Frequent Closed Subgraph Mining: A Multi-thread Approach

Lam B. Q. Nguyen, Ngoc-Thao Le, Hung Son Nguyen, Tri Pham, Bay Vo

Abstract: Frequent subgraph mining (FSM) is an interesting research field and has attracted a lot of attention from many researchers in recent years, in which closed subgraph mining is a new topic with many practical applications. In the field of graph mining, GraMi (GRAph MIning) is considered state-of-the-art, and many algorithms have been developed based on the improvement of this approach. In 2021, we proposed the CloGraMi algorithm based on GraMi to mine closed frequent subgraphs from a large graph rapidly and efficiently. However, with NP time complexity and extremely high cost in terms of running time, graph mining is always a challenging problem for all researchers. In this paper, we propose a parallel processing strategy aiming to improve the execution speed of our CloGraMi algorithm. Our experiments on six datasets, including both undirected and directed graphs, with different sizes, including large, medium and small, show that the new algorithm significantly reduces running time and improves performance, and has better performance compared to the original algorithm.