A Systematic Review of Controlling Rumors in Online Social Networks: Insights From Epidemiological Models

Online social networks have transformed global communication, enabling instant interaction but also accelerating the spread of rumors and misinformation that threaten public trust. While much research targets rumor detection, strategies for controlling rumor diffusion remain limited. This study systematically reviews 62 peer-reviewed papers focusing on control-oriented epidemiological (compartmental) models to analyze and mitigate rumor propagation across six major databases: Web of Science, IEEE Xplore, ProQuest, ScienceDirect, Engineering Village, and ACM Digital Library. Results show the Susceptible–Infected–Recovered (SIR) model dominates, with 87.10% of studies adopting deterministic approaches. Approximately 35% consider heterogeneous networks, and 75.81% rely on synthetic datasets, which restricts real-world validation. Common control measures include education & behavioral interventions, media & public communication, often optimized via Pontryagin's Maximum Principle. The review emphasizes the need for stronger empirical validation and adaptive modeling to enhance rumor management and ensure information integrity.