Currently, there is a significant increase in the number of incidents involving the use of bank cards, driven by the variety of fraud methods employed by malicious actors. Offenses in banking activities pose a substantial threat to the economic interests of modern society as a whole. The rise in fraudulent transactions using bank cards is explained by the growing number of bank cardholders, the popularization of online payments, and the transfer of salaries to cards. This article is dedicated to the development of machine learning methods to combat fraudulent transactions involving bank cards. The aim of the article is to develop a methodology for creating a fraud detection system that possesses high accuracy and adaptability concerning new fraudulent schemes. The article provides a description of the developed foundational solution for identifying cases of fraud with bank cards using machine learning techniques. The subject of the study is the mechanisms for implementing machine learning methods that arise in the process of preventing and detecting fraudulent transactions during financial operations with bank cards. The research methodology includes a set of complementary approaches. In the research process, general scientific and specialized methods were applied. Among the general scientific methods used were analysis, synthesis, generalization, and structural-logical analysis. The value of the research lies in the fact that it expands the understanding of the capabilities of machine learning in recognizing and combating fraudulent transactions with bank cards and demonstrates the effectiveness of modern artificial intelligence technologies in ensuring data protection. The study allows for a comprehensive overview of contemporary machine learning methods and technologies aimed at detecting and preventing fraudulent schemes. This is largely related to the ability to analyze large volumes of data, leading to the identification of patterns indicative of fraudulent activities. The developed methodology for detecting fraudulent transactions based on machine learning methods allows for targeted interventions in the most critical and vulnerable areas concerning financial operations with bank cards. For financial and credit organizations, the research provides value in justifying the need to adjust approaches to countering fraudulent operations in light of the emergence of new technologies and criminal schemes.
Maksim Vyacheslavovich Sokul' (Thu,) studied this question.