Адаптивный поведенческий профиль клиента коммерческого банка: концепция, формализация, методы построения

Андрей Александрович Сафонов; Виктор Петрович Чернов

doi:10.17308/meps/2078-9017/2026/7/28-37

Authors

Andrey Aleksandrovich Safonov Saint Petersburg State University of Economics
Viktor Petrovich Chernov Saint Petersburg State University of Economics

DOI:

https://doi.org/10.17308/meps/2078-9017/2026/7/28-37

Keywords:

customer behavioral profile, commercial bank, hyperbolic discounting, prospect theory, concept drift, gradient boosting, machine learning

Abstract

Importance: the article examines the behavioral profile of a commercial bank’s client, which is a set of characteristics that describe his financial preferences, his reactions to market incentives, and his digital footprints. Purpose: to formalize the concept of concept drift in relation to banking models of machine learning with the distinction between real and virtual types of drift, expressed in the change in the joint distribution of training data over time. Research design: the application of the quasi-hyperbolic specification of the Leibson model as a discounting model is substantiated. The model parameter is proposed as a feature estimated by the client’s transactional data and associated with the level of his financial stress. A four-level concept of the adaptive behavioral profile of the client is built on the basis of a combination of methodologies of neoclassical theory, behavioral economics and institutional constraints. Results: on the basis of the proposed concept, a three-component hybrid architecture of the model is substantiated, including clustering of the customer base, predictive models differentiated by business goals, and interpretation of forecasts.

Author Biographies

Andrey Aleksandrovich Safonov, Saint Petersburg State University of Economics

postgraduate student
Viktor Petrovich Chernov, Saint Petersburg State University of Economics

Dr. Sci. (Econ.), Full. Prof.

References

1. Ванина М.Ф., Ерохин А.Г. Применение методов машинного обучения в банковском скоринге // DSPA: Вопросы применения цифровой обработки сигналов, 2025, Т. 15, no. 1, c. 17-35.

2. Качурина А.С., Мостыка К.С. Методы анализа клиентской базы // Экономика и социум, 2016, no. 6-1 (25). Доступно: https://cyberleninka.ru/article/n/metodyanaliza-klientskoy-bazy-1 (дата обращения: 26.11.2025).

3. Соломко Д.С. Сегментация клиентской базы на основе RFM-анализа с применением методов машинного обучения / Д.С. Соломко, Д.А. Павлов // Информационное общество: современное состояние и перспективы развития: сборник материалов XIV международного форума, Краснодар, 12–17 июля 2021 года. Краснодар, Кубанский государственный аграрный университет имени И.Т. Трубилина, 2021, c. 405-409.

4. Шестернина М.В., Галиханов Р.Р. ABC-XYZ-анализ клиентов как основа управления клиентской базой предприятия // Телескоп, 2025, no. 1. Доступно: https://cyberleninka.ru/article/n/avsxyzanaliz-klientov-kak-osnova-upravleniyaklientskoy-bazoy-predpriyatiya (accessed: 08.02.2026).

5. Bartos Vojtech and Bauer Michal and Chytilová Julie and Levely Ian. Effects of Poverty on Impatience: Preferences or Inattention? (September 1, 2018). CERGEEI Working Paper Series, no. 623. Available at SSRN: http://dx.doi.org/10.2139/ ssrn.3247690 (accessed: 28.02.2026).

6. Esposito L., Mastromatteo G. Behavioral economics and the nature of neoclassical paradigm. Mind Soc 23, 45–78 (2024). https://doi.org/10.1007/s11299- 024-00303-y (accessed: 28.02.2026).

7. Frederick S., Loewenstein G., O'Donoghue T. Time Discounting and Time Preference: A Critical Review // Journal of Economic Literature, 2002, Vol. 40, no. 2, pp. 351-401.

8. Friedman J.H. Greedy Function Approximation: A Gradient Boosting Machine // The Annals of Statistics, 2001, Vol. 29, no. 5, pp. 1189-1232.

9. Kahneman D., Tversky A. Prospect Theory: An Analysis of Decision under Risk // Econometrica, 1979, Vol. 47, no. 2, pp. 263-291.

10. Laibson D. Golden Eggs and Hyperbolic Discounting // Quarterly Journal of Economics, 1997, Vol. 112, no. 2, pp. 443–478.

11. Lipman S.A., Boderie N.W., Been J.V., van Kippersluis H. Deposit? Yes, please! The effect of different modes of assigning reward – and deposit-based financial incentives on effort // Behavioural Public Policy, 2026, Vol. 10, no. 1, pp. 13-41.

12. Lu J., Liu A., Dong F., Gu F., Gama J., Zhang G. Learning under Concept Drift: A Review // IEEE Transactions on Knowledge and Data Engineering, 2019, Vol. 31, no. 12, pp. 2346–2363. Available at: https:// arxiv.org/pdf/2004.05785 (accessed: 28.02.2026).

13. Lundberg Scott & Lee Su-In. A Unified Approach to Interpreting Model Predictions, 2017.

14. Mohammad Zoynul Abedin, Petr Hajek, Taimur Sharif, Md. Shahriare Satu, Md. Imran Khan. Modelling bank customer behaviour using feature engineering and classification techniques. Research in International Business and Finance, Vol. 65, 2023, 101913. Available at: https:// doi.org/10.1016/j.ribaf.2023.101913 (accessed: 03.12.2025).

15. Widmer G., Kubat M. Learning in the Presence of Concept Drift and Hidden Contexts // Machine Learning, 1996, Vol. 23, pp. 69-101.