Research On Mining Maximal Frequent Subgraphs Approach

Mining Association Rules is the key task of data mining, frequent pattern mining is the necessary stage of association rules. The non-structured frequent pattern mining has been a very good development, but the non-structured data is difficult to express relation in the objects. With the development of structure frequent pattern mining, subgraphs pattern becomes into a new research focus. At present, the urgent problem in the field of the mining subgraphs is how to improve the subgraphs mining algorithm efficiency. Due to the frequent subgraphs mining generate a great results set, this restricts the algorithm performance to some extent, and mining maximal frequent subgraphs can effectively reduce the results set of the frequent subgraphs. Therefore, this paper focuses on the research to mining maximal frequent subgraphs algorithm to improve the efficiency of the algorithm and the major works are as follows.1) Based on the study of the typical mining subgraphs pattern algorithm, this paper proposes an algorithm MFME for mining maximal frequent subgraphs. Isomorphic relation in graphs is mapped to mapping edges set and maximal frequent subgraphs can be mining form mapping edges set through the algorithm MFME, and the structure of the mapping tree is only a depth-first traversal. Therefore, algorithm MFME can reduce time overhead of the subgraphs isomorphism judgment effectively, and the efficiency of algorithm MFME comparing with the similar algorithm SPIN is improved obviously.2) This paper discusses the data decomposition technology and parallelizes the algorithm MFME.