Graph Hybrid Summarization

One solution for processing and analysis of massive graphs is summarization. Generating a high quality summary is the main challenge of graph summarization. For the aims of generating a summary with a better quality for a given attributed graph, both the structural and attribute-based similarities must be considered. There are two measures, density and entropy, are used to evaluate the quality of structural and attribute-based summaries, respectively. For an attributed graph, a high quality summary is the one that covers the structure and vertex attributes, of-course, with the user-specified degrees of importance. Recently, two methods have been proposed for summarizing/clustering a graph based upon both the structure and vertex attribute similarities. In this paper, a new method is proposed for the hybrid summarization of a given attributed graph, and the quality of the summary generated by the developed method is compared with the quality of summaries generated by the recently proposed method, SGVR, for this purpose. The experimental results showed that the proposed method generates a summary with a better quality.