Automatic machine learning algorithms for fraud detection in digital payment systems

dc.contributor.authorKolodiziev, O.en
dc.contributor.authorMints, A. Yu.en
dc.contributor.authorSidelov, P.en
dc.contributor.authorPleskun, I.en
dc.contributor.authorLozynska, O.en
dc.contributor.authorМінц, О. Ю.uk
dc.date.accessioned2023-04-17T20:29:13Z
dc.date.available2023-04-17T20:29:13Z
dc.date.issued2020
dc.description.abstractData on global financial statistics demonstrate that total losses from fraudulent transactions around the world are constantly growing. The issue of pay-ment fraud will be exacerbated by the digitalization of economic relations, in particular the introduction by banks of the concept of "Bank-as-a-Service", which will increase the burden on payment services. The aim of this study is to synthesize effective models for detecting fraud in digital payment sys-tems using automated machine learning and Big Data analysis algorithms.Approaches to expanding the information base to detect fraudulent transactions have been proposed and systematized. The choice of performance metrics for building and comparing models has been substan-tiated.The use of automatic machine learning algorithms has been proposed to resolve the issue, which makes it possible in a short time to go through a large num-ber of variants of models, their ensembles, and input data sets. As a result, our experiments allowed us to obtain the quality of classification based on the AUC metric at the level of 0.977‒0.982. This exceeds the effectiveness of the classifiers developed by tradition-al methods, even as the time spent on the synthesis of the models is much less and measured in hours. The models' ensemble has made it possible to detect up to 85.7 % of fraudulent transactions in the sample. The accuracy of fraud detection is also high (79‒85 %).The results of our study confirm the effectiveness of using automatic machine learning algorithms to synthesize fraud detection models in digital payment systems. In this case, efficiency is manifested not only by the resulting classifiers' quality but also by the reduction in the cost of their development, as well as by the high potential of interpretability. Implementing the study results could enable financial institutions to reduce the financial and temporal costs of developing and updating active systems against payment fraud, as well as improve the effectiveness of monitoring financial transactions.en
dc.identifier.citationKolodiziev, O., Mints, A., Sidelov, P., Pleskun, I., & Lozynska, O. (2020). Automatic machine learning algorithms for fraud detection in digital payment systems. Eastern-European Journal of Enterprise Technologies, 5(9 (107), 14–26. doi:https://doi.org/10.15587/1729-4061.2020.212830.en
dc.identifier.citationKolodiziev O., Mints A., Sidelov P., Pleskun I., Lozynska O. Automatic machine learning algorithms for fraud detection in digital payment systems. Eastern-European Journal of Enterprise Technologies. 2020. Vol. 5. № 9(107). P. 14–26. DOI:https://doi.org/10.15587/1729-4061.2020.212830.en
dc.identifier.doihttps://doi.org/10.15587/1729-4061.2020.212830
dc.identifier.issn1729-3774
dc.identifier.orcidhttps://orcid.org/0000-0002-8032-005X
dc.identifier.urihttps://dspace.mipolytech.education/handle/mip/109
dc.language.isoenen
dc.publisherPC TECHNOLOGY CENTERen
dc.relation.ispartofEastern-European Journal of Enterprise Technologies. Vol. 5. № 9(107). P. 14–26.en
dc.subjectdigital paymentsen
dc.subjectmachine learningen
dc.subjectautomated synthesisen
dc.subjectfraud detectionen
dc.subjectdata scienceen
dc.titleAutomatic machine learning algorithms for fraud detection in digital payment systemsen
dc.typeArticle

Файли

Контейнер файлів

Зараз показуємо 1 - 1 з 1
Ескіз
Назва:
Automatic machine learning algorithms for fraud detection in digital payment systems.pdf
Розмір:
791.36 KB
Формат:
Adobe Portable Document Format

Ліцензійна угода

Зараз показуємо 1 - 1 з 1
Ескіз недоступний
Назва:
license.txt
Розмір:
10.29 KB
Формат:
Item-specific license agreed to upon submission
Опис: