Machine Learning Approach to Anti-Money Laundering: A Review

For decades, money laundering has been a global issue, posing a severe threat to economies and societies. Despite the fact that billions of dollars in fines from authorities and the dismantling of illegal businesses continue to make headlines, government, regulatory, and financial bodies are all fighting the problem in their own ways. Traditional techniques of detecting and monitoring unlawful activities, while improved to some extent, are insufficient to keep up with the rising pace of criminal activity and the volume of data and transactions that flow around the world. Machine Learning techniques for anti-money laundering implementation have already been offered, however they are still in the early stages of development. Deep Learning approaches have been discovered to work even better with the enormous datasets that have been provided. This paper attempts to highlight and compare some of the traditional and machine learning based techniques that have been used and proposed to curb money laundering.