摘要
风险预警是当前电子商务信用领域研究的热点,而电子商务信用风险与多种因素密切相关,具有随机性、时变规律,传统方法难以全面、科学地反映描述电子商务信用风险的变化特点,导致电子商务信用风险预警误差大。为了降低电子商务信用风险预警误差,获得理想的电子商务信用风险预警结果,提出大数据背景下的电子商务信用风险预警方法。首先,分析与电子商务信用风险相关的影响因素,提取数据,通过专家对电子商务信用风险进行打分,建立电子商务信用风险预警学习样本集合;然后,采用大数据分析技术对影响因素和电子商务信用风险值之间的关系进行建模,建立电子商务信用风险预警模型;最后,在相同平台上与传统方法进行对照测试。结果表明,所提方法的电子商务信用风险预警正确率高、预警时间短,改善了电子商务信用风险预警结果,对照测试结果验证了该方法的优越性。
Risk early warning is a hot topic in the field of e⁃commerce credit.The e⁃commerce credit risk is closely related to a variety of factors,which has randomness and time⁃varying characteristics.Therefore,the traditional methods is far away from describing the change characteristics of e⁃commerce credit risk comprehensively and scientifically,which results that the error of e⁃commerce credit risk early warning obtained with the traditional methods is large.In order to reduce the error of e⁃commerce credit risk early warning and get reasonable early warning results,an e⁃commerce credit risk early warning method under the background of big data is proposed.The influencing factors related to e⁃commerce credit risk are analyzed,the data are extracted and experts are employed to grade the e⁃commerce credit risk,so as to establish the learning sample set of e⁃commerce credit risk early warning.And then,the modeling of the relationship between the influencing factors and the e⁃commerce credit risk value is implemented with big data analysis technology to establish the e⁃commerce credit risk early warning model.A comparative test of the proposed method and the traditional methods was carried out on the same platform.The results show that the accuracy of the e⁃commerce credit risk early warning of the proposed method is high,and its early warning time is short,so the proposed method can improve the e⁃commerce credit risk early warning results.The results of the comparison test show the advantages of this method.
作者
薛淑娟
XUE Shujuan(Zhengzhou Sias University,Zhengzhou 451150,China)
出处
《现代电子技术》
2021年第3期74-78,共5页
Modern Electronics Technique
关键词
电子商务
信用风险
预警方法
影响因素分析
仿真实验
预警模型
e⁃commerce
credit risk
early warning method
influencing factor analysis
simulation experiment
early warning model
