Community Mining in Signed Networks Based on Dynamic Mechanism

Topology structure of the networks is crucial to understand their structural and functional characteristics. Networks may be social networks, internet, political networks, networks that involve natural language connections, and so on. Topics of community detection in signed networks have attracted a lot of attention in recent years. In this paper, a novel network model based on dynamic mechanism is proposed to detect the community structure in signed networks. Similarities between nodes are defined to select out positive neighbors with higher similarity and negative neighbors with lower similarity. Initial state values of nodes in networks are randomly generated. With the evolution of time, nodes with higher positive similarity would cluster together and nodes with lower negative similarity would get away. Finally, all nodes would cluster into different groups with different stable state values. By the Lyapunov stability theory, the proposed network model is proved to be uniformly stable. Lots of real networks and synthetic networks are tested to verify our proposed method. Besides, thorough comparisons demonstrate that the presented method is superior to three state-of-the-art algorithms.