A COMPARATIVE STUDY OF TRADITIONAL VS. MACHINE LEARNING APPROACHES IN CREDIT RISK ASSESSMENT
Keywords:
Inclusive Growth, PovertyAbstract
Credit risk assessment in the real estate supply chain is a critical component of financial risk management, influencing investment decisions and overall sector stability. Traditional credit risk models rely on historical financial data, credit scores, and financial ratios to evaluate borrower credibility. Although effective for regulatory and interpretability purposes, these methods often fail to capture rapid market changes, emerging risks, and complex financial dependencies. The advancement of artificial intelligence (AI) and machine learning (ML) has introduced more dynamic approaches to credit risk analysis. AI-driven models utilize big data, predictive analytics, and real-time market information to detect hidden patterns and behaviours that conventional methods overlook. This study presents a comparative evaluation of AI-based and traditional credit risk models, focusing on their predictive accuracy, adaptability, scalability, and limitations within the real estate sector. Findings indicate that AI models offer superior precision and responsiveness, helping financial institutions, lenders, and investors make more informed decisions and reduce potential losses. However, concerns regarding algorithmic transparency, ethical considerations, data privacy, and regulatory compliance continue to challenge widespread AI adoption. The study concludes that balancing technological innovation with robust regulatory frameworks is essential for responsible and effective credit risk management in the real estate supply chain.
Downloads
References
Anjum, A., et al. (2024). Comparative analysis of gradient boosting methods in financial risk prediction. Airlangga, T. (2024). Statistical testing of algorithmic performance in credit scoring models. Bhandary, S., & Ghosh, P. (2025). Evaluation of traditional credit scoring models. Chang, L., et al. (2024). XGBoost optimization for financial credit risk assessment. Dong, Y., et al. (2024). Data-driven analysis of ensemble model performance in credit risk. Haque, R., et al. (2024). Stacked ensemble approaches in credit scoring. Karthik, V., et al. (2025). Computational efficiency in AI-based credit risk evaluation. Khekare, R., et al. (2025). Handling class imbalance in credit default prediction. Mahajan, A., et al. (2024). Hybrid modeling strategies for mid-sized financial institutions. Shukla, P., et al. (2023). Random forest and deep learning models for credit risk prediction. Ukpabi, D., et al. (2025). Explainable AI for regulatory compliance in financial systems. Ujkani, E., et al. (2025). Industry adoption of XGBoost in large-scale financial credit scoring.
