Title: Community detection in complex networks using multi-objective bat algorithm
Authors: Iyad Abu Doush; We'am Bilal Alrashdan; Mohammed Azmi Al-Betar; Mohammed A. Awadallah
Addresses: Computer Science Department, American University of Kuwait, Salmiya, Kuwait. ' Computer Science Department, Yarmouk University, Irbid, Jordan ' Department of Information Technology, Al-Huson University College, Al-Balqa Applied University, P.O. Box 50, Al-Huson, Irbid, Jordan ' Department of Computer Science, Al-Aqsa University, P.O. Box 4051, Gaza, Palestine
Abstract: Community detection is the problem of identifying communities in which we aim to discover groups of nodes with high connectivity within the same group and with low connectivity outside the group. Community detection is considered to be a non-deterministic polynomial-time hard problem. Heuristic algorithms can be used to solve such a complex optimisation problem. Bat algorithm (BA) is a meta-heuristic optimisation algorithm. The BA can be used to model a multi-objective optimisation problem. In this paper, the multi-objective bat algorithm (MOBA) is adapted to model and solve the community detection problem. In order to evaluate the algorithm, four real-world datasets are used. The performance of the algorithm is compared with seven other methods from the literature. The comparison was in terms of two metrics to check the quality of the obtained community namely modularity (Q) and normalised mutual information (NMI). The results show that the proposed algorithm outperforms all algorithms in one dataset and that it is competitive in other cases.
Keywords: bat algorithm; community detection; multi-objective optimisation; multi-objective bat algorithm; MOBA.
DOI: 10.1504/IJMMNO.2020.106529
International Journal of Mathematical Modelling and Numerical Optimisation, 2020 Vol.10 No.2, pp.123 - 140
Received: 25 Sep 2018
Accepted: 24 Jun 2019
Published online: 09 Apr 2020 *