[1]潘永明,王雅杰,来明昭.基于IG-SVM模型的供应链融资企业信用风险预测[J].南京理工大学学报(自然科学版),2020,44(01):117-126.[doi:10.14177/j.cnki.32-1397n.2020.44.01.018]
 Pan Yongming,Wang Yajie,Lai Mingzhao.Credit risk prediction of supply chain financing enterprisesbased on IG-SVM model[J].Journal of Nanjing University of Science and Technology,2020,44(01):117-126.[doi:10.14177/j.cnki.32-1397n.2020.44.01.018]
点击复制

基于IG-SVM模型的供应链融资企业信用风险预测()
分享到:

《南京理工大学学报》(自然科学版)[ISSN:1005-9830/CN:32-1397/N]

卷:
44卷
期数:
2020年01期
页码:
117-126
栏目:
出版日期:
2020-02-29

文章信息/Info

Title:
Credit risk prediction of supply chain financing enterprisesbased on IG-SVM model
文章编号:
1005-9830(2020)01-0117-10
作者:
潘永明王雅杰来明昭
天津理工大学 管理学院,天津 300384
Author(s):
Pan YongmingWang YajieLai Mingzhao
School of Management,Tianjin University of Technology,Tianjin 300384,China
关键词:
供应链融资 信息增益 支持向量机 信用风险 分类预测
Keywords:
supply chain financing information gain support vector machine credit risks classified prediction
分类号:
F832.4; F276.3
DOI:
10.14177/j.cnki.32-1397n.2020.44.01.018
摘要:
为了提高对供应链融资中小企业信用风险预测的精度,在通过对中小企业信用风险评价研究基础上集成机器学习算法构建了能够提高信用风险预测的组合模型。该模型采用支持向量机(Support vector machine,SVM)建立供应链中小企业信用风险分类预测模型,并引入信息增益(Information gain,IG)提取对预测结果有显著贡献的特征变量,优化模型特征输入。在与其他模型的对比实验中可知,采用IG-SVM模型预测的测试样本精确度为97.62%,比单一SVM模型精度提高8.97%。采用IG进行特征优化,能进一步提高SVM模型的预测能力。
Abstract:
In order to improve the accuracy of the credit risk prediction of small and medium-sized enterprises in supply chain financing,based on the research on the credit risk evaluation of small and medium-sized enterprises,a combination model which can improve the credit risk prediction is constructed by integrating machine learning algorithm. In this model,support vector machine(SVM)is used to establish the credit risk classification prediction model of small and medium-sized enterprises in the supply chain,and information gain(IG)is introduced to extract the feature variables that have significant contribution to the prediction results and optimize the feature input of the model. Compared with other models,IG-SVM model has the highest accuracy of 97.62%,which is 8.97% higher than single SVM model. Using IG for feature optimization can further improve the prediction ability of SVM model.

参考文献/References:

