一种正负关联规则的快速查询扩展算法
首发时间:2013-02-04
摘要:将负关联规则引入到查询扩展研究中,提出了新的查询扩展模型,并设计了一种基于正负关联规则的快速查询扩展算法。该算法通过对文本事务数据库的布尔化表示及数据结构的合理分配,采用向量内积策略来产生频繁和非频繁特征词集,并从中挖掘出词间正负关联规则。实验结果表明,该算法能对原查询词进行快速有效的扩展,且仅需扫描一次文本数据库,并具有动态剪枝,不保留中间候选项和节省大量内存等优点,对信息检索中查询扩展的研究具有重要意义。
For information in English, please click here
Efficient query expansion based on positive and negative association rules
Abstract:This paper introduces negative association rules to the field of query expansion, and proposes new models of query expansion; meanwhile, we design an algorithm of query expansion based on positive and negative association rules. By converting the text database to Boolean Vector Matrix, and allotting equitable data storage structure, this algorithm can produce frequent and infrequent feature terms according to the inner vector product, and get positive and negative association rules between terms. Experimental results show that this algorithm can expand original query terms efficiently and effectively, and scan the database only once. Meanwhile, it has advantages such as pruning dynamically, without saving mid items, and saving lots of memories, which is important to the research of query expansion in information retrieval.
Keywords: data mining negative association rules information retrieval query expansion
论文图表:
引用
No.****
同行评议
勘误表
一种正负关联规则的快速查询扩展算法
评论
全部评论0/1000