借款描述对互联网金融信用风险的影响研究  

The Influence of Loan Description on Credit Risk of Internet Finance

作　　者：陈为民[1] 杨泽俊 陈依 CHEN Weimin;YANG Zejun;CHEN Yi(School of Business,Hunan University of Science and Technology,Xiangtan,Hunan 411201,China)

机构地区：[1]湖南科技大学商学院,湖南湘潭411201

出　　处：《财经理论与实践》2022年第6期24-30,共7页The Theory and Practice of Finance and Economics

基　　金：国家社会科学基金(17BGL057)。

摘　　要：基于互联网金融提供的客户借款描述,通过潜在语义分析挖掘借款描述文本内容的主题,运用二元分位数回归分析借款描述对互联网金融信用风险的影响。实证结果表明,借款描述中有关情感表达、个人信用和借款目的的描述与违约情况呈负相关,有关财务情况的描述与违约情况呈正相关。Based on the customer loan description provided by Internet finance,the binary quantile regression method is used to classify the text content in the loan description of the lending platform and the data obtained by content mining according to the latent semantic analysis(LSA),and the impact of the loan description on the credit risk of Internet finance is considered.The results show that the description of emotional expression,personal credit and borrowing purpose in the loan description is negatively correlated with the default,and the description of the financial situation is positively correlated with the default.

关 键 词：互联网金融 二元分位数回归 潜在语义分析 文本挖掘 

分 类 号：F830.589[经济管理—金融学]