E-SF2I: Dynamics modeling and multi-level verification of Online Social Network rumor propagation driven by emotional differentiation and dual-immunity-state coupling

Abstract

The open architecture of Online Social Networks (OSNs) facilitates information sharing but also exacerbates the risk of rumor propagation. In response to the challenge that existing propagation models struggle to quantify the synergistic effects of user emotional polarization and temporary neutral behavior, this paper proposes a novel E-SF2I (Emotion-Susceptible-Forwarding-Temporary Immunity-Permanent Immunity) rumor propagation dynamics model. In terms of theoretical modeling, a differential dynamic system with time-varying characteristics is constructed by coupling the emotion-driven forwarding mechanism with dual immunity state transitions. Based on the Lyapunov stability theory, the system equilibrium and its global asymptotic stability are rigorously demonstrated. A rumor propagation simulation algorithm based on the E-SF2I framework and a multi-level verification are further developed to verify the practical application value of the proposed model. Initially, the theoretical results derived from the model are validated using real network structure datasets (Facebook, Twitter, P2P). Subsequently, the validity of the proposed model is tested using 13 real rumor events on the Twitter dataset. Finally, the proposed model is compared with traditional and new models for three representative real rumor events. The results show that the proposed model achieves higher accuracy in predicting the propagation trends of rumor events.