The paper “Incorporating machine learning and a risk-based strategy in an anti-money laundering multiagent system”, authored by Claudio Alexandre, LASIGE Ph.D. student, and João Balsa, LASIGE integrated researcher, was published in “Expert Systems and Applications”, a top-ranked journal (#6 in Artificial Intelligence, according to Google Scholar metrics).
In the context of the development of mechanisms to fight money laundering activities, the main goal of this paper is to describe a multiagent system that incorporates machine learning and risk components to identify and flag suspicious bank clients. The use of machine learning techniques to build relevant client profiles, combined with a novel mechanism for risk evaluation, is integrated into a system to be used by financial institutions.
Besides using standard metrics, the methodology was evaluated by experts from an actual bank and the quality of the results of this work has been attested. It is worth highlighting that all cases classified by the system as being “high-risk” suspects had not been previously reported. Besides, 76% of fully confirmed suspects had never been reported by any other system in use in the bank.
The paper is available here.