基于权重标准化SimRank方法查询扩展技术研究.pdfVIP

基于权重标准化SimRank 方法的查询扩展技术研究* 马云龙，林鸿飞 大连理工大学信息检索研究室，大连，116024 E-mail: kevinma@ , hflin@ 摘 要：查询扩展是信息检索中的一项重要技术。传统的局部分析查询扩展方法利用伪相关文档作为候选 词集合，然而部分伪相关文档并不具有很高的相关性。本文利用真实的搜索引擎查询日志，建立了查询点 击图，经过多次图结构的转化得到能够反映词之间关联程度的词项关系图，并在图结构的相似度算法 SimRank 的基础上，提出了一种基于权重标准化的改进 SimRank 方法，该方法利用词项关系图中词项的 全局和间接关系，能够有效挖掘与原始查询相关联的扩展词。同时，为降低SimRank 算法的计算复杂度， 本文采用了剪枝等策略进行优化，使得计算效率有大幅提高。在TREC 标准数据集上的实验表明，本文的 方法可以有效地选择相关扩展词。MAP 指标较局部分析查询扩展方法提高了 1.81%，在 P@10 和 P@20 指标评价中效果分别提高了5.44%和3.73% 。 关键词：搜索引擎；查询扩展；查询日志；SimRank A Weight Normalized SimRank Approach for Query Expansion Yunlong Ma, Hongfei Lin Information Retrieval Laboratory, Dalian University of Technology, Dalian 116024 E-mail: kevinma@ Abstract: As an important technology in information retrieval, and Traditional method of query expansion uses the pseudo-relevant documents as the candidate words set. But, some of pseudo-relevant documents are not highly relevant. In our work, a query-click graph was built by a query log in real search engine. The term relationship graph which was obtained by several transformations could reflect the direct relationship of the terms. We proposed a normalized weight SimRank approach, a revised algorithm based on the SimRank, would be used for query expansion. In order to reduce the computational complexity of SimRank, strategies like pruning were used to optimize the algorithm. Experiments on large real AOL search engine query logs and a standard TREC corpus shows that our approach can discover the quality expansion terms effectively. The MAP of our approach is 1.81% higher than the query expansion based on pseudo relevance feedback, 5.44% higher on P@10, and 3.73% higher on P@20. Keywords: Search engine; Query