应用粒子群算法的重叠社区发现

Application of PSO algorithm in overlapping community detection

随着大数据的出现,越来越多研究者对复杂网络的社区发现感兴趣,现有社区发现算法大多为检测不重叠社区的.提出一种基于粒子群算法的重叠社区划分法,初始粒子群时考虑非法划分的产生,用标签传播法调整每个粒子的编码.在一种经典数据集上测试,验证了该算法有效性,能快速检测出网络中潜在的社区结构.

With the phenomenon of big data emerged.More and more researchers interested in finding community of complex network.But the existing community detection algorithms mostly assume that no overlaps exist.An overlapping community detection algorithm based on particle swarm optimization is proposed.In the initialization phase,a label propagation algorithm is utilized on optimization variables of each particle for coding adjustment,to avoid illegal community.In experiments,the algorithm is applied to two classic datasets to demonstrate the effectiveness of the algorithm,capability of detecting the potential community structure quickly in networks.