Network Analysis Enhancements to improve Anti Money Laundering Transaction Monitoring Systems in Banks

Financial crime, including fraud, corruption, and money laundering, poses a serious threat to the global financial system. To combat these risks, regulatory frameworks enforce Anti-Money Laundering (AML) and Combating the Financing of Terrorism (CFT) measures, requiring financial institutions to implement strict compliance controls. A key component of these efforts is Transaction Monitoring, which helps detecting and preventing suspicious activities. This thesis, developed in collaboration with the Models, Analytics, and Special Investigation office of Intesa Sanpaolo Group, proposes two proprietary algorithms to enhance transaction monitoring. By leveraging Network Analysis principles, these algorithms improve the detection of illicit activities and the quality of reports sent to the Italian Financial Intelligence Unit (UIF). The first algorithm analyzes individual customer activity in a novel manner, identifying suspicious transactions and counterparties based on velocity and fund rotation. The second applies machinelearning- based anomaly detection to large transaction volumes, uncovering unusual and possibly suspicious shifts in macroscopic financial behavior. Together, these approaches create a comprehensive framework for monitoring transactions at different levels of complexity, improving financial crime prevention within the Bank.