APPLICATION OF MACHINE LEARNING AND DEEP LEARNING TECHNOLOGIES IN ANTI-MONEY LAUNDERING FOR CRYPTOCURRENCIES: A STUDY ON THE BITCOIN ELLIPTIC DATASET

Abstract

This study explores the application of deep learning and machine learning technologies in the field of Anti-Money Laundering (AML) for cryptocurrencies.With the rapid growth of cryptocurrency markets, the associated money laundering activities have increasingly become a focal point for governments and financial institutions worldwide.Traditional AML measures face challenges in the digital realm, particularly in identifying and preventing illicit transactions involving cryptocurrencies.To address this, the study designs various algorithms including Deep Neural Networks (DNN), Random Forest (RF), K-Nearest Neighbors (KNN), and Naive Bayes (NB) to enhance the detection capabilities of suspicious transactions within the Bitcoin Elliptic dataset.Cryptocurrencies involve using cryptographic security measures for financial transactions, yet their anonymity and transnational nature make them susceptible to money laundering activities.By evaluating the performance of different machine learning models on the Bitcoin Elliptic dataset, this research analyzes their effectiveness in identifying illicit transactions.The results indicate that the Random Forest model performs best, achieving an overall accuracy of 95%, effectively distinguishing between most illegal and legal transactions while mitigating overfitting risks.Through these technological approaches, the study aims to enhance AML monitoring capabilities in cryptocurrency markets, providing reliable decision support for financial institutions and regulatory bodies.Future research directions may include exploring more complex deep learning models or ensemble learning methods to further improve classification accuracy across diverse datasets and enable real-time monitoring of emerging money laundering patterns.The integration of these technologies holds promise for strengthening the global AML framework, addressing the increasingly complex challenges posed by digital finance and illicit financial activities (