The Use of Machine Learning for the Purpose of Combating Bank Fraud

Abstract

Machine learning is a specific field of artificial intelligence that allows an information system to learn and improve itself autonomously using neural networks and deep learning, without the need for explicit programming, by ingesting large amounts of data. With the deepening of banking activity at the global level in cyberspace, the problems of a different kind, which are considered bank frauds, have become substantially more acute. Combating them required the examination of a gigantic volume of information, which as a task can only be achieved with artificial intelligence solutions, including the use of automatic learning technologies. Systems based on automatic learning are capable not only of analyzing the available information, but also of making certain predictions and self-learning conclusions, which allows not only the detection of existing forms of bank fraud, but to a certain extent to prevent it by detecting anomalies in customer behavior. At the moment, the global information industry has developed different forms of automatic learning, which can be adapted to the detection of fraud in different forms of banking activity. The development, implementation and further development of banking information systems based on automatic learning requires the completion of certain stages and the observance of mandatory rules.

Keywords

• machine learning
• bank
• customer
• fraud.