关联规则中改进FP-tree的最大频繁模式挖掘算法  

Mining algorithm of maximal frequent pattern based on improved FP-tree in association rules

作　　者：钱雪忠[1] 惠亮[1] 

机构地区：[1]江南大学信息工程学院,江苏无锡214122

出　　处：《计算机工程与设计》2010年第21期4635-4638,共4页Computer Engineering and Design

基　　金：江苏省自然科学基金项目(BK20003017)

摘　　要：关联规则挖掘是数据挖掘领域中重要的研究内容，最大频繁模式挖掘又是关联规则挖掘中的关键问题之一。针对已有的最大频繁模式挖掘算法存在的问题，通过对FP—Growth、FP—Max算法的分析，提出了基于改进FP—tree的最大频繁模式挖掘算法DFP—Max。该算法使用预测、剪枝的策略减少条件FP—tree个数，采用数字集匹配代替项集匹配的方式，减少超集检验的次数，并且避免了中间结果的组合连接，从而使算法达到较高的效率。实验结果表明，在支持度相对较小情况下，DFP—Max的效率是同类算法的2-5倍。Mining association rules is an important matter in data mining, in which mining maximal frequent pattern is a key problem in mining association rules. Aimed at the problem of the existing algorithms, an algorithm of mining maximal frequent patterns - DFP-Max algorithm based on improved FP-tree is put forward after the algorithms of FP-Growth and FP-Max is analyzed. This algorithm uses the strategy of prediction and pruning to reduce the number of generated condition FP-tree, which can not only reduce the checking times but also avoid the combination connection of intermediate results by using digital set matching instead of the testing strategy of itemsets matching. The experiment shows that the efficiency of DFP-Max is two to five times as much as that of the similar algorithms in the case of a relatively small support.

关 键 词：关联规则 数据挖掘 FP—tree 最大频繁项集 超集检验 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]