Predicting the likelihood of systemic financial crises in 13 African countries using historical data from 1860–2014. The dataset includes indicators like inflation rates, banking and debt crises. Machine learning is used to model crisis risk based on these economic features.