Conflict-Event Prediction for UN Peacekeeping

In 2021, the CSS and the UN Operations and Crisis Centre (UNOCC) launched a joint research project to explore the feasibility of an event-prediction system for UN peacekeeping. The project involves a close collaboration between domain experts from the CSS, machine-learning scholars from ETH’s technical departments (including the Department of Computer Science), and practitioners working at the UNOCC. By combining machine learning with event records collected by UN peacekeepers, the project aims to lay the foundation for a software platform that can support data processing and decision-making procedures within UN peace missions.

Currently, the project focusses on the application of deep learning to conflict-event data. This is a novel research area with a limited body of existing work within the community of machine-learning scholars. As such, the project’s ongoing research has a foundational character. The aim is to understand how the advantages that deep neural networks have shown in other application domains can be ‘carried over’ to conflict-event data. The project’s long-term vision is to develop a collaborative Human-AI platform in which different machine-learning models (including deep neural networks) and a diverse pool of human experts can jointly make event forecasts for the same conflict zones. Adopting such a ‘human-in-the-loop’ approach is a shared goal for both the ETH team and the UNOCC.

This project is one element of the new partnership between the UN and ETH Zurich. It has also been mentioned by the Swiss Permanent Representative, Pascale Baeriswyl, in the UN Security Council in August 2023: “The [ETH] is developing a prototype of an AI-assisted analysis tool for the [UNOCC] which could explore AI’s potential for peacekeeping, particularly for the protection of civilians and peacekeepers”.

ETH PIs: Dr. Sascha Langenbach (CSS), Prof. Andreas Wenger (CSS), Prof. Menna El-Assady (Department of Computer Science)

Project members & partners: Francesco Re (ETH), Prof. Allard Duursma (ETH), Jan-Nico Zaech (ETH), Kersten Jauer (UN Complex Risk Analytics Fund), George Khoury (UN Operations and Crisis Centre), Rajkumar Cheney Krishnan (UN Operations and Crisis Centre)