DOI: 10.17587/prin.13.247-256

Building the Scale for Fraud Detection on the Internet Using ML

L. V. Zhukova^1,4, lvzhukova@hse.ru, I. M. Kovalchuk^2,4, ikovalchuk@ec-leasing.ru, A. A. Kochnev ^3,4, akochnev@ec-leasing.ru, V. R. Chugunov⁴, vchugunov@ec-leasing.ru,
¹ National Research University Higher School of Economics, Moscow, 101000, Russian Federation,
² MIREA — Russian Technological University, Moscow, Russian Federation,
³ Plekhanov Russian University of Economics, Moscow, Russian Federation,
⁴ JSC "EC-LEASING" CO., Moscow, 117587, Russian Federation

Corresponding author: Zhukova Ludmila V., Senior Lecturer, National Research University Higher School of Economics, Moscow, 101000, Russian Federation, Leading Analysist, JSC "EC-LEASING" Co., Moscow, 117587, Russian Federation, E-mail: lvzhukova@hse.ru

The widespread digitalization of the modern society and the development of information technology have increased the number of methods of interaction between financial institutions and potential consumers of financial services. At the same time, the advent of new financial products inevitably leads to an increase in threats, and the use of information technology facilitates the continuous "improvement" of fraudulent schemes and unfair services that negatively impact both the financial market as a whole and its individual participants, such as financial institutions and their clients. Due to the development of modern society, most financial transactions have moved to the Internet, including the fraudulent ones. When services are provided remotely, it is more difficult to trace and prosecute the beneficiary, but there are still ways to stop fraudulent activity. They can be characterized as labour-consuming, as the monitoring and analysis of huge amounts of unstructured information (BigData) located on the Internet take great amount of time and effort. The solution to detecting illegal activity in the financial market is based on open data intelligence, the application of machine learning methods and systems analysis techniques. The article examines the types of financial services provided on the Internet, among which fraudulent activities are most common. In order to identify illegal financial services, criteria are identified and grouped according to their contribution to the decision-making process. The main result of the study is the construction of the scale of a complex indicator, which is used to develop a mathematical model based on the selected criteria and machine learning methods to identify the extent of illegality of financial services provided on the Internet.

Keywords: to unfair practices, financial services, Internet fraud, machine learning, system analysis, mathematical model, big data, complex indicator, open data analysis, Internet monitoring

pp. 247–256

For citation:
Zhukova L. V., Kovalchuk I. M., Kochnev A. A., Chugunov V. R. Building the Scale for Fraud Detection on the Internet Using ML, Programmnaya Ingeneria, 2022, vol. 13, no. 5, pp. 247—256.