A study on multi-information diffusion model considering dual social reinforcement effect from the perspective of evolutionary game

The dynamic interactions of multi-information in online social networks present new challenges for understanding and forecasting information dissemination trends, especially the bounded rational decision-making of users when faced with various information. This article introduces evolutionary game theory to analyze user strategies amidst varying information. By using the Fermi function to calculate the imitation probability of netizens and considering the attractiveness of information and the dual social reinforcement effect, a class of G-SFDFRR multi-information delay propagation models has been established. The propagation threshold is calculated using the next-generation matrix method, and the global asymptotic stability of the system is analyzed using the time-delay Lyapunov function. Empirical analysis based on a Twitter dataset validated the model's effectiveness, showing an improvement in the fitting degree of 38.91 and 19.21 % over the SCIR and SICMR models, respectively. Further quantitative analysis through numerical calculations revealed that evolutionary games can delay the peak of information dissemination and accelerate the decline of rumors, highlighting the key role of strategic interaction in curbing rumor spread. Additionally, regulating the positive and negative social reinforcement effects and propagation probabilities can optimize the effect of rumor refutation, with short-term forced silence measures being more effective than long-term ones.