Discovery of Correlated Sequential Subgraphs from a Sequence of Graphs

Dynamic graphs or a sequence of graphs attract much attention recently. In this paper, as a first step towards finding significant patterns hidden in dynamic graphs, we consider the problem of mining successive sequence of subgraphs which appear frequently in a long sequence of graphs. In addition, to exclude insignificant patterns, we take into account the mutual dependency measured by *** -correlation coefficient among the components in patterns. An algorithm named CorSSS, which utilizes the generality ordering of patterns effectively, is developed for enumerating all frequent and correlated patterns. The effectiveness of CorSSS, is confirmed through the experiments using real datasets.

[1]  Hans-Peter Kriegel,et al.  Pattern Mining in Frequent Dynamic Subgraphs , 2006, Sixth International Conference on Data Mining (ICDM'06).

[2]  Ryohei Orihara,et al.  A Sequential Pattern Mining Method based on Sequential Interestingness , 2008 .

[3]  Jianyong Wang,et al.  Efficient mining of frequent sequence generators , 2008, WWW.

[4]  Wilfred Ng,et al.  Correlation search in graph databases , 2007, KDD '07.

[5]  Yiming Yang,et al.  The Enron Corpus: A New Dataset for Email Classi(cid:12)cation Research , 2004 .

[6]  Hui Xiong,et al.  Hyperclique pattern discovery , 2006, Data Mining and Knowledge Discovery.

[7]  Ramakrishnan Srikant,et al.  Mining Sequential Patterns: Generalizations and Performance Improvements , 1996, EDBT.

[8]  Unil Yun.  WIS: Weighted Interesting Sequential Pattern Mining with a Similar Level of Support and/or Weight , 2007 .

[9]  Neil Gershenfeld,et al.  MIT-Media Lab , 1991, ICMC.

[10]  Jaideep Srivastava,et al.  Selecting the right interestingness measure for association patterns , 2002, KDD.

[11]  Jiawei Han,et al.  CloseGraph: mining closed frequent graph patterns , 2003, KDD '03.

[12]  Takashi Washio,et al.  A Fast Method to Mine Frequent Subsequences from Graph Sequence Data , 2008, 2008 Eighth IEEE International Conference on Data Mining.

[13]  Tomonobu Ozaki,et al.  Mining Correlated Subgraphs in Graph Databases , 2008, PAKDD.

[14]  Jiawei Han,et al.  gSpan: graph-based substructure pattern mining , 2002, 2002 IEEE International Conference on Data Mining, 2002. Proceedings..

[15]  Christian Borgelt,et al.  Canonical Forms for Frequent Graph Mining , 2006, GfKl.

[16]  Jiawei Han,et al.  BIDE: efficient mining of frequent closed sequences , 2004, Proceedings. 20th International Conference on Data Engineering.

[17]  Tomonobu Ozaki,et al.  Discovery of Internal and External Hyperclique Patterns in Complex Graph Databases , 2008, 2008 IEEE International Conference on Data Mining Workshops.