Project Description
The gradual drawdown of US operations in Afghanistan and Iraq have led the world to a degree of complacency about terrorism. Yet, the threat is omnipresent as we are seeing continuing terror attacks in Africa via groups such as Al-Shebaab, Boko Haram, and Jama’at Nusrat al-Islam wal-Muslimin (JNIM), in Asia via groups such as Lashkar-e-Taiba and Abu Sayyaf, and others.
Counterterrorism organizations need the power to be able to predict the approximate time frames and types of attacks that terror groups might carry out in the coming months. Armed with such knowledge, they can better direct intelligence collection resources, allocate counter-terrorism resources, and decide on optimal security strategies.
NTEWS, the Northwestern Terror Early Warning System, is an AI-driven platform that can make predictions about the types of attacks that terror groups will carry out in the coming 6 months. By reducing all data about a terror group to a table that is intelligible to machine learning algorithms, NTEWS can automatically learn predictive models and explain them in simple, easy to understand terms. Using a set of eight machine learning models, each fine tuned to a group and a specific type of attack, NTEWS can predict the likelihood of that type of attack by that group sometime in the next month or the next 2 months, 3 months, 4 months, 5 months, or 6 months. Types of attacks that NTEWS tries to predict include attacks on security installations, on security forces, on public sites (e.g. markets), abductions, and more. We emphasize that NTEWS doesn’t predict attacks on specific locations (e.g. a specific security installation) or an exact date. NTEWS can also explain many of its predictions via simple temporal probabilistic rules or tp-rules of the form “If condition C is true during the current month, then we can expect attacks of type T by group G sometime in the next 1,2,3,4,5, or 6 months.” When the precondition C is true in the current month, the probability of attacks of type T is high, but when it is false, the probability is very low.
To date, NTEWS has learned models and can make predictions about six groups: Abu Sayyaf, Lashkar-e-Taiba, Indian Mujahideen, Al-Shabaab, Boko Haram, and Jama’at Nusrat al-Islam wal-Muslimin (JNIM). We are pleased to announce the release of monthly NTEWS Terror Forecast reports on these six groups as of October 17 2024. We plan to release a report during the middle of each month.
If you wish to receive these reports, please send an email with your name, organization, and contact information to: vss@northwestern.edu