about multi-keyword information retrieval based on HowNet

FU He-gang,XU Chen-xia   

  1. Department of Computer Science,Chongqing University,Chongqing 400044,China
  • Received:2007-10-10 Revised:2008-01-15 Online:2008-07-11 Published:2008-07-11
  • Contact: FU He-gang



  1. 重庆大学 计算机学院,重庆 400044
  • 通讯作者: 傅鹤岗

Abstract: This paper presents a methodology on multi-keyword relationship which based on HowNet,and gets the kernel-keyword by the analysis,then sets multi-level weight.The design solves the problem of string matching in the information retrieval when it has multi-keyword.The experimental results indicate that precision has greatly improvement.

Key words: multi-keyword, HowNet, sememe, kernel-keyword

摘要: 提出了一种基于知网的多关键词语义关系分析方法,并利用其得出核心关键词,设置分级权重,从而解决了搜索引擎在面对多关键词检索时只作机械性的字符串匹配的缺陷。实验结果表明,查准率有明显提高。

关键词: 多关键词, 知网, 义原, 核心关键词