Varied density based graph clustering algorithm for social networks

• Published in: 2017 International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC) pp. 520 - 524
• Main Authors: Sowjanya, M. Venkata, Padmaja, T. Maruthi
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01-02-2017
• Subjects: Clustering algorithms; Community; DBSCAN; DENGRAPH; Facebook; Heuristic algorithms; incremental clustering; Mobile communication; Partitioning algorithms; Prediction algorithms; Social Networks
• ISBN: 1509032428; 9781509032426
• DOI: 10.1109/I-SMAC.2017.8058404

Detecting communities from large social networks, which are graphical in structure is a critical issue. Recently, many attempts were made to provide feasible solutions and Density Based Graph Clustering (DENGRAPH) is one among them. However, DENGRAPH fails to predict the varied density clusters across the given network, which is significant for dynamic data sets like social networks. In this paper a varied density based DENGRAPH (Varied Density DENGRAPH) clustering algorithm with new merging and reduction criteria for social network is presented. The performance of the proposed approach is illustrated by Facebook social network data. The results have shown that, obtained clusters with varied DENGRAPH are more compact than the clusters with DENGRAPH.