Traffic flow prediction with big data: A learning approach based on SIS-complex networks

This paper proposes a susceptible-infected-susceptible complex networks (SIS-Complex Networks) model by the combination between SIS epidemic model and complex networks theory to predict the network traffic flow. It generalizes the SIS theory to the whole traffic network system. In order to establish it, we firstly obtain the traffic structure information from the OpenStreetMap and utilize the web spider to collect real-time traffic data. Then, in terms of the SIS-Complex networks model, we provide the quantitative description of congestion propagation between traffic structure and to predict the network traffic flow. Simulation results which show that the method proposed by this paper can successfully fit and predict the traffic flow are provided as well.