Memory Efficient Mining of Maximal Itemsets using Order Preserving Generators

In this paper, we propose a memory efficient algorithm for maximal frequent itemset mining from transactional datasets. We propose OP-MAX* (Order Preserving – MAXimal itemset mining) algorithm, which mines all the maximal itemsets from transactional datasets with less space and time. Our methodology uses a memory efficient maximality checking technique to generate frequent maximal itemsets. We have also incorporated several optimization techniques to improve the mining efficiency. Experiments involving publicly available datasets show that our algorithm takes less memory and less computation time than other algorithms in most cases.