• • 上一篇    

基于LSTM-CNN的中小企业信用风险预测

王鑫1, 王莹1,2   

  1. 1. 北京信息科技大学经济管理学院, 北京 100192;
    2. 智能决策与大数据应用北京市国际科技合作基地, 北京 100192
  • 收稿日期:2022-05-10 修回日期:2022-07-01 发布日期:2022-11-04
  • 通讯作者: 王莹,Email:yeaishuijiao@126.com.
  • 基金资助:
    国家重点研发计划课题(2019YFB1405303),国家自然科学基金重点项目(71932002)资助课题.

王鑫, 王莹. 基于LSTM-CNN的中小企业信用风险预测[J]. 系统科学与数学, 2022, 42(10): 2698-2711.

WANG Xin, WANG Ying. Credit Risk Prediction of Small and Medium-Sized Enterprises Based on LSTM-CNN[J]. Journal of Systems Science and Mathematical Sciences, 2022, 42(10): 2698-2711.

Credit Risk Prediction of Small and Medium-Sized Enterprises Based on LSTM-CNN

WANG Xin1, WANG Ying1,2   

  1. 1. School of Economics and Management, Beijing Information Science and Technology University, Beijing 100192;
    2. Intelligent Decision Making and Big Data Application, Beijing International Science and Technology Cooperation Base, Beijing 100192
  • Received:2022-05-10 Revised:2022-07-01 Published:2022-11-04
针对中小企业的信用风险预测问题,文章提出了一种基于长短期记忆网络(Long Short-Term Memory,LSTM)-卷积神经网络(ConvolutionalNeural Network,CNN)的中小企业信用风险预测方法.首先,依据国标《企业信用评价指标》,结合中小企业的特点,构建中小企业信用风险预测指标体系,包括守信意愿、守信能力和守信表现三方面的财务与非财务指标;然后,优化LSTM-CNN的网络结构和参数,并使用Dropout方法与Batch归一化方法防止过拟合;最后,采集上市中小企业数据,对数据进行缺失值、标准化与过采样处理,利用LSTM-CNN自动提取信用风险特征,并进行信用风险预测.实验结果表明,文章构建的指标体系能够更为全面的反映信用风险状况,基于LSTM-CNN的中小企业信用风险预测效果优于其他信用风险预测模型,克服了传统方法无法对中小企业时序数据进行动态预测、忽视中小企业发展潜力与时间延续性的局限.
Aiming at the credit risk prediction of small and medium-sized enterprises,this paper proposes a credit risk prediction method based on Long Short-Term Memory (LSTM)-Convolutional Neural Network (CNN) of small and medium-sized enterprises.Firstly,according to the national standard "Enterprise Credit Evaluation Index" and the characteristics of small and medium-sized enterprises,this paper proposes a credit risk prediction index system of small and medium-sized enterprises.The index system includes three kinds of financial and non-financial indicators:Credit intention,credit ability and credit performance.Then,this paper optimizes the network structure and parameters of LSTM-CNN,and applies Dropout and Batch Normalization methods to prevent over fitting.Finally,collecting the information of the listed small and medium-sized enterprises,and after missing value processing,standardization and oversampling,LSTM-CNN is applied to automatically extract features and predict credit risk.The experimental results show that the index system of this paper comprehensively reflect the credit risk situation.The credit risk prediction effect of small and medium-sized enterprises based on LSTM-CNN is better than the comparative models,which overcomes the limitations of traditional methods that cannot dynamically predict the time series data,and ignore the development potential and time continuity of small and medium-sized enterprises.

MR(2010)主题分类: 

