北京邮电大学学报

  • EI核心期刊

北京邮电大学学报 ›› 2006, Vol. 29 ›› Issue (6): 107-110.doi: 10.13190/jbupt.200606.107.meix

• 研究报告 • 上一篇    下一篇

一种基于语义关联的查询优化方法

梅 翔, 孟祥武, 陈俊亮, 徐萌   

  1. 北京邮电大学 网络与交换技术国家重点实验室, 北京 100876
  • 收稿日期:2005-12-27 修回日期:1900-01-01 出版日期:2006-12-30 发布日期:2006-12-30
  • 通讯作者: 梅 翔

A Query Refinement Scheme Based on Semantic Association Ranking

MEI Xiang, MENG Xiang-wu, CHEN Jun-Liang, XU Meng   

  1. State Key Laboratory of Networking and Switching Technology, Beijing University of Posts and Telecommunications, Beijing 100876, China
  • Received:2005-12-27 Revised:1900-01-01 Online:2006-12-30 Published:2006-12-30
  • Contact: MEI Xiang

摘要:

提出了一种综合词法关系和语义分析的查询优化方法。通过对查询关键字词法特性和本体实例之间语义关联强弱的分析,提高了查询关键字到本体概念映射的完整性和准确率,帮助搜索引擎对用户的意图做出有效推测,解决了传统搜索引擎不能很好理解用户意图的问题。设计实现了相应的查询优化系统,实验表明该系统可以有效提高搜索的准确率。

关键词: 搜索引擎, 查询优化, 语义关联, 词网

Abstract:

A query refinement scheme, which synthesizes the synonymic relation and the semantic association, is presented. By analyzing the synonymic characteristic of the key words and the intensity of semantic association among the ontology instances, the accuracy and completeness of the mapping from key word to ontology entities can be improved. A query refinement system is designed and implemented. The experimentation results show the system can optimize query input efficiently.

Key words: Web search, query refinement, semantic association, WordNet

中图分类号: