The end of the Cold War and the rise of non-state actors on the international stage has heralded resurgence of conflicts around the world. We believe that the ready availability of online news sources, as well as advances in information extraction, statistical machine learning and stochastic modeling can be harnessed to predict the outbreak of serious international conflicts. The goals of our project are
- Event Extraction: Identifying relevant news stories and extracting who did what to whom and when,
- Conflict Prediction: Using the extracted events data for developing conflict prediction algorithms,
- Terror Networks: Use events data in conjunction with other specialized knowledge collections (e.g., START and ISVG) to build models of the evolution of terrorist networks around the world.