An Improved Gray Wolf Optimization Algorithm with a Novel Initialization Method for Community Detection

Community discovery (CD) under complex networks is a hot discussion issue in network science research. Recently, many evolutionary methods have been introduced to detect communities of networks. However, optimization-based community still suffers from two problems. First, the initialization population quality current algorithm not good, resulting slow convergence speed, and final performance needs be further improved. Another important that CD inconsistent detection at different scales, showing dramatic drop as scale increases. To address such issues, this paper proposes an based on novel initial method improved gray wolf optimization (NIGWO) tackle above problems same time. In paper, strategy proposed generate high-quality greatly accelerate speed evolution. The effectively fused elite substructure features dependency other among nodes. Moreover, GWO presented with new search strategies. An hunting prey stage retain excellent substructures populations quickly improve structure. Furthermore, mutation strategies node level are designed encircling stage. Specifically, boundary nodes mutated according function efficiency save computation assumption. Numerous experiments proven our obtains more most compared 11 state-of-the-art algorithms.