重叠社区发现的两段策略

复杂网络中的社区特别是重叠社区在信息传播与推荐、舆情控制、商业营销等领域中具有重要作用。在实际的网络中,由于有些节点天然地属于多个功能团体,重叠社区的挖掘越来越受到重视。提出了一种重叠社区挖掘的两段策略算法:初始社区抽取与社区合并。在社区抽取阶段,选择网络中最大度节点及其紧密的邻居节点作为初始社区,将与此初始社区联系紧密的节点也一并加入;在社区合并阶段,如果两个社区合并之后使得模块度增加,则合并这两个社区。用包括大规模网络在内的3个实际网络对所提算法进行了测试,结果表明,该算法可有效挖掘网络中的重叠社区。

Two-phase Strategy on Overlapping Communities Detection

Communities, especial overlapping communities in complex networks are significant in many fields such as in- formation spreading and recommending,public opinion controlling,and commercial marketing. Overlapping communities detecting is attracting increasing attentions since some nodes may naturally belong to several groups in real-world net- works. This paper proposed an overlapping community detecting algorithm based on two phase strategies; initial com- munity extracting and community merging. In extracting phase, a node with maximal degree and its tight neighbors arc selected as an initial community, and nodes tight with the community arc also included. In merging phase, two communi- ties are merged if the modularity gets larger after merging. Three real-world complex networks including a large-scale one were used to evaluate the algorithm. Experimental results demonstrate that the proposed algorithm is efficient for detecting overlapping communities in complex networks.