Community detection in social networks using a hybrid swarm intelligence approach

Article type: Research Article

Authors: Ghasabeh, Alireza^* | Abadeh, Mohammad Saniee

Affiliations: Faculty of Electrical and Computer Engineering, Tarbiat Modares University, Tehran, Iran

Correspondence: [*] Corresponding author: Alireza Ghasabeh, Faculty of Electrical and Computer Engineering, Tarbiat Modares University, Tehran, Iran. E-mail:a.ghasabeh@modares.ac.ir

Abstract: The problem of community detection has been widely studied in numerous research papers in the last decade. There have been several proposed solutions for this problem; however the challenges of this problem have not been fully addressed yet. In this paper, we hybridize the idea of Ant Colony clustering, which is a local search solution, with global search ability of Honey Bee Hive Optimization to detect communities faster and more accurately. We use the dancer bees for exchanging information among nodes, and a node is considered as an ant in the ant colony clustering. Experimental results on real world networks and also artificial generated graphs show superior performance of our algorithm in comparison to other previous approaches.

Keywords: Community detection, ant colony optimization, swarm intelligence, honey-bees optimization

DOI: 10.3233/KES-150326

Journal: International Journal of Knowledge-based and Intelligent Engineering Systems, vol. 19, no. 4, pp. 255-267, 2015