Community detection is the process of detecting communities in complex networks. Communities are important structures that can help us further study the properties of complex networks. In recent years, swarm intelligence algorithms have been applied to community detection and have achieved remarkable results. However, these existing algorithms have limited search ability and easily fall into the problem of local optima. In this paper, we propose a new community detection approach based on an improved whale optimization algorithm (WOA). The WOA is applied to a discrete symbol space in solving the community detection problem, therefore topology structure-based search strategies, adjustment and mergence policies, and evolutionary population method are designed to improve the efficiency and effectiveness of the method. Then, a whale optimization algorithm with evolutionary population for community detection (EP-WOCD) is proposed. Extensive experiments are conducted to compare the EP-WOCD with other state-of-the-art algorithms on both artificial and real-world social networks. Experimental results show that the EP-WOCD is effective and stable.

社区检测是在复杂网络中检测社区的过程。社区是重要的结构，可以帮助我们进一步研究复杂网络的属性。近年来，群体智能算法已应用于社区检测，并取得了显著成果。但是，这些现有算法搜索能力有限，容易陷入局部最优问题。在本文中，我们提出了一种基于改进的鲸鱼优化算法（WOA）的新的社区检测方法。WOA被应用于离散符号空间以解决社区检测问题，因此设计了基于拓扑结构的搜索策略，调整与合并策略以及进化种群方法，以提高该方法的效率和有效性。然后，提出了一种进化种群的鲸鱼群落优化算法（EP-WOCD）。进行了广泛的实验，以将EP-WOCD与人工和现实世界社交网络上的其他最新算法进行比较。实验结果表明，EP-WOCD是有效且稳定的。