基于多模型融合的互联网信贷个人信用评估方法.docx

基于多模型融合的互联网信贷个人信用 评估方法 m ra 周国曷白鹏飞 安琪Nicolaas Fransde ROOIJ李楠 m ra 周国曷 华南师范大学华南先进光电子研究院深圳市国华光 电科技有限公司 深圳市国华光电研究院 摘要： 针对网络个人信用有效评分缺失的问题，分析了互联网信贷个人信用评估数据 的特点，选用支持向量机、随机森林和XGBoost分别建立了信用预测模型，并对 3种单一模型进行了投票加权融合?基于互联网信贷数据的特点，在特征工程中 对样本集特征进行了离散化、归一化和特征组合等处理?为増加对比，对实验数 据集进行了 FICO评估核心一一Logistic回归分析?实验结果表明：3种单一算法 性能均优于Logistic回归，XGBoost表现优于支持向量机和随机森林模型，预 测相对准确;投票融合模型的表现比单一模型更好，模型分辨能力更优秀，预测 精度更高，更适用于互联网信贷个人信用评估. 关键词： 个人信用评估;互联网信贷;支持向量机;随机森林;XGBoost;模型融合; 作者简介：白鹏飞，副研究员，Ema订:Baipf@scnu. edu. cn. 收稿日期:2017-03-06 基金：国家自然科学基金委员会-荷兰国家基金机构间合作重点项目(NSFC-NW0) (51561135014) Internet Credit Personal Credit Assessing Method Based on Multi一Model Ensemble BAI Pengfei AN Qi Nicolaas Frans de ROOIJ LI Nan ZHOU Guofu South China Academy of Advanced Opto electronics, South Chirm Normal University; Shenzhen Guohua Optoelectronics Technology Co., Ltd; Abstract： To solve the problem of the missing of the effective scores of online personal credits, the characteristics of internet personal credit assessment data arc analyzed. Support vector machinc （ SVM） 、 remdom forest （ RE） , and XGBoost have been adopted to establish the credit forecasting model in the paper, respectively. The voting fusion of the proposed models is conducted. Based on the data characteristics of internet credit data, discretization, normalization, and feature combi neiti on arc adop ted to experime ntal data set in fca turc eng inc cring. In order to improve the contrast, the logistic regression analysis-the core of FICO assessment is carried out. The experimentai results show that the performanee of the three established algorithm are better than logistic regression. The performance of XGBoost are better than SVM and RF model in the accuracy prediction. The performancc of voting fusion model is better than that of single model, with outstanding model resolution and prediction accuracy, which is more suitable for internet personal credi t assessment. Keyword： personal crcdit assessing; online lending; support vcctor machine;