Credit rating with a monotonicity-constrained support vector machine model

作者：

Highlights：

• We proposed a novel monotonicity constrained SVM model for credit rating.

• We evaluate the performance of the model with real-world data sets.

• The proposed method can correct the loss of monotonicity in the data.

• The proposed method can improve the performance as compared to the conventional SVM.

摘要

•We proposed a novel monotonicity constrained SVM model for credit rating.•We evaluate the performance of the model with real-world data sets.•The proposed method can correct the loss of monotonicity in the data.•The proposed method can improve the performance as compared to the conventional SVM.

论文关键词：Credit rating,SVM,Monotonicity constraint,Prior domain knowledge,Data mining

论文评审过程：Available online 12 June 2014.

论文官网地址：https://doi.org/10.1016/j.eswa.2014.05.035