Information diffusion prediction in mobile social networks with hydrodynamic model

Abstract

Mobile social networks have gained tremendous popularity among hundreds of millions of Internet users due to their fast information spreading and strong inter-person influence. However, the high complexity of social interactions and the intrinsic dynamics of mobile social networks make it challenging to model the spreading mechanism delicately and enable precise prediction of information diffusion. In this paper, we are the first to exploit physical hydrodynamics to model the process of information diffusion in mobile social networks. With our proposed hydrodynamic information diffusion prediction model (hydro-IDP), we can accurately capture the information diffusion process from both temporal and spatial perspectives, and shed more light on the information spreading characteristics (e.g., information popularity, user influence, social platform diffusivity, etc.). We also conduct a large-scale trace-driven validation to verify the accuracy of our model. The results show that the hydro-IDP model is competent to characterize and predict the process of information propagation in mobile social networks.