Zobraziť v EDS

Using Artificial Intelligence and Deep Learning Applications in Credit Risk Analysis.

Uložené v:
Podrobná bibliografia
Názov: Using Artificial Intelligence and Deep Learning Applications in Credit Risk Analysis.
Autori: Faris, Asmaa, Elhachloufi, Mostafa
Zdroj: Cuestiones de Fisioterapia; 2025, Vol. 54 Issue 4, p7104-7122, 19p
Predmety: CREDIT analysis, CREDIT risk, CREDIT risk management, ARTIFICIAL intelligence, CREDIT management
Abstrakt: This study aims to develop a high-performing predictive model capable of classifying a credit file as *good* or *bad*, in order to help financial institutions reduce the risks of default and optimize their credit decision-making processes. A deep learning model was trained on a dataset of 45,211 client records, using key variables such as *balance*, *loan*, *contact*, and *pdays*. The approach is based on an advanced classifier evaluated using standard performance metrics, including precision, recall, F1-score, and accuracy. The model demonstrated remarkable performance, with an overall accuracy of approximately 0.88, outperforming traditional scoring methods. The analysis of the most influential variables confirmed their relevance in credit file classification, and the model was also effective in identifying high-risk profiles, thus enhancing credit portfolio management. While these results highlight the transformative potential of deep learning in credit risk assessment, challenges remain, particularly regarding interpretability and regulatory compliance. It is crucial to develop more transparent approaches to ensure decision explainability and responsible adoption by financial institutions. Future research should also explore the integration of unstructured data to further refine credit risk assessment. [ABSTRACT FROM AUTHOR]
Copyright of Cuestiones de Fisioterapia is the property of Cuestiones de Fisioterapia and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Databáza: Biomedical Index
Popis
Abstrakt:This study aims to develop a high-performing predictive model capable of classifying a credit file as *good* or *bad*, in order to help financial institutions reduce the risks of default and optimize their credit decision-making processes. A deep learning model was trained on a dataset of 45,211 client records, using key variables such as *balance*, *loan*, *contact*, and *pdays*. The approach is based on an advanced classifier evaluated using standard performance metrics, including precision, recall, F1-score, and accuracy. The model demonstrated remarkable performance, with an overall accuracy of approximately 0.88, outperforming traditional scoring methods. The analysis of the most influential variables confirmed their relevance in credit file classification, and the model was also effective in identifying high-risk profiles, thus enhancing credit portfolio management. While these results highlight the transformative potential of deep learning in credit risk assessment, challenges remain, particularly regarding interpretability and regulatory compliance. It is crucial to develop more transparent approaches to ensure decision explainability and responsible adoption by financial institutions. Future research should also explore the integration of unstructured data to further refine credit risk assessment. [ABSTRACT FROM AUTHOR]
ISSN:11358599