RISK ASSESSMENT OF THE ANTI-MONEY LAUNDERING AND CYBER SECURITY SYSTEMS’ CONVERGENCE

Abstract

The growth of financial and cyber threats leads to the most significant losses in the financial sector. Only complex approaches can be the most effective to counteract them, which require processes of systems' convergence of financial monitoring and cyber security. Therefore, this article is devoted to determining the convergence risk for different countries. The study proposes a scientific and methodological approach to its evaluation, which involves implementing four stages. The empirical database was formed by the National Cyber Security Index and the Anti-Money Laundering Index for 114 countries in 2022. The first index characterizes the level of development of the country's cyber security system. The second indicator reflects the degree of criminal proceeds legalization risk. Calculations were made using the Python programming language. In the first stage, countries were clustered according to the money laundering risk performed using such methods as "Silhouette analysis" and "K-means" clustering. As a result, seven clusters were obtained, which make it possible to identify countries by the possibilities of criminal income legalization. In the second stage, a similar procedure was carried out to obtain countries' segments regarding the level of their cyber security. As a result, six clusters were obtained, which allowed identifying the level of development of the cyber threat countermeasure system. An integral convergence index was proposed and calculated in the third stage using normalization and geometric mean methods. Based on the cluster analysis results, nine risk groups of the systems' convergence for countering financial and cyber risks were determined for different countries. The fourth stage was devoted to developing a predictive model of convergence risk based on a classification decision tree. The quality of the model turned out to be high, although, for the second classification group, the model will not be able to make the correct prediction. The proposed approach is of practical importance for improving countries' strategies for combating financial and cybercrimes.