An enhanced multi-objective biogeography-based optimization for overlapping community detection in social networks with node attributes

Community detection is one of the most important and interesting issues in social network analysis. In recent years, simultaneous considering nodes' attributes topological structures networks process community has attracted attentions many scholars, this consideration been recently used some methods to increase their efficiencies enhance performances finding meaningful relevant communities. But problem that these tend find non-overlapping communities, while real-world include communities often overlap extent. order solve problem, an evolutionary algorithm called MOBBO-OCD, which based on multi-objective biogeography-based optimization (BBO), proposed paper automatically overlapping a with node synchronously density connections similarity network. extended locus-based adjacency representation OLAR introduced encode decode Based OLAR, rank-based migration operator along novel two-phase mutation strategy new double-point crossover are evolution MOBBO-OCD effectively lead population into path. assess performance metric alpha_SAEM paper, able evaluate goodness both partitions two aspects linkage structure. Quantitative evaluations reveal achieves favorable results quite superior 15 algorithms literature.