Explainable Machine Learning for Road Accident Severity Prediction

Initial public release of the Road Accident Severity XAI framework. This release includes: Cross-country road accident severity prediction using harmonized UK, France, and Ethiopia data. Cost-sensitive multiclass modeling, external validation, domain-shift analysis, and SHAP explainability. An evidence-to-action road-safety intervention-priority framework. Reproducible notebooks, curated figures, tables, and dataset-source documentation. Raw and processed datasets are not included. Please consult DATA_SOURCES.md and data/README.md for access instructions and reuse terms. If you use this software, please cite it using the metadata below.