A multi-objective evolutionary algorithm based on mixed encoding for community detection

Abstract

Community structure is one of the most significant features of complex networks and community detection is a crucial method to analyze community structure. Existing representations in community detection are inflexible and easily generate invalid solutions. To address the drawbacks, this paper proposed a multi-objective evolutionary algorithm based on mixed encoding (MOGAME). The algorithm combines the locus-based representation and labels-based representation, which can avoid generating invalid solution and improve the performance. Extensive experiments on both synthetic and real-word networks show that the proposed algorithm performs better than the existing algorithms with respect to accuracy and stability.

An adaptive dynamic community detection algorithm based on multi-objective evolutionary clustering

“…In the real world, various large-scale complex systems can be abstracted as complex networks, such as social networks, protein interaction networks and transportation networks (Yang et al ., 2023; Wu et al ., 2020; Huang et al ., 2021). These networks have complex structures and diverse node behaviors, so the whole network can be divided into different community structures (Ma and Dong, 2017; Zhang et al ., 2023).…”