An Ant Colony Random Walk Algorithm for Overlapping Community Detection

Discovery of communities is a very effective way to understand the properties of complex networks. An improved ant colony algorithm based on random walk has been proposed in this paper. Inspired by the framework proposed in AntCBO, firstly, a list of node importance is obtained through calculation. The nodes in the network will be sorted in descending order of importance. Secondly, on the basis of random walk, a matrix is constructed to measure the similarity of nodes and we can use this matrix and pheromone to get the heuristic information. Thirdly, an improved ant's location discovery strategy is proposed. After the movement of ants, every node will keep a list of labels and the proposed post processing will give the result of overlapping community detection. Finally, a test in real-world networks is given. The result shows that this algorithm has better performance than existing methods in finding overlapping community structure.