Home // SOTICS 2025, The Fifteenth International Conference on Social Media Technologies, Communication, and Informatics // View article

Modeling Quarantine Intervention for Varied Toxic Intensities

Authors:
Nitin Agarwal

Keywords: toxicity; social media; epidemiological modeling; hate speech; COVID-19; Brazil; Per

Abstract:
This study employs a Susceptible-Exposed-Infected-Quarantine-Recovered (SEIQR) epidemiological framework to analyze the spread of toxicity in online environments, integrating toxicity intensity stratification to capture the complexity of toxicity propagation. Using datasets from coronavirus disease of 2019 (COVID-19) and social movement discussions, we conduct sensitivity analysis to evaluate key parameters influencing toxicity diffusion. The results reveal that splitting toxicity into moderate and high levels significantly reduces model error rates, enhancing predictive accuracy across all datasets. Additionally, our findings indicate that the basic reproduction number ($R_0$) is highly sensitive to exposure and quarantine rates, emphasizing the critical role of enhanced moderation and adaptive quarantining in suppressing toxicity. Moreover, quarantine interventions and content demotion strategies are shown to significantly curb toxicity intensity while maintaining engagement dynamics. These insights provide a foundation for policy-driven interventions, enabling social media platforms to implement optimized content moderation, algorithmic intervention, and network-level strategies to mitigate online toxicity and promote healthier digital ecosystems.

Pages: 1 to 6

Copyright: Copyright (c) IARIA, 2025

Publication date: September 28, 2025

Published in: conference

ISSN: 2326-9294

ISBN: 978-1-68558-301-9

Location: Lisbon, Portugal

Dates: from September 28, 2025 to October 2, 2025