《计算机应用研究》|Application Research of Computers

互联网金融平台中高违约风险用户识别算法

Identification algorithm of high breaching risk member for Internet financial platform

免费全文下载 (已被下载 次)  
获取PDF全文
作者 阳晓慧,郭炳晖,米志龙,郑志明
机构 北京航空航天大学 大数据科学与脑机智能高精尖创新中心 数学、信息与行为教育部重点实验室 数学与系统科学学院,北京 100191
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2019)03-010-0691-05
DOI 10.19734/j.issn.1001-3695.2017.09.0928
摘要 以某互联网金融平台的用户交易数据为对象,通过分析其中借贷逾期违约的传播行为,提出通过传播特征构建模型算法识别互联网金融平台的高风险用户。在构建基于阈值传播和随机传播的SIS和SIR模型的基础上,将模型转换为可评价用户风险值的算法,并进一步与实际违约数据进行验证对比。对比结果显示,在前5%和10%高风险群体划分条件下,算法具有较高的召回率和良好的结构关联性。
关键词 风险传播;复杂网络;互联网金融;识别算法
基金项目 国家自然科学基金资助项目(11401017,11671025)
国家自然科学基金重大项目(11290141)
本文URL http://www.arocmag.com/article/01-2019-03-010.html
英文标题 Identification algorithm of high breaching risk member for Internet financial platform
作者英文名 Yang Xiaohui, Guo Binghui, Mi Zhilong, Zheng Zhiming
机构英文名 KeyLaboratoryofMinistryofMathematicalInformation&BehaviorEducation,SchoolofMathematics&SystemsScience,BeijingAdvancedInnovationCenterforBigData&BrainComputing,BeihangUniversity,Beijing100191,China
英文摘要 This paper studied loan transaction data of an Internet financial platform, and identified the high-risk members by analyzing the propagation behavior of the loan network. It established the SIS model and SIR model based on threshold propagation and random propagation, respectively. After that, it generated an algorithm to evaluate the users’ risk value. Furthermore, it compared it with the actual defaulting data. In terms of the top 5% and 10% high risk group division, the results show that it can achieve high recall rate and good structural correlation with the algorithm.
英文关键词 risk propagation; complex network; Internet banking; identification algorithm
参考文献 查看稿件参考文献
  [1] 熊亚骅, 熊一鹏, 李婷. 互联网金融网贷客户违约风险研究进展[J] . 金融经济, 2015(12):70-73. (Xiong Yahua, Xiong Yipeng, Li Ting. Research progress on default risk of Internet financial network loan customers[J] . Financial & Economy, 2015(12):70-73. )
[2] 王书斌, 谭中明, 陈艺云. P2P网贷违约风险及其传染性评估综述[J] . 武汉金融, 2017(6):40-44, 14. (Wang Shubin, Tan Zhongming, Chen Yiyun. Review of default risk and infectiousness assessment of P2P network loan[J] . Wuhan Finance Monthly, 2017(6):40-44, 14. )
[3] 王倩, Hartmannwendels T. 信用违约风险传染建模[J] . 金融研究, 2008(10):162-173. (Wang Qian, Hartmannwendels T. Modelling of contagion effects[J] . Journal of Financial Research, 2008(10):162-173. )
[4] 李杰汀. 基于复杂网络的关联信用风险传染建模与仿真研究[D] . 成都:电子科技大学, 2015. (Li Jieting. Research on correlative credit risk infection modeling and simulation based on complex networks[D] . Chengdu:University of Electronic Science and Technology of China, 2015. )
[5] 陈庭强, 何建敏. 基于复杂网络的信用风险传染模型研究[J] . 软科学, 2014, 28(2):111-117. (Chen Tingqiang, He Jianmin. Credit risk contagion model based on complex network[J] . Soft Science, 2014, 28(2):111-117. )
[6] 王申侃. P2P网贷平台应对借款违约的风险管理研究[D] . 北京:对外经济贸易大学, 2016. (Wang Shenkan. Research on risk ma-nagement of P2P online loan platform to cope with borrowing default[D] . Beijing:University of International Business and Economics, 2016. )
[7] 丁岚, 骆品亮. 基于Stacking集成策略的P2P网贷违约风险预警研究[J] . 投资研究, 2017, 36(4):41-54. (Ding Lan, Luo Pinliang. Early warning research on P2P network loan default risk based on stacking integration strategy[J] . Review of Investment Studies, 2017, 36(4):41-54. )
[8] 唐剑琴. 基于决策树算法的P2P网贷借款人违约风险度量研究[D] . 长沙:湖南师范大学, 2016. (Tang Jianqin. Research on default risk measurement of borrowers of P2P net loans based on decision tree algorithm[D] . Changsha:Hunan Normal University, 2016. )
[9] Zhou Jie, Liu Zonghua. Epidemic spreading in complex networks[J] . Frontiers of Physics in China, 2008, 3(3):331-348.
[10] Dorogovtsev S N, Mendes J F F. Evolution of networks[J] . Advances in Physics, 2002, 51:1079-1187.
[11] Newman M E J. Spread of epidemic disease on networks[EB/OL] . (2002-04-20). https://arxiv. org/abs/cond-mat/0205009.
[12] Zhang Zike, Liu Chuang, Zhan Xiuxiu, et al. Dynamics of information diffusion and its applications on complex networks[J] . Physics Reports, 2016, 651(9):1-34.
收稿日期 2017/9/25
修回日期 2017/11/21
页码 691-695,700
中图分类号 TP301.6
文献标志码 A