Scholarlink

The rapid spread of disinformation on social media platforms poses significant threats to public discourse and democratic institutions. This study introduces a novel fractal-fractional epidemiological model, SEDAZR, that partitions users into Susceptible, Exposed, Disinformed, Anti-disinformed, Skeptic, and Recovered compartments. Leveraging real-world datasets from Twitter, Telegram, and TikTok, the model captures memory-dependent dynamics and nonlinear user transitions observed in online environments. We derive the basic reproduction number \(\mathcal {R}_0\) and conduct sensitivity analysis using Latin Hypercube Sampling and Partial Rank Correlation to identify key parameters influencing disinformation spread. Theoretical validation is achieved through existence, uniqueness, and Ulam-Hyers stability analyses, confirming the model’s robustness under perturbations. Numerical simulations demonstrate the influence of memory effects and transmission rates on user behavior, while model fitting shows strong alignment with platform-specific data. This integrated framework offers practical insights for designing adaptive mitigation strategies and informs future extensions into multi-platform, demographically aware disinformation control systems.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Disinformation Contagion: Integrating Data-Driven Insights with Theoretical Model

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Nitin Agarwal

摘要

The rapid spread of disinformation on social media platforms poses significant threats to public discourse and democratic institutions. This study introduces a novel fractal-fractional epidemiological model, SEDAZR, that partitions users into Susceptible, Exposed, Disinformed, Anti-disinformed, Skeptic, and Recovered compartments. Leveraging real-world datasets from Twitter, Telegram, and TikTok, the model captures memory-dependent dynamics and nonlinear user transitions observed in online environments. We derive the basic reproduction number \(\mathcal {R}_0\) and conduct sensitivity analysis using Latin Hypercube Sampling and Partial Rank Correlation to identify key parameters influencing disinformation spread. Theoretical validation is achieved through existence, uniqueness, and Ulam-Hyers stability analyses, confirming the model’s robustness under perturbations. Numerical simulations demonstrate the influence of memory effects and transmission rates on user behavior, while model fitting shows strong alignment with platform-specific data. This integrated framework offers practical insights for designing adaptive mitigation strategies and informs future extensions into multi-platform, demographically aware disinformation control systems.