[1] 程砚秋. 基于不均衡数据的小企业信用风险评价[J]. 运筹与管理,2016,25(6):181-189.
Cheng Yanqiu. Credit risk assessment of small enterprises based on unbalanced data[J]. Operations Research and Management Science,2016,25(6):181-189.
[2]李梦宇,周莹. 供应链融资风险传染度量及贷款利率定价[J]. 统计与决策,2015(20):152-156.Li Mengyu,Zhou Ying. Risk contagion measurement and loan rate pricing of supply chain financing[J]. Statistics and Decision-making,2015(20):152-156.
[3]刘祥东,王未卿. 我国商业银行信用风险识别的多模型比较研究[J]. 经济经纬,2015,32(6):132-137.
Liu Xiangdong,Wang Weiqing. A comparative study of multiple models for credit risk identification of commercial banks in China[J]. Economic Survey,2015,32(6):132-137.
[4]Zhang Shuai,Hong Min. Design of credit risk evaluation system based on the fuzzy neural network and expert system[J]. Applied Mechanics and Materials,2014,543-547:4523-4527.
[5]Zhong Haoming,Miao Chunyan,Shen Zhiqi,et al. Comparing the learning effectiveness of BP,ELM,I-ELM,and SVM for corporate credit ratings[J]. Neurocomputing,2014,128:285-295.
[6]Mizgier K J,Wagner S M,Holyst J A. Modeling defaults of companies in multi-stage supply Chain networks[J]. International Journal of Production Economics,2012,135(1):14-23.
[7]周文坤,王付成. 供应链融资模式下中小企业信用风险评估研究——基于左右得分的模糊TOPSIS算法[J]. 运筹与管理,2015,24(1):209-215.
Zhou Wenkun,Wang Fucheng. Research on credit risk assessment of SMEs under supply chain financing mode:fuzzy TOPSIS based on left and right scores[J]. Operations Research and Management Science,2015,24(1):209-215.
[8]方匡南,范新妍,马双鸽. 基于网络结构Logistic模型的企业信用风险预警[J]. 统计研究,2016,33(4):50-55.
Fang Kuangnan,Fan Xinyan,Ma Shuangge. Early warning of enterprise credit risk based on logistic model of network structure[J]. Statistical Research,2016,33(4):50-55.
[9]刘艳春,崔永生. 供应链金融下中小企业信用风险评价——基于SEM和灰色关联度模型[J]. 技术经济与管理研究,2016(12):14-19.
Liu Yanchun,Cui Yongsheng. Credit risk assessment of small and medium-sized enterprises under supply chain finance:based on SEM and grey relational degree model[J]. Journal of Technical Economics & Management,2016(12):14-19.
[10]Govindan K,Fattahi M. Investigating risk and robustness measures for supply chain network design under demand uncertainty:A case study of glass supply chain[J]. International Journal of Production Economics,2017(183):680-699.
[11]Yang Yang. The research on the credit risk transmission mechanism of commercial banks under the perspective of supply chain financing[J]. West China Finance,2016(12):10-14.
[12]Zhu You,Xie Chi,Wang Gangjin,et al. Comparison of individual,ensemble and integrated ensemble machine learning methods to predict China’s SME credit risk in supply chain finance[J]. Neural Computing and Applications,2017(28):41-50.
[13]肖奎喜,王满四,倪海鹏. 供应链模式下的应收账款风险研究——基于贝叶斯网络模型的分析[J]. 会计研究,2011(11):65-71.
Xiao Kuixi,Wang Mansi,Ni Haipeng. Research on accounts receivable risk in supply chain mode:analysis based on Bayesian network model[J]. Accounting Research,2011(11):65-71.
[14]江训艳. 基于BP神经网络的商业银行信用风险预警研究[J]. 财经问题研究,2014(S1):46-48.
Jiang Xunyan. Research on credit risk early warning of commercial bank based on BP neural network[J]. Research on Financial and Economic Issues,2014(S1):46-48.
[15]Huang Xiaobing,Liu Xiaolian,Ren Yuanqian. Enterprise credit risk evaluation based on neural network algorithm[J]. Cognitive Systems Research,2018,52:317-324.
[16]淳伟德,肖杨. 供给侧结构性改革期间系统性金融风险的SVM预警研究[J]. 预测,2018,37(5):36-42.
Chun Weide,Xiao Yang. SVM early warning of systematic financial risk during supply-side structural reform[J]. Forecasting,2018,37(5):36-42.
[17]Zhang Lang,Hu Haiqing,Zhang Dan. A credit risk assessment model based on SVM for small and medium enterprises in supply chain finance[J]. Financial Innovation,2015,1(1):14.
[18]Wang Gang,Ma Jian. A hybrid ensemble approach for enterprise credit risk assessment based on Support Vector Machine[J]. Expert Systems with Applications,2012,39(5):5325-5331.
[19]李健,张金林. 供应链金融的信用风险识别及预警模型研究[J]. 经济管理,2019,41(8):178-196.
Li Jian,Zhang Jinlin. Research on credit risk identification and early warning model of supply chain finance[J]. Business Management Journal,2019,41(8):178-196.
[20]刘炳春,郑红梅,张斌. 基于IG-LASSO模型的城市空气质量指数混合预测研究[J]. 环境科学与技术,2017,40(11):144-148.
Liu Bingchun,Zheng Hongmei,Zhang Bin. Mixed prediction of urban air quality index based on IG-LASSO model[J]. Environmental Science & Technology,2017,40(11):144-148.
[21]Cortes C,Vapnik V. Support-vector networks[J]. Machine Learning,1995,20(3):273-297.
[22]赵莉. 基于支持向量机的云计算资源负载预测模型[J]. 南京理工大学学报,2018,42(6):687-692.
Zhao Li. Cloud computing resource load forecasting model based on support vector machine[J]. Journal of Nanjing University of Science and Technology,2018,42(6):687-692.
[23]Delmerico J,Isler S,Sabzevari R,et al. A comparison of volumetric information gain metrics for active 3D object reconstruction[J]. Autonomous Robots,2018(4):1-12.
[24]方匡南,吴见彬,朱建平,等. 信贷信息不对称下的信用卡信用风险研究[J]. 经济研究,2010,45(S1):97-107.
Fang Kuangnan,Wu Jianbin,Zhu Jianping,et al. Research on credit card credit risk under asymmetric credit information[J]. Economic Research Journal,2010,45(S1):97-107.
[25]肖瑾. 浅析供应链金融中信用风险的评价体系构建[J]. 经贸实践,2018(19):125.
Xiao Jin. On the construction of credit risk evaluation system in supply chain finance[J]. Economic and Trade Practice,2018(19):125.
[26]Zhu You,Zhou Li,Xie Chi,et al. Forecasting SMEs’credit risk in supply chainfinance with an enhancedhybrid ensemble machine learning approach[J]. International Journal of Production Economics,2019(211):22-33.

备注/Memo

备注/Memo:
收稿日期:2019-04-17 修回日期:2019-09-11
基金项目:国家自然科学基金(71503180); 天津市教委社会科学重大项目(2017JWZD16)
作者简介:潘永明(1963-),男,教授,主要研究方向:企业融资与计算机网络,E-mail:79411339@qq.com; 通讯作者:王雅杰(1995-),女,硕士生,主要研究方向:企业融资与计算机网络,E-mail:807063173@qq.com。
引文格式:潘永明,王雅杰,来明昭. 基于IG-SVM模型的供应链融资企业信用风险预测[J]. 南京理工大学学报,2020,44(1):117-126.
投稿网址:http://zrxuebao.njust.edu.cn
更新日期/Last Update: 2020-02-29