As fundamental changes in information systems drive digitalization, the heavy reliance on computers today significantly increases the risk of fraud. Existing literature promotes machine learning as a potential solution approach for the problem of fraud detection as it is able able to detect patterns in large datasets efficiently. However, there is a lack of clarity and awareness on which components and functionalities of machine learning-based fraud detection systems exist and how these systems can be classified consistently. We draw on 54 identified relevant machine learning-based fraud detection systems to address this research gap and develop a taxonomic scheme. By deriving three archetypes of machine learning-based fraud detection systems, the taxonomy paves the way for research and practice to understand and advance fraud detection knowledge to combat fraud and abuse.
Matschak, Tizian; Trang, Simon; and Prinz, Christoph, "A TAXONOMY OF MACHINE LEARNING-BASED FRAUD DETECTION SYSTEMS" (2022). ECIS 2022 Research Papers. 173.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.