基于Copulas框架的混合式查询扩展方法

基于语义资料和局部分析的混合式查询扩展可以同时提供具有语义相关性和时效性的扩展结果,但如何有效地混合不同相似度度量指标是尚未解决的问题。提出了一种基于Copulas框架的混合式查询扩展方法,在统一框架内实现了不同类型相似度度量指标的合并。该方法基于语义分析及词语共现分析方法,分别计算扩展词与用户查询词的语义及统计相似概率,进而在Copulas框架下融合扩展词集,选取最高质量的扩展词形成查询扩展。实验结果表明,该方法充分利用了语义及词语共现分析查询扩展方法的优点,有效地弥补了两者的不足,提高了搜索结果的查准率,具有更优的搜索性能。

Combined Query Expansion Method Based on Copulas Framework

Hybrid query expansion methods based on semantic and local analysis can provide time-sensitive extension results with semantic correlation.However,how to effectively combine two different kinds of similarity metrics has not been solved.This paper proposd a hybrid query expansion method based on the Copulas framework to implement the combination of different types of similarity metrics.Based on the query expansion methods of semantic and word co-occurrence analysis,the proposed method respectively calculates the semantical and the statistical similar probabilities between expansion-words and the query words submitted by the user.It then selects high quality extension words to obtain the final extension word set.The experimental results show that the method makes full use of the advantages of the two kinds of query expansion methods to improve the precision ratio factor.The method has better search performance.