Mining frequent maximal cliques efficiently by global view graph

• Published in: 2012 9th International Conference on Fuzzy Systems and Knowledge Discovery pp. 1362 - 1366
• Main Authors: Guanling Lee, Sheng-Lung Peng, Shih-Wei Kuo, Yi-Chun Chen
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.05.2012
• Subjects: Algorithm design and analysis; Data mining; Frequent subgraph mining; Graph mining; Graph summary; Itemsets; Knowledge based systems; Maximal clique mining; Testing; Topology
• ISBN: 9781467300254; 146730025X
• DOI: 10.1109/FSKD.2012.6233927

Graph mining problem has been a popular research issue in recent years. Many kind of data can be represented as a graph and solve the particular problem by using a specific graph algorithm. Recently, the applications of graph mining are growing quantity. In this paper, the main subject is to find a specific topology called clique which is maximal and frequent in a set of graphs. In our approach, the graphs are first summarized into a global view graph. It is shown that any clique contains in the graph database, there must exist an isomorphic subgraph in the summarized graph according to our summarization process. Therefore, the frequent maximal clique mining process will focus on the global view graph. Moreover, by comparing with other existing methods, a set of experiments is performed to show the benefit of our approach.