The Big Ban Theory: A Pre- and Post-Intervention Dataset of Online Content Moderation

The Big Ban Theory (TBBT) is a large-scale dataset designed to support systematic research on the effects of online content moderation interventions. The dataset includes 25 moderation interventions of varying type, severity, and scope (e.g., community bans, community quarantines, community bans with migration, and post removals). TBBT comprises more than 38 million comments, collected from Reddit communities affected by moderation interventions. For each intervention, the dataset provides standardized metadata together with pseudonymized user activity data covering the three months preceding and following the enforcement of the intervention. This pre- and post-intervention design enables consistent, comparable, and reproducible analyses of behavioral changes associated with moderation interventions. This Dataset is described in the paper: Cerulli, A., Cima, L., Tessa, B., Tardelli, S., & Cresci, S. (2026). The Big Ban Theory: A Pre-and Post-Intervention Dataset of Online Content Moderation Actions. Proceedings of the International AAAI Conference on Web and Social Media. If you use this data, please cite the original paper. For additional details on the dataset structure and organization, please refer to the README file and the paper.