An evolutionary game-theoretic modeling for heterogeneous information diffusion

In this paper, we model and analyze the information diffusion in heterogeneous social networks from an evolutionary game perspective. Users interact with each other according to their individual fitness, which are heterogeneous among different user types. We first study a model where in each social interaction the payoff of a user is independent of the type of the interacted user. In such a case, we derive the information diffusion dynamics of each type of users as well as that of the overall network. The evolutionarily stable states (ESSs) of the dynamics are determined accordingly. Afterwards, we investigate a more general model where in each interaction the payoff of a user depends on the type of the interacted user. We show that the local influence dynamics change much more quickly than the global strategy population dynamics and the former keeps track of the latter throughout the information diffusion process. Based on this observation, the global strategy population dynamics are derived. Finally, simulations are conducted to verify the theoretical results.