基于排序矩阵和树的关联规则挖掘算法

最大频繁项集的生成是影响关联规则挖掘的关键问题，Apriori算法从大量的候选频繁项集产生频繁项集的过程是非常耗时的过程。本文提出了一种新的算法，该算法结合项集的有序特性构造矩阵，使生成树的每一层结点从左往右按支持度大小升序排列，这样得到的候选频繁项集的集合是最小的，大大减少了候选频繁项集的数量，而且能保持频繁项集的完整性，从而节约了计算开销，提高了算法的效率。

Algorithm for Association Rule Minning Based on Ordinal Matrix and Tree

Generating the frequent itemsets is a key problem of association rule mining. It is important that determining the frequent itemsets from a huge amount of candicate itemsets is the most time-consuming part of the process in Apriori algorithm. This paper proposes a new algorithm,which combines the ordinal character of itemsets to create matrixes and makes the nodes ascending order by support count in the tree,so that the sets of the candidate frequent itemsets is the least totally. It can greatly reduce the candidate frequent itemsets, keeps the completion of frequent itemsets, reduces the cost of computing,and improve the efficiency of algorithm.