Algorithm Research for Mining Maximal Frequent Itemsets Based on Item Constraints

Frequent item mining has been extensively used in association rules mining. The goal of frequent itemset mining is to discover all the itemsets whose supports in the database exceed a user-specified threshold. However, it often generates a large number of candidate itemset, which reduce the effectiveness of the mining algorithms. Constraint-based mining enables users to provide restraints on mining their interested association rules and can greatly improve the efficiency of mining tasks. In this paper, we propose a fast constraint-based algorithm for mining maximal frequent itemsets. The algorithm introduces item-constraints into the Eclat algorithm, and adopts itemset extension pruning strategy to prun search space. Empirical evaluation showed that the algorithm is very effective and can solve the lack of constrained frequent itemsets algorithm in mining long pattern and intensive database.