基于知识图谱的小微企业贷款申请反欺诈方案

doi:10.11959/j.issn.2096-0271.2019035

Abstract

Abstract:

While major commercial banks have been providing various business loans,the risk of loan fraud has arisen at the same time.In order to overcome this challenge,an anti-fraud model was proposed based on the full-scale enterprise portrait and enterprise relation graph.By integrating multi-source information to form a concrete enterprise profile,and quantifying the interactions among enterprise entities,the fraud risk of new SMEs’loan applications could be quantitatively evaluated.Experiments show that compared with purely considering enterprise’s attributes,the additional use of relational information contributes a 5% increase in the AUC of the test set,which is more helpful for banking institutions to accurately assess the corporate fraud risk.

Key words: anti-fraud,enterprise portrait, relation graph

CLC Number:

TP393

Panshi JIN,Guangming WAN,Lizhong SHEN. Knowledge graph-based fraud detection for small and micro enterprise loans[J]. Big Data Research, 2019, 5(4): 100-112.

Figures/Tables 7

References 22

[1]	张潇飞 . 商业银行小微企业贷款信用风险研究[J]. 智富时代, 2015(2).
	ZHANG X F . Research on credit risk of small and micro enterprise loans in commercial banks[J]. The Fortune Times, 2015(2).
[2]	陈隆, 闫真宇, 邓舒仁 . 对当前小微企业融资问题的若干思考[J]. 浙江金融, 2018(1): 17-23.
	CHEN L , YAN Z Y , DENG S R . Reflections on current financing problems of small and micro enterprises[J]. Zhejiang Finance, 2018(1): 17-23.
[3]	孙自通 . 小微企业信贷业务流程与法律实务[M]. 北京: 中华工商联合出版社, 2017.
	SUN Z T . Small and micro enterprise credit business process and legal practice[M]. Beijing: All-China Federation of Industry and CommercePress, 2017.
[4]	于沛丰 . 大数据金融破解小微企业融资难的分析[J]. 全国流通经济, 2018,2189(29): 88-89.
	YU P F . Analysis of big data finance cracking the financing difficulties of small and micro enterprises[J]. China Circulation Economy, 2018,2189(29): 88-89.
[5]	陈平, 王晓婷, 黄一朕 . 商业银行企业级反欺诈实践与趋势[J]. 中国银行业, 2017(11): 23-25.
	CHEN P , WANG X T , HUANG Y Z . Commercial bank’s enterprise-level anti-fraud practice and trend[J]. China Banking, 2017(11): 23-25.
[6]	丁濛濛 . 基于规则引擎的互联网金融反欺诈研究[J]. 电脑知识与技术, 2018,14(1): 1-3.
	DING M M . Internet finance antifraud research based on rule engine[J]. Computer Knowledge and Technology, 2018,14(1): 1-3.
[7]	仵伟强, 后其林 . 基于机器学习模型的消费金融反欺诈模型与方法[J]. 现代管理科学, 2018(10): 51-54.
	WU W Q , HOU Q L . Consumer finance anti-fraud model and method based on machine learning model[J]. Modern Management Science, 2018(10): 51-54.
[8]	何湘东, 魏吉勇 . B2B平台的反欺诈问题研究[J]. 信息安全与技术, 2016,7(11): 47-51.
	HE X D , WEI J Y . Research on the B2B platform anti-fraud problem[J]. Information Security and Technology, 2016,7(11): 47-51.
[9]	李苏, 周小惠, 宝哲 . 基于支持向量机的商业银行对中小信贷企业选择方法的研究[J]. 数学的实践与认识, 2018(11): 299-305.
	LI S , ZHOU X H , BAO Z . Research of loan enterprise selection for bank based on support vector machine[J]. Mathematics in Practice and Theory, 2018(11): 299-305.
[10]	张杰, 赵峰 . 基于支持向量机的中小企业技术信贷违约预测[J]. 统计与决策, 2013(20): 66-69.
	ZHANG J , ZHAO F . SME technology credit default forecast based on support vector machine[J]. Statistics and Decision, 2013(20): 66-69.
[11]	邱耀, 杨国为 . 基于XGBoost算法的用户行为预测与风险分析[J]. 工业控制计算机, 2018,31(9): 47-48.
	QIU Y , YANG G W . User behavior prediction and risk analysis based on XGBoost algorithmv[J]. Industrial Control Computer, 2018,31(9): 47-48.
[12]	陈俊清 . 神经网络模型在互联网金融反欺诈领域的研究与实践[J]. 中国金融电脑, 2016(8): 42-46.
	CHEN J Q . Research and practice of neural network model in the field of internet financial anti-fraud[J]. Financial Computer of China, 2016(8): 42-46.
[13]	胡鹏飞 . 金融科技在互联网金融行业性风险防范领域的应用[J]. 大数据, 2018,4(1): 117-123.
	HU P F . Application of FinTech in internet financial industry risk prevention[J]. Big Data Research, 2018,4(1): 117-123.
[14]	樊盛博 . 金融社交网络在伪卡欺诈发现中的应用研究[J]. 中国金融电脑, 2017(3): 65-71.
	FAN S B . Research on the application of financial social network in the detection of pseudo-card fraud[J]. Financial Computer of China, 2017(3): 65-71.
[15]	BEUTEL A , AKOGLU L , FALOUTSOS C . Fraud detection through graph-based user behavior modeling[C]// The 22nd ACM SIGSAC Conference on Computer and Communications Security,October 12-16,2015,Denver,USA. New York:ACM Press, 2015: 1696-1697.
[16]	PEROZZI B , AL-RFOU R , SKIENA S . DeepWalk:online learning of social representations[C]// The 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining,August 24-27,2014,New York,USA. New York:ACM Press, 2014: 701-710.
[17]	GROVER A , LESKOVEC J . Node2vec:scalable feature learning for networks[C]// The 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining,August 13-17,2016,San Francisco,USA. New York:ACM Press, 2016: 855-864.
[18]	TANG J , QU M , WANG M ,et al. LINE:large-scale information network embedding[C]// The 24th International Conference on World Wide Web,May 18-22,2015,Florence,Italy.[S.l.:s.n. ], 2015: 1067-1077.
[19]	DONG Y , CHAWLA N V , SWAMI A ,et al. Metapath2vec:scalable representation learning for heterogeneous networks[C]// The 23rd ACM SIGKDD International Conference on Knowledge Discovery＆ Data Mining,August 13-17,2017,Halifax,Canada. New York:ACM Press, 2017: 135-144.
[20]	NEVILLE J . Iterative classification[M]. Heidelberg: SpringerPress, 2000.
[21]	KE G , MENG Q , FINLEY T ,et al. Lightgbm:a highly efficient gradient boosting decision tree[C]// The 31st Conference on Neural Information Processing Systems,December 4-9,2017,Long Beach,USA.[S.l.:s.n. ], 2017: 3146-3154.
[22]	SNOEK J , LAROCHELLE H , ADAMS R P . Practical bayesian optimization of machine learning algorithms[C]// The 25th International Conference on Neural Information Processing Systems,December 3-6,2012,Lake Tahoe,USA.Miami:Curran Associates Inc. , 2012: 2951-2959.

