employee
student
student
The article discusses the features of a machine learning model that helps banks and other credit institutions to more accurately assess credit risk. The scientific article reflects the problems of classical information processing methods, which are increasingly unable to cope with the growing volume of data and rapid changes in customer behavior in today's reality. The authors of the article explore the question: "Why are banks starting to switch to more flexible algorithms, and what prevents them from using AI without restrictions?" The article presents conclusions on the use of a modern machine learning model that includes a new generation of credit risk management tools. This area allows the banking sector to adapt flexibly to the increasingly complex financial environment and the rapidly growing demands of the market system. Based on research and practical examples, the article shows which models are most commonly used and how they affect the performance of monetary analysis.
credit risk, customer financial behavior, traditional scoring approaches, logistic regression, gradient boosting, neural networks, regulatory authorities
1. Abramov A. A., Mel'nikova T. E. Primenenie metodov mashinnogo obucheniya v kreditnom skoringe // Finansy i kredit. — 2021. — № 8. — S. 34–47.
2. Evstigneev E. N. Bol'shie dannye i ih rol' v upravlenii kreditnymi riskami // Den'gi i kredit. — 2020. — № 5. — S. 52–63.
3. Polyakov P. V. Sovremennye metody kreditnogo skoringa: ot logisticheskoy regressii k ansamblevym modelyam // Bankovskoe delo. — 2022. — № 4. — S. 11–21.
4. Bank Rossii. Metodicheskie rekomendacii po primeneniyu modeley mashinnogo obucheniya pri upravlenii riskami. — M.: Central'nyy bank Rossiyskoy Federacii, 2021. — 48 s.
5. Savin A. Yu. Generativnye modeli v bankovskoy analitike: vozmozhnosti i ogranicheniya // Prikladnaya informatika. — 2023. — T. 18, № 2. — S. 90–103.
6. World Bank. Digital Credit Scoring and Financial Inclusion. — Washington, DC: World Bank Publications, 2020.
7. Gordeev I. N. Algoritmicheskaya spravedlivost' i riski diskriminacii v kreditnom skoringe // Ekonomicheskaya politika. — 2022. — T. 17, № 3. — S. 145–162.
8. Basel Committee on Banking Supervision. Principles for the Use of Artificial Intelligence Models in Banking. — Basel: BIS Publications, 2021. — 34 p.
9. Chang V., Abdulkareem K., Althubiti S. Credit Risk Prediction Using Machine Learning // Electronics. — 2024. — Vol. 13, No. 2. — P. 1–15.
