一种基于F-矩阵的最大频繁项目集快速挖掘算法

A Fast Mining Algoritm of Maximum Frequent Itemsets Based on F-matrix

最大频繁项目集挖掘是多种数据挖掘应用研究的一个重要方面,最大频繁项目集的快速挖掘算法研究是当前研究的热点。传统的最大频繁项目集挖掘算法要多遍扫描数据库并产生大量的候选项目集。为此,该文提出了基于F-矩阵的最大频繁项目集快速挖掘算法FMMFIBFM,FMMFIBFM采用FP-tree的存储结构,仅须扫描数据库两遍且不产生候选频繁项目集,有效地提高了频繁项目集的挖掘效率。实验结果表明,FMMFIBFM算法是有效可行的。

Mining maximum frequent itemsets is a major aspect of data mining researches.Efficient mining algorithm research of maximum frequent itemsets is the key problem.Conventional algorithms of maximum frequent itemsets need generate a lots of candidate sets,and need repeatedly scan the database,especially when there exist prolific patterns and or long patterns.In order to overcome the drawbacks of traditional mining algorithms,in this paper,the authors introduce algorithm FMMFIBF (fast mining maximum frequent itemsets based on F-matrix) ,FMMFIBF only scan database twice by using FP-tree structure,and need not to generate any candidate itemsets,so mining efficiency of maximum frequent itemsets is obviously improved.Experimental result shows that FMMFIBF algorithm is effective and efficient.