Adaptive surrogate-assisted evolutionary multi-objective community detection algorithm with core node learning

Community detection is a widely researched area in network science. In recent years, multi-objective evolutionary algorithms (MOEAs) have been widely used in community detection. Continuous coding is able to transform the discrete problem into a continuous one. However, the original continuous coding ignores the relationship between node structures, resulting in a low quality encoded population, which degrades the performance of community detection. In addition, continuous coding needs to be decoded into label-based coding in the optimization process to compute the objective function. To alleviate this, we design the surrogate model adaptive switching strategy that selects the optimal surrogate model for the task. Then, the adaptive surrogate-assisted evolutionary multi-objective community detection algorithm with core node learning is proposed. A core node learning method is used to improve the connection between nodes in augmented sequential coding, which helps initialize the population using the node similarity matrix. The core nodes of the network are then obtained based on the node weights. Thereafter, these core nodes are used to construct a surrogate model between the continuous coding and the objective function. The surrogate model is updated during the optimization process, while effectively improves the accuracy and efficiency of community detection tasks. Experimental results on synthetic and real networks show that the algorithm has better performance compared with seven community detection algorithms.