Novel hybrid ensemble credit scoring model with stacking-based noise detection and weight assignment
作者：
Highlights：
• A novel hybrid ensemble credit scoring model is proposed.
• A stacking-based noise detection method enhances the label noise adaptability of classifiers.
• The Bayesian optimization algorithm is applied to optimize the hyperparameters of classifiers.
• The overall performance of the proposed model is improved by the new weight assignment method.
• The proposed model outperforms benchmark ensemble models.
摘要
•A novel hybrid ensemble credit scoring model is proposed.•A stacking-based noise detection method enhances the label noise adaptability of classifiers.•The Bayesian optimization algorithm is applied to optimize the hyperparameters of classifiers.•The overall performance of the proposed model is improved by the new weight assignment method.•The proposed model outperforms benchmark ensemble models.
论文关键词：Cloud model,Credit scoring,Ensemble modeling,Machine learning,Noise detection
论文评审过程：Received 5 December 2021, Revised 8 February 2022, Accepted 12 March 2022, Available online 18 March 2022, Version of Record 22 March 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.116913
