Efficient Mining of Maximal Patterns using Order Preserving Generators

In this paper, we address the problem of maximal frequent pattern mining from transactional datasets. Many of the existing algorithms for mining maximal patterns are based on frequent patterns which consume large amount of time & space. We propose OP-MAX (order preserving-maximal pattern mining) algorithm, which mines all the maximal patterns from transactional datasets with less space and time. Our methodology computes frequent closed maximal patterns and outputs maximal patterns among them. We also incorporate several optimization techniques to improve the mining efficiency. Experiments involving publicly available datasets show that our algorithm outperforms in time complexity when compared to the recently proposed FP-max, Eclat and MAFIA algorithms.