Evaluating Moderation in Online Social Network
Letizia Milli, Laura Pollacci, and Riccardo Guidotti

TL;DR
This paper introduces a simulation framework extending the SEIZ epidemic model to evaluate the effectiveness of generic and profile-aware moderation strategies in controlling toxic content spread on social networks.
Contribution
It presents a novel simulation-based approach incorporating user psychological profiles for personalized moderation, advancing the understanding of intervention impacts.
Findings
Profile-aware moderation significantly reduces toxicity spread.
Generic interventions are less effective than personalized strategies.
The framework allows systematic exploration of moderation parameters.
Abstract
The spread of toxic content on online platforms presents complex challenges that call for both theoretical insight and practical tools to test intervention strategies. In this novel research paper, we introduce a simulation-based framework that extends the classical SEIZ (Susceptible-Exposed-Infected-Skeptic) epidemic model to capture the dynamics of toxic message propagation. Our simulator incorporates active moderation mechanisms through two distinct variants: a basic moderator, which implements uniform, non-personalized interventions, and smart moderator, which leverages user-specific psychological profiles based on Dark Triad traits to apply personalized, threshold-driven moderation. By varying parameter configurations, the simulator allows for systematic exploration of how different moderation strategies influence user state transitions over time. Simulation results demonstrate…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsHate Speech and Cyberbullying Detection · Misinformation and Its Impacts · Spam and Phishing Detection
