Crafting peace and building a safer world
Ensuring trust and safety in the digital world

Machine Learning for Peace

University of Pennsylvania

Recent advances in technology have endowed governments with new repressive tools and even the ability to anticipate citizen action and engage in preemptive response. The Machine Learning for Peace Project (MLP) aims to level the playing field by equipping policymakers and civil society actors working to defend democracy with new tools to navigate increasingly sophisticated approaches to repression. MLP provides policymakers and practitioners with high-frequency, up-to-date data on the occurrence of 32 types of important political events in more than 35 countries around the world. Furthermore, these data are combined with machine learning tools to generate monthly forecasts of future changes in political conditions. A website allows users to interact with these data and view monthly forecasts about future events that can be used to anticipate and plan for changing political conditions.