Fighting Money Laundering With Statistics and Machine Learning

Money laundering is a profound global problem. Nonetheless, there little scientific literature on statistical and machine learning methods for anti-money laundering. In this paper, we focus in banks provide an introduction review of the literature. We propose unifying terminology with two central elements: (i) client risk profiling (ii) suspicious behavior flagging. find that characterized by diagnostics, i.e., efforts to explain factors. On other hand, flagging non-disclosed features hand-crafted indices. Finally, discuss directions future research. One major challenge need more public data sets. This may potentially be addressed synthetic generation. Other possible research include semi-supervised deep learning, interpretability, fairness results.