Anti-Money Laundering Tracker Using Graph-AI
The Anti-Money Laundering Tracker models transaction networks using the IBM AMLSim synthetic dataset, which includes approximately 9,945 account records and 68,289 transactions.
Product · 3 documents
The Anti-Money Laundering Tracker models transaction networks using the IBM AMLSim synthetic dataset, which includes approximately 9,945 account records and 68,289 transactions.
The paper introduces a multi-horizon reconciliation architecture that coordinates daily batch, hourly incremental partitioning, and real-time listener interception over a single transaction stream.
On a six-million-transaction synthetic dataset, the proposed ensemble attains a recall of 0.9184 at a false-positive rate of 0.0186.