Fast algorithm for mining minimal generators of frequent closed itemsetsand their applications

Abstract

The number of frequent closed itemsets (FCIs) are usually fewer thannumbers of frequent itemsets. However, it is necessary to find MinimalGenerators (mGs) for mining association rule from them. The finding mGsapproaches based on generating candidate lose timeliness when the number offrequent closed itemsets are large. In this paper, we present MG-CHARM, anefficient algorithm for finding all mGs of frequent closed itemsets. Based onthe mGs properties mentioned in section 2.4, we develop an algorithm which doesnot generate candidates by mining directly the mGs of frequent closed itemsetsat mining FCIs. Thus, the time for finding mGs of frequent closed itemsets isinsignificant. Experiment shows that the time of MG-CHARM is fewer than the timeof finding mGs after finding all closed itemsets (CHARM), especially in case ofthe the length of each frequent closed itemset is long. We also presentapplications of mGs for mining (minimal) non-redundant association rules. ©2009 IEEE.