基于完全化语义的关键词检索研究  

Research on keyword retrieval based on perfect semantics

作　　者：黎方正[1] 谢东[1] 

机构地区：[1]中南大学信息科学与工程学院,长沙410083

出　　处：《计算机应用研究》2010年第10期3794-3797,共4页Application Research of Computers

基　　金：湖南省自然科学基金资助项目(07JJ3119);湖南省教育厅科研基金资助项目(08B040)

摘　　要：已有方法解决了结果集要求是关于关键词检索结果全集而带来的无检索结果的问题,但非空子集作为结果集有较大冗余。提出了一种基于完全化语义的关键词检索方法,对检索关键词进行语义标准化处理,识别处理后采用不同的检索评分器评分,基于检索核给出了针对不同类型的检索算法,生成连接元组集,取得相邻拓展连接元组集,转换得到SQL检索语句,将所有的查询结果返回给用户。实验表明,采用了检索核的方法可以很好地避免数据冗余情况。Existing methods resolve the problem,which can not return retrieval results since the result set is respond to the univeral set of the keyword retrieval result. However,a non-null subset has more redunndancy as the result set. This paper proposed a keyword retrieval method based on perfect semantics,it made the semantics standradiztion for retrieval keywords. The way employeed different retrieval scorers to score keywords after distinguished keywords. Proposed the different type of retrieval algorithms based on the retrieval focus to generate linked tuple sets,which was transformed to a SQL retrieval sentence to return all query results for users. The experiments show that the method with retrieval focus avoids data redunndancy well.

关 键 词：关系数据库 关键词检索 检索核 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]