A novel heterogeneous ensemble credit scoring model based on bstacking approach

作者：

Highlights：

• A novel heterogeneous ensemble credit scoring model (i.e., bstacking) is proposed.

• Several novel classifiers have been introduced as base learners.

• Bstacking significantly outperforms several state-of-the-art benchmark models.

• We discuss accurate yet complex credit scoring model from a regulatory perspective.

摘要

•A novel heterogeneous ensemble credit scoring model (i.e., bstacking) is proposed.•Several novel classifiers have been introduced as base learners.•Bstacking significantly outperforms several state-of-the-art benchmark models.•We discuss accurate yet complex credit scoring model from a regulatory perspective.

论文关键词：Credit scoring,Heterogeneous ensemble,Bagging,Stacking

论文评审过程：Received 19 July 2017, Revised 8 October 2017, Accepted 9 October 2017, Available online 10 October 2017, Version of Record 18 October 2017.

论文官网地址：https://doi.org/10.1016/j.eswa.2017.10.022