基于神经网络模型的网络借贷高危企业信用风险的识别研究

doi:10.11959/j.issn.2096-109x.2017.00222

Abstract

Abstract:

The rapid development of online lending alleviates the difficulty of financing for small and micro enterprises to a certain extent,but it also exposes the credit risk identification problem of online lending platform.In order to fully identify the characteristics of high-risk network lending enterprises,small and medium-sized network lending companies were selected as samples,and indicators that were highly correlated with risk identification were chosen as indicators variables.And by using the BP neural network algorithm model,the credit risk identification rate and credit risk classification accuracy rate of high risk network lending enterprises,under different conditions,were obtained.The results show that the credit risks of high-risk network lending enterprises are highly recognized,and have the characteristics of high recall rate and high accuracy.

Key words: high risk online lending enterprise risk identification, index screening, neural network, recall rate

CLC Number:

Mao-guang WANG,Zi-jun ZHU. Credit risk identification of high-risk online lending enterprises based on neural network model[J]. Chinese Journal of Network and Information Security, 2017, 3(12): 15-21.

Figures/Tables 10

变量类别	变量名称	数据类型	分类依据
企业基本指标	公司类型	离散变量	私营类=-4，股份制=-3，合伙制=-2，国有制=-1，集体所有制=1，联营企业=2，外商投资类=3，港澳台投资类=4，缺失=0
	开业时间	连续变量	(0,90]天=5，(90,180]天=4，(180,360]天=3，(360,1200]天=2，(1200,∞)=1，缺失=0
	核准开业时间	连续变量	(-∞,0]天以下=6，(0,100]天=5，(100,180]天=4，(180,360]天=3，(360,600]天=2，(600,∞)=1，缺失=0
	是否有第三方征信	二元变量	有=1，无=0
	是否接受过风投	二元变量	有=1，无=0
	自动投标	二元变量	有=1，无=0
财务指标	平均收益	连续变量	[0.05,0.1)=1，[0.1,0.15)=2，[0.15,0.2)=3，(0.2,∞)=4
	平均投资期限	连续变量	(0,30]天=1，(30,60]天=2，(60,∞)天=3
	平均标的金额	连续变量	(0,5×10⁴]元=1，(5×10⁴,5×10⁵]元=2，(5×10⁵,10⁶]元=3，(10⁶,5×10⁶]元=4，(5×10⁶,10⁷]元=5
资本情况	注册资金	连续变量	(0,5×10⁶]元=1，(5×10⁶,2×10⁷]元=2，(2×10⁷,5×10⁷]元=3，(5×10⁷,10⁸]元=4，(10⁸,∞]元=5，缺失=0
	实缴资金	连续变量	(0,10⁶]元=1，(10⁶,5×10⁶]元=2，(5×10⁶,10⁷]元=3，(10⁷,5×10⁷]元=4，(5×10⁷,∞]元=5，缺失=0
	实缴资金比例	连续变量	(0,0.01]=1，(0.01,0.1]2，(0.1,0.5]=3，(0.5,1]=4，缺失=0
平台资本保障	平台管理费	二元变量	有=1，无=0
	平台充值费	二元变量	有=1，无=0
	平台提现费	二元变量	有=1，无=0
	平台VIP费	二元变量	有=1，无=0
	平台转让费	二元变量	有=1，无=0
	债券转让	二元变量	有=1，无=0
	保障模式	离散变量	非融资性担保公司=-4，融资性担保公司=-3，平台垫付=-2，小贷公司=-1，风险准备金=1，抵押质押=2，中介平台=3，其他=4，缺失=0
评价指标	第三方评价	二元变量	有=1，无=0
评价指标	用户推荐度	连续变量	0%=0，(0,0.1]=1，(0.1,0.25]=2，(0.25,0.5]=3，(0.5,0.99]=4，100%=5
结果变量	平台风险状态	二元变量	有=1，无=0

