UKRLOGOS Group Open Repository

USING ARTIFICIAL INTELLIGENCE FOR SCORING AND CREDITWORTHINESS EVALUATION OF CLIENTS IN DIGITAL BANKING

Maksym Zdorovyi; Oleksandr Horokh

Conference proceedings: Raccolta di articoli scientifici «ΛΌГOΣ» con gli atti della VIII Conferenza scientifica e pratica internazionale «Ricerche scientifiche e metodi della loro realizzazione: esperienza mondiale e realtà domestiche» (Bologna, Repubblica Italiana; 19 dicembre, 2025)

Section: Finance and Banking; Taxation, Accounting and Auditing

Publication date: 2025/12/19

Pages: 32-41

DOI: 10.36074/logos-19.12.2025.005

ISBN: 978-617-8440-86-2

Publisher: Associazione Italiana di Storia Urbana

Language: en

PDF for indexing Original PDF in OJS archive DOI

Abstract

Credit scoring is the process by which banks and lenders evaluate a borrower’s creditworthiness – essentially predicting the risk that a customer will default on a loan obligation. Traditionally, credit scoring has relied on statistical models and expert-designed scorecards using a limited set of financial variables (such as income, outstanding debts, and past repayment history). These traditional models, most commonly logistic regression-based scorecards, became industry-standard because of their interpretability and regulatory acceptance. In essence, a simple weighted sum of factors (e.g. debt-to-income ratio, credit history length, etc.) produces a score, and the simplicity of such models makes it easy for risk managers and regulators to understand how each factor influences the decision. Over decades, methods like linear discriminant analysis and logistic regression proved effective and were relatively easy to implement and explain. However, these conventional approaches have limitations in predictive power because they assume a linear or simple relationship between inputs and credit risk. This has opened the door for more complex approaches.

Author affiliations

References

  1. Стрюк А.М. & Рассовицька М.В. (2014) Система хмаро орієнтованих засобів навчання як елемент інформаційного освітньо-наукового середовища ВНЗ. Інформаційні технології і засоби навчання, (4), 150–158. Вилучено з: http://journal.iitta.gov.ua/index.php/itlt/article/view/1087/829.
  2. Bank of England & Financial Conduct Authority. (2024). Artificial intelligence in UK financial services – 2024. Bank of England Report. Retrieved from https://www.bankofengland.co.uk/report/2024/artificial-intelligence-in-uk-financial-services-2024 (accessed Nov 18, 2025).
  3. Bank of Italy. (2022). Artificial intelligence in credit scoring: an analysis of some experiences in the Italian financial system (Occasional Paper No. 721). Rome: Bank of Italy. Available at SSRN: http://dx.doi.org/10.2139/ssrn.4462973 (accessed Nov 18, 2025).
  4. Bonaccorsi di Patti, E., Calabresi, F., De Varti, B., Federico, F., Affinito, M., & Antolini, M. et al. (2022). Artificial intelligence in credit scoring. An analysis of some experiences in the Italian financial system. Bank of Italy Occasional Papers, 721. Retrieved from https://www.bancaditalia.it/pubblicazioni/qef/2022-0721/index.html (accessed Nov 18, 2025).
  5. Consumer Financial Protection Bureau. (2019, August 6). An update on credit access and the Bureau’s first No-Action Letter. (Blog post by P. A. Ficklin & P. Watkins). Washington, DC: CFPB. Retrieved from https://www.consumerfinance.gov/about-us/blog/update-credit-access-and-no-action-letter/ (accessed Nov 18, 2025).
  6. MathWorks. (n.d.). Interpretability and explainability for credit scoring. MATLAB Risk Management Toolbox Documentation. Retrieved from https://www.mathworks.com/help/risk/interpretability-and-explainability-for-credit-scoring.html (accessed Nov 18, 2025).
  7. McKinsey & Company. (2022, August 9). Making financial services available to the masses through AI. (Interview with A. Wang, CFO of WeBank, by V. Chung). Retrieved from https://www.mckinsey.com/industries/financial-services/our-insights/making-financial-services-available-to-the-masses-through-ai (accessed Nov 18, 2025).
  8. Roberts, T. (2019). Credit Scoring Approaches: Guidelines.
  9. Washington, DC: The World Bank Group. Retrieved from https://thedocs.worldbank.org/en/doc/935891585869698451-0130022020/original/CREDITSCORINGAPPROACHESGUIDELINESFINALWEB.pdf (accessed Nov 18, 2025).
  10. Upstart Network Inc. (2019). Update on CFPB No-Action Letter – Access to Credit findings. Upstart Company News. Retrieved from https://www.upstart.com/news/an-update-from-cfpb-on-upstarts-no-action-letter (accessed Nov 18, 2025).
  11. World Bank & CGAP. (2018). Financial consumer protection
  12. and new forms of data processing beyond credit reporting. World
  13. Bank Discussion Note. Available at World Bank Documents: http://documents.worldbank.org/curated/en/403611493134249446/pdf/WPS8040.pdf.
  14. World Bank Group. (2019). The role of credit reporting in supporting financial regulationt. Washington, DC: World Bank. Available at https://documents1.worldbank.org/curated/en/262691559115855583/pdf/Credit-Reporting-Knowledge-Guide-2019.pdf.