An ant colony approach in the detection of communities in complex networks

Community structure is one of the most important properties of complex networks. Over the years, various approaches had been used to detect communities in complex networks. In this paper, the ant colony optimization (ACO) algorithm is implemented to detect the initial communities. By utilizing the similarity between the nodes as the heuristic information, this variant of ACO is developed to optimize the densities of the communities within a limited number of iterations. Once the initial communities are detected, other community detection methods are implemented to obtain the final detection results. The ACO algorithm can produce good initial communities in both synthetic and real-world networks. By combining the proposed algorithm with the existing community detection methods, good quality final communities can be detected in those networks. The results are comparable to the community detection results which are obtained by using the existing community detection methods. The distinct feature of the proposed algorithm is the prevention of some community detection methods in getting trivial detection results, where a method fails to detect any community structure in a network.