Generation of complex bipartite graphs by using a preferential rewiring process.

It is important in computer science, sociology, and so on to investigate complex bipartite graphs from a viewpoint of statistical physics. We propose a model to generate complex bipartite graphs without growing; the bipartite graphs are assumed to have two sets of the fixed numbers of nodes and a fixed number of edges between nodes belonging to different sets of nodes. In this model, essential ingredients are a preferential rewiring process and a fitness distribution function. By using the preferential rewiring process, we confirm that a bipartite graph reaches a stationary state after a sufficiently long time has passed. We find that the obtained bipartite graph has a scale-free-like property when a suitable fitness distribution is used. It turns out that a condensation of edges takes place in the cases of certain fitness distributions.