Community detection in the textile-related trade network using a biased estimation of distribution algorithm

It is very important to discovery potential customers or groups for the traditional textile enterprises to enhance its’ market competitiveness. Aiming to discover the implicit characters of the textile-related trade system, A biased estimation of distribution algorithm was proposed to detect the community structure in this paper. This algorithm combined a biased search and a simulated annealing selections strategy which were used to improve both the convergence speed and the accuracy of the EDAs to discovery the communities structure for complex system by maximizing the modularity density. The biased search is efficient by taking into account an asymmetric similarity between any pairs of nodes in network according to the different characteristics and local environment of nodes. The proposed algorithm was applied to detect the community structure for a textile-related trade network with the scale-free character extracted from a set of textile companies by uniquely leveraging each node with economic behavior, and the result show that the algorithm is efficient and competent