Development of an Intelligent Module for Monitoring and Analysis of Client's Bank Transactions

Abstract

In the paper were analyzed the problems in the field of information security of payment systems based on plastic cards. Authorization data used for cashless payments can be compromised and used by intruders for unauthorized access to the user's bank account and conducting fraudulent transactions. To assess the main characteristics of the module being developed, an analytical review of the most well-known banking transaction risk control systems was conducted. A new method based on cluster analysis and the vectors method is proposed for identifying possible fraudulent operations using payment cards. Has been implemented an efficient algorithm for converting information from a user's payment profile to a string of numbers in an n-number system. To perform this conversion, special prepared in advance conversion maps are used. The results of completing the steps are illustrated by figures and tables. The correctness of the proposed methods and algorithms is confirmed by experiments conducted on real samples.