Optimal Control for Positive and Negative Information Diffusion Based on Game Theory in Online Social Networks.

In online social networks, information diffusion presents a complicated dynamic process that accompanies users' lives, in which the link between multi-information symbiosis and conflict is frequently overlooked. We investigate two aspects that influence the process of diffusion by examining the phenomena of information dissemination between individuals and the general environment in online social networks. To characterize that complicated process, an effective model of positive and negative information diffusion based on evolutionary game theory is developed. To reduce the impact of negative information on the network ecosystem as much as possible while increasing the propagation of positive information, real-time control mechanisms are provided, and a corresponding optimal problem of control resource allocation is constructed. Considering some special circumstances that cannot monitor the diffusion process in real time, the pulse control strategies are offered. The efficiency of control systems is validated by theoretical evidence and experiments, and the impact of various factors on information dispersion is examined. The work will contribute to network management by providing a theoretical framework for understanding and managing the distribution of diverse types of information.