References 10

[1]	BARTH M E , BEAVER W H , LANDSMAN W R . Value-relevance of banks'fair value disclosures under SFAS No.107[J]. Accounting Review, 1996,71(4): 513-537.
[2]	刘新海 . 阿里巴巴集团的大数据战略与征信实践[J]. 征信, 2014,32(10): 10-14.
	LIU X H . The strategy of big data for Alibaba Group and its practices of credit reference[J]. Credit Reference, 2014,32(10): 10-14.
[3]	肖曼君, 欧缘媛, 李颖 . 我国P2P网络借贷信用风险影响因素研究——基于排序选择模型的实证分析[J]. 财经理论与实践, 2015,36(1): 2-6.
	XIAO M J , OU Y Y , LI Y . Research on the factors affecting credit risk of P2P lending in China:an empirical analysis based on the sort selection model[J]. The Theory and Practice of Finance and Economics, 2015,36(1): 2-6.
[4]	刘宇璟, 林裘绪, 窦先琴 . 小微企业创业者素质评价指标体系构建[J]. 统计与决策, 2015,(20): 185-188.
	LIU Y J , LIN Q X , DOU X Q . The construction of the quality evaluation index system for the entrepreneurs of small and micro enterprises[J]. Statistics and Decision, 2015,(20): 185-188.
[5]	邬建平 . 电子商务信用风险评估模型[J]. 统计与决策, 2016,(11): 69-71.
	WU J P . Credit risk assessment model of e-commerce[J]. Statistics and Decision, 2016,(11): 69-71.
[6]	ATIYA A F . Bankruptcy prediction for credit risk using neural networks:a survey and new results[J]. IEEE Transactions on Neural Networks, 2001,12(4):929.
[7]	叶斌, 雷燕 . 关于BP网中隐含层层数及其节点数选取方法浅析[J]. 商丘职业技术学院学报, 2004(6): 52-53.
	YE B , LEI Y . On the method of selecting the number of hidden layers and the number of nodes in BP network[J]. Journal of Shangqiu Vocational Technical College, 2004,(6): 52-53.
[8]	熊志斌 . 运用遗传神经网络模型对我国上市公司财务危机的预测[J]. 统计与决策, 2008,(14): 32-34.
	XIONG Z B . Using genetic neural network model to predict financial crisis of Listed Companies in China[J]. Statistics and Decision, 2008,(14): 32-34.
[9]	翟万里 . 基于人工神经网络的商业银行信用风险评估模型研究[D]. 长沙:长沙理工大学, 2013.
	ZHAI W L . Research on credit risk assessment model of commercial banks based on artificial neural network[D]. Changsha：Changsha University of Science and Technology, 2013.
[10]	刘建伟, 刘媛, 罗雄麟 . 深度学习研究进展[J]. 计算机应用研究, 2014,31(7): 1921-1930.
	LIU J W , LIU Y , LUO X L . Progress in research on deep learning[J]. Application Research of Computers, 2014,31(7): 1921-1930.

Metrics

Recommended 0

No Suggested Reading articles found!

属性	含义
X1	网贷企业名称
X2	网贷企业的法人代表
X3	网贷公司性质、公司类型
X4	网贷企业的开业时间
X5	网贷企业核准开业时间
X6	资讯平台对网贷企业的评价打分
X7	网贷企业的平均收益
X8	网贷企业的注册资本
X9	网贷企业实际缴付的资本
X10	网贷平台是否允许投资自动投标业务
X11	网贷平台是否允许投资者进行债券转让
X12	网贷平台资金是否有托管机构
X13	网贷平台投资的标是否有担保
X14	担保机构名称
X15	网贷平台管理费用的情况
X16	网贷平台充值费用的情况
X17	网贷平台提现费用的情况
X18	网贷平台转让费用的情况
X19	投资期限在1个月的标的所占比重
X20	投资期限在3个月的标的所占比重
X21	投资期限在4～6个月的标的所占比重
X22	投资期限在6个月以上的标的所占比重
X23	标的金额在[0,10⁵)元所占比重
X24	标的金额在[10⁵～10⁶)元所占比重
X25	标的金额在[10⁶,10⁷)元所占比重
X26	标的金额在1 000万元及以上所占比重
X27	贵宾客户费用情况
X28	费用支付方式情况
X29	网贷企业的近期交易成交量
X30	网贷平台累计待还金额
X31	平台最近30天待还金额
X32	平台净流入金额
X33	平台的投资人数
X34	平台的借款人数
X35	对网贷平台进行评价的人数
X36	用户对于网贷企业的推荐程度（%）
X37	是否为高危网贷企业

平台收益	平台状态		WOE	IV
平台收益	0	1	WOE	IV
1	1 051	715	0.71	0.33
2	80	489	-1.49	0.36
3	12	378	-3.12	0.71
总计	1 143	1 582	—	1.40

