Research on Community Detection Algorithm for Complex Networks with Multi-Attribute Nodes

The study of complex network has attracted more and more attention from researchers, and community detection has gradually become one of the most significant directions. The majority of existing community detection algorithms regard the connections between nodes as single relationship, which are not fully satisfied with the practical needs. This paper described complex networks with multi-attribute nodes and proposed a community detection algorithm (CDMAN). The algorithm represented a node's information in real world as a set of node's attributes in networks, then the properties and structural similarity of nodes were employed to take further partition. Experimental results on real datasets demonstrate that the algorithm (CDMAN) outperforms the traditional structure algorithm (LPA) and is proved to be more feasible and promising.