基于XGBoost和SHAP方法的个人信贷风险评估研究
Research on the Personal Credit Risk Assessment Based on XGBoost and SHAP Methods
伍洁 1陈迪芳 1李瑞彤 1石景阳1
作者信息
- 1. 湖北汽车工业学院 数理与光电工程学院,湖北 十堰 442002
- 折叠
摘要
为实现全面准确地评估个人信贷风险,首先,研究了借贷人的各项个人信息指标在信用风险评估中的重要性;接着,基于Python编程语言采用XGBoost集成学习方法搭建了一套个人贷款信用评估模型;随后,结合SHAP方法筛选出合理的信用评估指标,完善了评估模型;最后,基于LabVIEW平台开发个人贷款信用评估系统.研究结果表明:最终筛选的指标能更有效地评估个人信贷风险,可以为金融行业提供一个更有效的个人信贷风险评估系统.
Abstract
To achieve a comprehensive and accurate assessment of personal credit risk,firstly,the importance of various personal information indicators of borrowers in credit risk assessment is studied.Next,based on Python programming language and XGBoost integrated learning method,a personal loan credit assessment model is constructed.Subsequently,reasonable credit assessment indicators are selected by using the SHAP method to improve the assessment model.Finally,it develops a personal loan credit evaluation system based on the LabVIEW platform.The research results indicate that the final selected indicators can more effectively evaluate personal credit risk,and can provide a more effective personal credit risk assessment system for the financial industry.
关键词
信贷风险/XGBoost算法/SHAP/信用评估/PythonKey words
credit risk/XGBoost algorithm/SHAP/credit evaluation/Python引用本文复制引用
基金项目
湖北省大学生创新创业训练计划(S202210525055)
教育部产学合作协同育人项目(202101087049)
湖北省大学生创新创业训练计划(S202210525056)
出版年
2024
国家科技期刊平台
NSTL
维普
万方数据