Topology-traffic coupling: An ingredient for modelling weighted networks

Marc Barthelemy

Department of Physics, CEA, bp 12, 91680 Bruyeres-Le-Chatel Cedex, France

In addition to topological complexity, real-world networks display a gradation in the intensity strength between nodes-the weights of the links (or the "traffic" in the case of transportation networks). In particular, weighted networks can exhibit broad distributions and non-trivial correlations of weights that cannot be explained in terms of the underlying topological structure. These results call for the need of the modelling of complex networks which goes beyond purely topological models. I will present a model of weighted network formation which introduces a dynamical coupling between topology and weights by rearranging weights when a new link is introduced in the system. This simple model provides an explanation for most of the features observed in real-world networks and suggests that the topology-traffic coupling is an important ingredient for modelling complex weighted networks.