RAQAMLI IQTISODIYOT SHAROITIDA O‘ZBEKISTON TIJORAT BANKLARIDA KREDIT SKORING MODELLARINING TUSHUNTIRILUVCHANLIGI VA MODEL RISKINI BOSHQARISHNI TAKOMILLASHTIRISH
Keywords:
Keywords: Kalit so‘zlar: raqamli iqtisodiyot, raqamli bank, kredit skoringi, tushuntiriladigan sun’iy intellekt, model riski, SHAP, model reyestri, raqamli boshqaruv.Abstract
Annotatsiya. Tezisda raqamli iqtisodiyot sharoitida O‘zbekiston tijorat banklarida sun’iy intellektga asoslangan kredit skoring modellarining tushuntiriluvchanligi va model riskini boshqarishni yagona raqamli hayotiy siklga birlashtirish masalasi tadqiq etiladi. Ilmiy adabiyotlar va rasmiy bank statistikasi asosida ma’lumotlar lineage’i, model pasporti va versiyalar reyestri, SHAP asosidagi izohlar, mustaqil validatsiya, drift monitoringi hamda inson nazoratini qamrab oluvchi raqamli boshqaruv konturi taklif etiladi. Uning samarasi bashorat aniqligi bilan birga modelni joriy etish tezligi, monitoring xarajatlari, auditga tayyorlik va raqamli kredit qarorlarining shaffofligi orqali baholanadi.
References
1.Ayari H., Guetari R., Kraiem N. Machine learning powered financial credit scoring: a systematic literature review // Artificial Intelligence Review. 2026. Vol. 59. Art. 13. DOI: 10.1007/s10462-025-11416-2.
2.Basel Committee on Banking Supervision. Newsletter on Artificial Intelligence and Machine Learning. BCBS Newsletter 27. Basel: BIS, 2022.
3.Bussmann N., Giudici P., Marinelli D., Papenbrock J. Explainable AI in Fintech Risk Management // Frontiers in Artificial Intelligence. 2020. Vol. 3. Art. 26. DOI: 10.3389/frai.2020.00026.
4.Central Bank of the Republic of Uzbekistan. Information on major indicators of commercial banks as of July 1, 2026. Tashkent, 2026. URL: https://cbu.uz/en/statistics/bankstats/4230544/ (accessed: 13.09.2026).
5.European Banking Authority. Follow-up Report on the Use of Machine Learning for Internal Ratings-Based Models. Paris, 2023.
6.Han S., Jung H. NATE: Non-pArameTric approach for Explainable credit scoring on imbalanced class // PLOS ONE. 2024. Vol. 19(12). e0316454. DOI: 10.1371/journal.pone.0316454.
7.Hlongwane R., Ramabao K., Mongwe W. A novel framework for enhancing transparency in credit scoring: Leveraging Shapley values for interpretable credit scorecards // PLOS ONE. 2024. Vol. 19(8). e0308718. DOI: 10.1371/journal.pone.0308718.
8.Lundberg S.M., Lee S.-I. A Unified Approach to Interpreting Model Predictions // Advances in Neural Information Processing Systems. 2017. Vol. 30. P. 4765-4774.
9.Perez-Cruz F., Prenio J., Restoy F., Yong J. Managing explanations: how regulators can address AI explainability // FSI Occasional Papers. 2025. No. 24. Basel: BIS.
10.Shams Khoozani Z., Seera M., Lim C.P. Towards transparent financial AI: a systematic review of graph learning and explainable methods for credit risk and fraud detection // Artificial Intelligence Review. 2026. DOI: 10.1007/s10462-026-11645-z.
11.O‘zbekiston Respublikasi Prezidentining 2024-yil 14-oktabrdagi PQ-358-son qarori. “Sun’iy intellekt texnologiyalarini 2030-yilga qadar rivojlantirish strategiyasini tasdiqlash to‘g‘risida”. – Qonunchilik ma’lumotlari milliy bazasi, 2024. – URL: https://lex.uz/docs/-7158604.