MaxMining: A Novel Algorithm for Mining Maximal Frequent Itemset

Abstract

We present a new algorithm for mining maximal frequent itemsets, MaxMining, from big transaction databases. MaxMining employs the depth-first traversal and iterative method. It re-represents the transaction database by vertical tidset format, travels the search space with effective pruning strategies which reduces the search space dramatically. MaxMining removes all the non-maximal frequent itemsets to get the exact set of maximal frequent itemsets directly, no need to enumerate all the frequent itemsets from smaller ones step by step. It backtracks to the proper ancestor directly, needless level by level, ignoring those redundant frequent itemsets. We found that MaxMining can be more effective to find all the maximal frequent itemsets from big databases than many of proposed algorithms with ordinary pruning strategies.