Abstract:
One of the tasks related to the study of the of complex networks is the task of revealing communities structure – splitting all vertices into groups (communities), so that the vertices of each group are more closely related to each other than to the rest of the graph. A popular algorithm for detecting
communities is the Blondel, based on the maximization of Newman-Girvan modularity, a common criterion for assessing the quality of community divisions. This article is devoted to the analysis of its features and work results, as well as possible modifications. The test results are analyzed both on the generated graphs and on real data.
Keywords:graph structure, social network analysis, community detection, big data.
