A comparative study on base classifiers in ensemble methods for credit scoring
作者：
Highlights：
• Small improvements in the systems about credit scoring can suppose great profits.
• Ensembles of classifiers achieve the better results for credit risk assessment.
• To look for the best base classifier used in ensembles on credit datasets is an important task.
• Via experiments, it is shown that the credal decision tree classifier is the best one to be used in ensembles.
• The study uses several of the most successful ensemble schemes and single classifiers.
摘要
•Small improvements in the systems about credit scoring can suppose great profits.•Ensembles of classifiers achieve the better results for credit risk assessment.•To look for the best base classifier used in ensembles on credit datasets is an important task.•Via experiments, it is shown that the credal decision tree classifier is the best one to be used in ensembles.•The study uses several of the most successful ensemble schemes and single classifiers.
论文关键词：Credit scoring,Ensembles of classifiers,Base classifier,Decision trees,Imprecise Dirichlet model,Uncertainty measures
论文评审过程：Received 1 August 2016, Revised 12 December 2016, Accepted 13 December 2016, Available online 14 December 2016, Version of Record 21 December 2016.
论文官网地址：https://doi.org/10.1016/j.eswa.2016.12.020
