UT Institute of Computer Science Graduation Theses Registry

Name

Kristjan Paavel

Abstract

Transaction monitoring is essential in financial systems to comply with regulatory standards and prevent financial crimes. This thesis develops a transaction labeling system for Wise, focusing on detecting and categorizing suspicious transactions to enhance anti-money laundering (AML) efforts. The proposed system improves Wise's operational processes and existing transaction monitoring by eventually utilizing dynamic and adaptable machine learning techniques to better identify and address new financial crime strategies. The implementation includes a new backend architecture and a frontend interface integrated with an existing service for efficient transaction labeling. While centered on AML requirements, the system is designed to be adaptable for use by other financial crime teams within Wise.
Keywords: Transaction monitoring, AML, machine learning, financial crimes, Wise

Graduation Thesis language

English

Graduation Thesis type

Bachelor - Computer Science

Supervisor(s)

Karl-Martin Miidu, Roshni Chakraborty

Defence year

2024