SDM: A frequent subgraph mining algorithm for sequential directed graphs

Abstract

Current research on frequent subgraph mining mainly focuses on the mining of undirected graphs, but in fact, the research on the mining of directed graphs has more practical significance. A new algorithm for mining frequent subgraphs based on directed graphs, SDM, is proposed. This algorithm uses breadth-first search strategy to mine frequent subgraphs by establishing a hierarchical tree space. And provide an efficient method in the process of frequent subgraph generation and counting. Experimental research shows that compared with other frequent subgraph mining algorithms, SDM has achieved a significant performance improvement.