This dataset contains 2,500 synthetic, high-fidelity cross-border e-commerce chargeback events.
It is engineered specifically for Lead Data Scientists, Risk Engineers, and Fintech startups training machine learning models for anomaly detection, friendly fraud classification, and chargeback win-probability scoring.
