基于模糊聚类的多分辨率社区发现方法

Multiresolution Community Detection Based on Fuzzy Clustering

针对网络结构的复杂性和群体划分的不确定性,该文提出一种基于模糊聚类的多分辨率社区结构发现方法。该方法用模糊方法来处理网络节点间的相似性,以实现社区结构的模糊划分。基于节点间的局部交互信息,考虑节点间的模糊关系和网络拓扑结构相似性传递,实现网络社区的层次聚类。并通过调节模糊参数,挖掘出不同分辨率下的社区结构。同时为了避免主观地确定社区数目,引入一种新的模块度以度量社区划分结果。实验证明该方法能够有效且稳定地揭示潜在的社区结构。

Focusing on the complexity of network structure and the indeterminacy of community partition, this paper puts forward a novel fuzzy clustering method for uncovering community structures. In contrast to previous studies, the proposed method disposes the similarity of connecting vertices with fuzzy relation. Based on local interactive information, it considers the fuzzy relation between vertices and the transitive similarity in network topology to divide vertices into communities. In addition, multiresolution communities can be detected by adjusting fuzzy parameter. In order to avoid subjectivity in the selection of cluster number, a new modularity is introduced to evaluate the effectiveness of the clustering analysis. It is proved by experiments that the method is efficient and stable to detect underlying communities.