1. Fuzzy Community Detection Based on Elite Symbiotic Organisms Search and Node Neighborhood Information
- Author
-
Jing Xiao, Yan-Jiao Wang, and Xiaoke Xu
- Subjects
Modularity (networks) ,Speedup ,Computer science ,business.industry ,Applied Mathematics ,Node (networking) ,Stability (learning theory) ,Network topology ,Fuzzy logic ,Computational Theory and Mathematics ,Artificial Intelligence ,Control and Systems Engineering ,Convergence (routing) ,Artificial intelligence ,business ,Metaheuristic - Abstract
In the last decade, fuzzy community detection has received increasing attention, since it} can not only uncover community structure of a network, but also reflect the membership degrees of each node to multiple communities. Although some pioneers proposed a few algorithms for finding fuzzy communities, there is still room for further improvement in the quality of detected fuzzy communities. In this study, a metaheuristic-based modularity optimization algorithm, named Symbiotic Organisms Search Fuzzy Community Detection (SOSFCD) is proposed. On the one hand, an improved bio-inspired metaheuristic algorithm, Elite Symbiotic Organisms Search (Elite-SOS), is designed as optimization strategy to improve the global convergence of fuzzy modularity optimization. On the other hand, a Neighbor-based Membership Modification (NMM) operation is proposed to intensify exploitation ability and speed up convergence, by efficiently utilizing local information (i.e., node neighborhood) of network topology. Experimental results on both of synthetic and real-life networks with different scales and characteristics show that SOSFCD can find max-modularity fuzzy partitions and coverings, which outperforms many state-of-the-art algorithms in terms of accuracy and stability.
- Published
- 2022