METHOD OF IMPROVING AML CHECKS IN THE FIELD OF CRYPTOCURRENCY USING BLOCKCHAIN ​​TECHNOLOGY AND MACHINE LEARNING

Abstract

The article investigates the problem of improving Anti-Money Laundering (AML) procedures in the cryptocurrency domain under conditions of rapid blockchain and WEB 3.0 development. The growing number of active crypto wallets and transactions in decentralized networks necessitates advanced approaches to detecting illegal financial activities, particularly money laundering operations. Traditional AML mechanisms, based on sequential transaction analysis and rule-based logic, are characterized by high computational costs and limited adaptability to emerging fraud scenarios. The aim of the study is to develop a method for enhancing AML verification by integrating blockchain technology and machine learning algorithms. The paper analyzes the current state of the AML service market, compares commercial providers, and identifies key pricing factors influencing transaction verification costs. A hypothesis is substantiated regarding the need to change the paradigm of initial transaction assessment: instead of assuming a transaction is safe by default, a model of conditional initial restriction with subsequent rating adjustment based on machine learning results is proposed. A mathematical model is introduced to formalize each verification stage: initial scoring, weighted parameter analysis, ML-based classification, dynamic rating update, system response, and final decision-making based on threshold values. To validate the proposed method, a theoretical experiment is described using a dataset of blockchain transactions and implementing Logistic Regression, Random Forest, and Neural Networks. The evaluation framework includes Accuracy, Precision, Recall, and F1-score metrics.The expected results demonstrate the potential to achieve high transaction classification accuracy (above 90%) while reducing false positive rates. The practical value of the research lies in decreasing computational resource consumption, increasing adaptability of AML systems, and strengthening cybersecurity in cryptocurrency financial operations.