基于数据挖掘的互联网金融平台风险治理研究  

Research on Risk Management of Internet Financial Platform Based on Data Mining

作　　者：张昶[1,2] 李晓峰[1] 任媛媛[2] ZHANG Chang;LI Xiao-feng;REN Yuan-yuan(Beijing University of Posts and Telecomunications,Beijing 100876,China;Shijiazhuang Posts and Telecommunications Technical College,Shijiazhuang 050021,China)

机构地区：[1]北京邮电大学,北京100876 [2]石家庄邮电职业技术学院,石家庄050021

出　　处：《价值工程》2019年第8期148-151,共4页Value Engineering

基　　金：河北省高等学校科学研究项目-青年基金项目:基于人工神经网络与决策树算法的P2P互联网金融平台风险治理研究;项目编号:QN2018318

摘　　要：随着网络技术的飞速发展,P2P互联网金融平台催生了大量的理财和借贷行为。但由于互联网两端存在着信息不对称性,会产生大量的借贷信用风险问题。本文利用国内某大型互联网金融平台的借贷数据,基于数据挖掘的思路和方法,对数据进行了预处理、挖掘建模以及结果的分析,主要通过决策树算法找到借贷违约人的普遍特征,挖掘出隐藏在数据背后的知识和模式,并提出互联网金融平台的借贷风险治理方案,降低了信息不对称性,优化互联网金融平台的资源配置。With the rapid development of network technology, a large number of financial management and lending behaviors have been happened on peer-to-peer Internet financial platforms. However, due to the information asymmetry at both sides of the Internet, a large number of loan credit risk problems will arise. Based on the thoughts and methods of data mining, this paper uses the loan data of large Internet financial platforms in China to do data pre-process, model mining and results analysis. The decision tree algorithm is used here to find general characteristics of loan defaulters,and the knowledge and patterns hidden behind the data are mined. The loan risk management scheme of Internet financial platform is proposed, which can reduce the information asymmetry and optimize the resource allocation of Internet financial platform.

关 键 词：P2P互联网金融平台 信息不对称性 借贷信用风险 数据挖掘 决策树算法 

分 类 号：F724.6[经济管理—产业经济]