article · Journal Of Big Data
Predicting credit risk is an essential operational requirement for financial institutions seeking to manage lending decisions effectively. A machine learning system combines a stacked classifier with an information-gain filter-based feature selection method to predict credit risk across multiple datasets. The architecture links Random Forest, Gradient Boosting, and Extreme Gradient Boosting sequentially as base estimators. Performance evaluation was conducted using accuracy, F1-score, and area under the curve metrics across three benchmark datasets from Australia, Germany, and Taiwan. When benchmarked against alternative classification techniques including artificial neural networks, decision trees, and k-nearest neighbour models, the stacked approach achieved superior overall predictive results. Specifically, the model yielded area under the curve scores of 0.934 for the Australian data, 0.944 for the German data, and 0.870 for the Taiwan data, outperforming both standalone estimators and standard baseline models.
Accurate credit risk assessments help financial organisations minimise defaults and allocate capital safely. By combining multiple algorithms and selecting the most informative features, this method improves predictive reliability across distinct regional datasets. Such improvements assist lenders in making more dependable lending evaluations, supporting broader institutional stability and responsible credit allocation.
This technique is applicable to automated credit risk assessment engines used by banks and lending organisations. As the model has been tested only on standard benchmark datasets, it remains at the stage of applied experimental research. Operational adoption would require engineering the sequential pipeline into production banking workflows, alongside validation against live institution-specific transaction data.
AI-generated from the published abstract. Always read the original work before citing.
Abstract Credit risk prediction is a crucial task for financial institutions. The technological advancements in machine learning, coupled with the availability of data and computing power, has given rise to more credit risk prediction models in financial institutions. In this paper, we propose a stacked classifier approach coupled with a filter-based feature selection (FS) technique to achieve efficient credit risk prediction using multiple datasets. The proposed stacked model includes the following base estimators: Random Forest (RF), Gradient Boosting (GB), and Extreme Gradient Boosting (XGB). Furthermore, the estimators in the Stacked architecture were linked sequentially to extract the best performance. The filter- based FS method that is used in this research is based on information gain (IG) theory. The proposed algorithm was evaluated using the accuracy, the F1-Score and the Area Under the Curve (AUC). Furthermore, the Stacked algorithm was compared to the following methods: Artificial Neural Network (ANN), Decision Tree (DT), and k-Nearest Neighbour (KNN). The experimental results show that stacked model obtained AUCs of 0.934, 0.944 and 0.870 on the Australian, German and Taiwan datasets, respectively. These results, in conjunction with the accuracy and F1-score metrics, demonstrated that the proposed stacked classifier outperforms the individual estimators and other existing methods.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1186/s40537-024-00882-0
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
