Enhancing Anti-Money Laundering Systems with Machine Learning: A Comparative Analysis of Supervised Models

Abstract

Money laundering is a crucial worldwide problem which is posing consistent challenge to financial institutions and global economies, consenting illicit funds infiltrate legal financial systems and destabilize economic stability. By ascertaining unusual trends in massive amount of financial transactions, Machine learning (ML) is emerged as a dominant tool to address this challenge. This research discusses the employment of Random Forests and other state-of-the-art ML systems to effectually perceive money laundering actions. A broad development for feature engineering, model training, and performance valuation is defined in the work by a Kaggle dataset of anonymised bank transactions classified as either real or suspect. The Random Forest model's capability to categorize suspicious transactions is established by its excellent accuracy. By signifying the potential contribution of ML to the anticipation of financial crimes, these outcomes set the foundation for robust anti-money laundering scheme.