A New Data Structure to Enhance the Speed of Frequent Pattern Mining

Frequent pattern mining is a crucial task in data mining and plays an important role in data mining applications. The PrePost(+) algorithm is one of the most leading algorithms in this field which has taken an important step toward frequent pattern mining by utilizing PPC data structure in the tree and lists. In this paper, we present a novel data structure, namely, Pre-order Size Code (PSC) to replace PPC data structure with so as to speed up tree building and completion phases and consequently the speed of frequent pattern mining. In contrast to PPC data structure, there is no need for separate pre-order and post-order tree traversals to complete the PSC in tree nodes. To evaluate the performance of the PrePost(+) with the PSC proposed data structure which named PreSize algorithm, we have conducted experiments to compare it with three state-of-the-art algorithms, on a variety of real data sets. The experimental results show that PreSize is a high performance on running time.