A Production-Collected Online Banking Fraud Detection Dataset from a Live Cloud-Based Deep Learning System

This dataset contains 56,962 online banking transactions collected through a live, cloud-deployed fraud detection system over a 30-day period in January 2026. The data were gathered through a REST API endpoint hosted on a Ubuntu Linux virtual private server, where a trained hybrid CNN-LSTM model processed each incoming transaction in real time. The dataset contains 98 confirmed fraudulent transactions (0.172% fraud rate). Each record includes 30 transaction features alongside model-generated outputs comprising fraud probability score, risk level, confidence estimate, action recommendation (BLOCK or ALLOW), and response latency in milliseconds.