This web application is a visual dashboard for our global conflict modeling capstone project. The goal for this project was to build an optimal and explainable forecasting model that uses data from the Violence & Impacts Early-Warning System (VIEWS) to predict global conflict and potential fatalities.

This project is inspired by the “Fast forward: Forecasting global emerging threats” challenge from the Department of National Defence and the Canadian Armed Forces (DND/CAF). The challenge was for Canadian companies to develop precision forecasting models to help the DND/CAF predict the emergence of future conflicts around the world. Though this was initially intended to be the objective, the project was branched off into a separate process due to an incompatibility with the structure of the expectations in the capstone, with the scope, goals, and constraints being adjusted accordingly. Our client initially developed an Adaptive Markov Model for Violence Extrapolation and Rationalization (AMMVER) for this challenge, which will be used purely for inspiration.

This project was commissioned by our client, VIMY Analytics, a Canadian software company that specializes in explainable AI solutions for aerospace and defence contractors. Aerospace and defence requires a high level of certainty through mathematical rigor, so their solutions aim to prioritise precision, explainability, and reliability. Some of their current projects include their Flight-Level Adaptive Routing Engine (FLARE) and their Strategic High-Latitude Integrated Vessel Emergency Routing (SHIVER).

This web application has 4 main pages:

• Home Page: an introduction to the project and a brief explanation of its background and how to navigate the site. You are here!
• Map Page: an interactive heatmap of the world, with detailed prediction results for each country by year.
• Data Page: the raw result data in table form. Also available for download.
• Credits Page: the list of developers and a special thanks to our client and supervisor.