A benchmark of machine learning approaches for credit score prediction
作者：
Highlights：
• A credit risk scoring models benchmarking in P2P platform has been proposed.
• The class imbalance problem has been addressed using different sampling strategies.
• The evaluation has been made on a real P2P data-set composed by 877,956 samples.
• The outcomes have been compared with state-of-the art approaches on different metrics.
• The three best models have been also evaluated in terms of their explainability.
摘要
•A credit risk scoring models benchmarking in P2P platform has been proposed.•The class imbalance problem has been addressed using different sampling strategies.•The evaluation has been made on a real P2P data-set composed by 877,956 samples.•The outcomes have been compared with state-of-the art approaches on different metrics.•The three best models have been also evaluated in terms of their explainability.
论文关键词：Credit score prediction,Benchmark,Supervised learning,Machine learning,Explainable artificial intelligence
论文评审过程：Received 25 May 2020, Revised 7 September 2020, Accepted 7 September 2020, Available online 9 September 2020, Version of Record 12 September 2020.
论文官网地址：https://doi.org/10.1016/j.eswa.2020.113986