Metrics

Recommended 0

No Suggested Reading articles found!

关系	表示	说明
企业与个人的对应关系		C₁,C₂∈(企业)，C₁≠C₂ P∈(个人) r₁,r₂∈(干系人关系)
企业干系人与企业干系人的对应关系		C₁,C₂∈(企业)，C₁≠C₂ P₁,P₂∈(个人)，P₁≠P₂ r₁,r₂∈(干系人关系) r₃∈(亲属关系)
企业与相关属性的对应关系		C₁,C₂∈(企业),C₁≠C₂ Attr_c∈(企业属性) r₄∈(企业属性拥有关系)
企业干系人与相关属性的对应关系		C₁,C₂∈(企业),C₁≠C₂ P₁,P₂∈(个人),P₁≠P₂ Attr_p∈(干系人属性) r₁,r₂∈(干系人关系) r₅,r₆∈(干系人拥有属性关系)
企业与企业的对应关系		C₁,C₂,C₃∈(企业),C₁≠C₂≠C₃ r₇∈(担保关系)

数据集	样本数量/个	正样本/个	负样本/个	正负样本比
训练集	112 708	312	112 396	2.7‰
测试集	28 177	77	28 100	2.7‰

模型	特征	测试集AUC值	测试集KS值
模型1	节点属性（60维）	0.791	0.466
模型2	节点属性+邻居特征（61维）	0.844	0.581

Knowledge graph-based fraud detection for small and micro enterprise loans

RichHTML

PDF下载

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 7

References 22

Related Articles 1

Metrics

Recommended 0