关键词: 社区挖掘, 二分网络, 蚁群优化算法, 二分模块度

Abstract: Detecting communities from networks receives much attention in recent decades, especially from bipartite networks. Detecting communities from bipartite network is very important in the research on the theory and applications of complex network analysis. This paper proposes an algorithm based on ant colony optimization for detecting community structures from bipartite networks. The algorithm firstly transforms the problem of community detection into the problem of ant colony optimization, then constructs a graph model for ants foraging. Meanwhile, this paper redefines heuristic information according the degree of vertexes. Each ant chooses its path according to the pheromone and heuristic information on each path to construct a solution. The quality of solution obtained by each ant is measured by its bipartite modularity. The experimental results show that the proposed algorithm can not only accurately identify the number of communities of a network, but also obtain higher quality of community detection.

Key words: detecting communities, bipartite network, ant colony optimization, bipartite modularity