实际类别		预测类别		总计
实际类别	GOOD(低风险企业)		BAD(高风险企业)	总计
GOOD(低风险企业)	TP		FN	P(实际为GOOD)
BAD(高风险企业)	FP		TN	N(实际为BAD)
总计	p(被分为GOOD)		n(被分为BAD)	P+N

学习率	迭代次数	训练数据	测试数据	高风险企业数	高风险企业分类正确数据	高风险企业识别召回率	神经网络正确率
0.025	10	900	1 825	693	692	99.86%	78.79%
0.025	25	900	1 825	693	686	98.99%	83.01%
0.025	50	900	1 825	693	688	99.28%	79.12%
0.025	100	900	1 825	693	693	100.00%	39.12%
0.025	200	900	1 825	693	692	99.86%	40.33%
0.025	500	900	1 825	693	560	80.81%	36.71%

[1]	Liquan CHEN, Yuhang ZHU, Yu WANG, Zhongyuan QIN, Yang MA. New hash function based on C-MD structure and chaotic neural network [J]. Chinese Journal of Network and Information Security, 2023, 9(3): 1-15.
[2]	Guanyun FENG, Cai FU, Jianqiang LYU, Lansheng HAN. Insider threat detection based on operational attention and data augmentation [J]. Chinese Journal of Network and Information Security, 2023, 9(3): 102-112.
[3]	Zhao CAI, Tao JING, Shuang REN. Survey on Ethereum phishing detection technology [J]. Chinese Journal of Network and Information Security, 2023, 9(2): 21-32.
[4]	Zezhou HOU, Jiongjiong REN, Shaozhen CHEN. Security evaluation for parameters of SIMON-like cipher based on neural network distinguisher [J]. Chinese Journal of Network and Information Security, 2023, 9(2): 154-163.
[5]	Rongna XIE, Zhuhong MA, Zongyu LI, Ye TIAN. Encrypted traffic classification method based on convolutional neural network [J]. Chinese Journal of Network and Information Security, 2022, 8(6): 84-91.
[6]	Dibin SHAN, Xuehui DU, Wenjuan WANG, Aodi LIU, Na WANG. Access control relationship prediction method based on GNN dual source learning [J]. Chinese Journal of Network and Information Security, 2022, 8(5): 40-55.
[7]	Hao CHEN, Feng WANG, Weiming ZHANG, Nenghai YU. Carrier-independent deep optical watermarking algorithm [J]. Chinese Journal of Network and Information Security, 2022, 8(4): 110-118.
[8]	Dian LIN, Li PAN, Ping YI. Research on the robustness of convolutional neural networks in image recognition [J]. Chinese Journal of Network and Information Security, 2022, 8(3): 111-122.
[9]	Xinya WANG, Guang HUA, Hao JIANG, Haijian ZHANG. Survey on intellectual property protection for deep learning model [J]. Chinese Journal of Network and Information Security, 2022, 8(2): 1-14.
[10]	Tong QIAO, Hongwei YAO, Binmin PAN, Ming XU, Yanli CHEN. Research progress of digital image forensic techniques based on deep learning [J]. Chinese Journal of Network and Information Security, 2021, 7(5): 13-28.
[11]	Zhenglong WANG, Baowen ZHANG. Survey of generative adversarial network [J]. Chinese Journal of Network and Information Security, 2021, 7(4): 68-85.
[12]	Jinyin CHEN, Dunjie ZHANG, Guohan HUANG, Xiang LIN, Liang BAO. Adversarial attack and defense on graph neural networks: a survey [J]. Chinese Journal of Network and Information Security, 2021, 7(3): 1-28.
[13]	Peijie LI, Li ZHANG, Yunfei XIA, Liming XU. Architecture design of re-configurable convolutional neural network on software definition [J]. Chinese Journal of Network and Information Security, 2021, 7(3): 29-36.
[14]	Hao CHEN, Ping YI. Code vulnerability detection method based on graph neural network [J]. Chinese Journal of Network and Information Security, 2021, 7(3): 37-45.
[15]	Qingyin TAN, Yingming ZENG, Ye HAN, Yijing LIU, Zheli LIU. Survey on backdoor attacks targeted on neural network [J]. Chinese Journal of Network and Information Security, 2021, 7(3): 46-58.

Credit risk identification of high-risk online lending enterprises based on neural network model

RichHTML

PDF下载

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 10

References 10

Related Articles 15

Metrics

Recommended 0