Important subgraph discovery using non-dominance criterion

Graph mining techniques have received a lot of attention to discover important subgraphs based on certain criteria.  These techniques have become increasingly important due to the growing number of applications that rely on graph-based data.  Some examples are: (i) microarray data analysis in bioinformatics, (ii) transportation network analysis, (iii) social network analysis.  In this study, we propose a graph decomposition algorithm using the non-dominance criterion to identify important subgraphs based on two characteristics: edge connectivity and diameter.  The proposed method uses a multi-objective optimization approach to maximize the edge connectivity and minimize the diameter.  In a similar vein, identifying communities within a network can improve our comprehension of the network's characteristics and properties.  Therefore, the detection of community structures in networks has been extensively studied.  As a result, in this paper an innovative community detection method is presented based on our approach.  The performance of the proposed technique is examined on both real-life and synthetically generated data sets.