电商企业信用风险预警模型的缺失值鲁棒性  

Missing Value Robustness of Credit Risk Early Warning Model for E-Commerce Enterprises

作　　者：陈艳 蒋伟杰 CHEN Yan;JIANG Weijie(Department of Economic Management,Fuzhou University Zhicheng College,Fuzhou,China,350002;College of Mathematics and Computer Science/College of Sofeware,Fuzhou University,Fuzhou,China,350108)

机构地区：[1]福州大学至诚学院经济管理系,福州350002 [2]福州大学数学与计算机科学学院/软件学院,福州350108

出　　处：《福建电脑》2021年第8期56-59,共4页Journal of Fujian Computer

基　　金：福建省中青年教师教育科研项目(No.JAS180840);福州大学至诚学院课程改革项目(No.ZJ1928、No.ZJ2016)资助。

摘　　要：电子商务信用风险评估是建设信用体系的重要环节。在企业电子商务数据采集存在缺失值的情况下,本文比较了BP神经网络、支持向量机、决策树、极限学习机以及对应的集成模型在含缺失值预测样本的鲁棒性。实证数据分析结果显示,极限学习机及其集成模型在上述情况下优于其他模型。E-commerce credit risk assessment is an important part of the construction of credit system.This paper compares the robustness of the integrated models corresponding to BP neural network,support vector machine,decision tree and extreme learning machine in the case of missing value of enterprise e-commerce data.The analysis results of empirical data show that extreme learning machine and its integrated model are better than other models.

关 键 词：信用风险预警 缺失值 机器学习 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]