Ben Hellmanzik, business informatics student at the University of Rostock, will defend his master's thesis on "Fraud detection in payment transactions: improvement potential for AI systems through knowledge sharing" on 13.10.2021 at 13:00.
The reviewers and supervisors were Dr. Anne Gutschmidt and Prof. Kurt Sandkuhl from the University of Rostock.
The defense will take place in a virtual ZOOM room. All interested parties please register by Wednesday, 13.10.2021, 08:00 a.m., to submit dial-in data by email to Prof. Kurt Sandkuhl (kurt.sandkuhluni-rostockde).
Abstract:
Fraud has caused billions of dollars in damages in recent years. Financial institutions, law enforcement agencies, and academic research have made it their mission to understand, detect, and prevent fraud. In current studies, artificial intelligence solutions, including machine learning and rule-based systems, are being used to detect fraud in a short time. Since current rule-based fraud detection methods use experts who develop features, for example, it makes sense to incorporate their insights into the development of machine learning methods. However, in order to make sense of these insights, an understanding of how these subject matter experts work must be developed, as well as of current developments in research and general fraud detection. This paper investigates the extent to which improving payment fraud detection by incorporating the knowledge of subject matter experts into AI-based solution is useful and feasible. It is based on a case study of a German financial institution.