()
[1] 范柏乃,朱文斌.中小企业信用评价指标的理论遴选与实证分析.科研管理, 2003, 24(6):83-88.(Fan B N, Zhu W B. An empirical study and choose of credit evaluation indexes of small-medium enterprises. Science Research Management, 2003, 24(6):83-88.)
[2] 王重仁,韩冬梅.基于卷积神经网络的互联网金融信用风险预测研究.微型机与应用, 2017, 36(24):44-46, 50.(Wang C R, Han D M. Prediction of credit riskin Internet financial industry based on convolutional neural network. Microcomputer&Its Applications, 2017, 36(24):44-46, 50.)
[3] 王鑫,王莹,陈进东.我国中小微企业信用评价研究现状与发展趋势.征信, 2021, 39(5):62-70.(Wang X, Wang Y, Chen J D. Research status and development trend of credit evaluation of micro, small and medium-sized enterprises in China. Credit Reference, 2021, 39(5):62-70.)
[4] Wei N, Su Y. Credit risk evaluation of SMEs based on supply chain financing. Management Science and Engineering, 2016, 10(2):51-56.
[5] 张朝辉,刘佳佳,冉惠.基于贝叶斯与神经网混合算法的电商信用评价方法研究.情报科学, 2020, 38(2):81-87.(Zhang C H, Liu J J, Ran H. Research on e-commerce credit evaluation method based on Bayesian and neural network hybrid algorithms. Information Science, 2020, 38(2):81-87.)
[6] 张发明,王伟明,李小霜. TOPSIS-GRA法下的企业动态信用评价方法及其应用.运筹与管理, 2018, 27(9):132-138.(Zhang F M, Wang W M, Li X S. Enterprise dynamic credit revaluation method based on TOPSIS-GRA and its application. Operations Research and Management Science, 2018, 27(9):132-138.)
[7] Szegedy C, Vanhoucke V, Ioffe S, et al. Rethinking the inception architecture for computer vision. Computer Science, 2015.
[8] 李新娟.基于长短期记忆与卷积结合的深度学习模型及试应用.硕士论文.上海大学,上海, 2018.(Li X J. Deep learning model based on the combination of long and short memory and convolution and its application. Master Thesis. Shanghai University, Shanghai, 2018.)
[9] 赵红蕊,薛雷.基于LSTM-CNN-CBAM模型的股票预测研究.计算机工程与应用, 2021, 57(3):203-207.(Zhao H R, Xue L. Research on stock forecasting based on LSTM-CNN-CBAM model. Computer Engineering and Applications, 2021, 57(3):203-207.)
[10] GB/T 23794-2015,企业信用评价指标.(GB/T 23794-2015, Enterprise credit evaluation indicators.)
[11] 肖汉杰,桑秀丽.相关向量机超参数优化的网络安全态势预测.计算机应用, 2015, 35(7):1888-1891.(Xiao H J, Sang X L. Network security situation prediction based on hyper parameter optimization of relevance vector machine. Journal of Computer Applications, 2015, 35(7):1888-1891.)
[12] Nair V, Hinton G E. Rectified linear units improve restricted Boltzmann machines. Proceedings of the 27th International Conference on Machine Learning (ICML-10), 2010, 807-814.
[13] Ioffe S, Christlan S. Batch normalization:Accelerating deep network training by reducing internal covariate shift. International Conference on Machine Learning, 2015.
[14] Koltchinskii V, Panchenko D. Complexities of convex combinations and bounding the generalization error in classification. The Annals of Statistics, 2005, 33(4):40-42.
[15] 周颖,苏小婷.基于最优指标组合的企业信用风险预测.系统管理学报, 2021, 30(5):817-838.(Zhou Y, Su X T. Credit risk prediction of company based on optimal feature set. Journal of Systems&Management, 2021, 30(5):817-838.)
[16] 李毅,姜天英,刘亚茹.基于不平衡样本的互联网个人信用评估研究.统计与信息论坛, 2017, 32(2):84-90.(Li Y, Jiang T Y, Liu Y R. Internet personal credit assessment research based on the perspective of unbalanced sample. Statistics&Information Forum, 2017, 32(2):84-90.)
[1] 谢楠, 何海涛, 王宗润. 社交网络环境下中小企业投融资市场的风险共担机制研究[J]. 系统科学与数学, 2022, 42(7): 1753-1768.
[2] 郭晓春, 马玉梅, 曹萍萍. 网络环境下考虑目标顾客偏好的服务产品选择方法[J]. 系统科学与数学, 2022, 42(7): 1769-1787.
[3] 张佳敏, 王莹. 考虑在线评论的中小食品企业品牌价值评价[J]. 系统科学与数学, 2022, 42(10): 2665-2679.
[4] 王秋雨, 张举勇. 一种基于三维对齐方式的深度学习人脸识别算法[J]. 系统科学与数学, 2021, 41(7): 2035-2045.
[5] 琚春华,陈冠宇,鲍福光. 基于kNN-Smote-LSTM的消费金融风险检测模型[J]. 系统科学与数学, 2021, 41(2): 481-498.
[6] 张立文,朱周帆,郝鸿. 基于深度学习的乘用车市场预警模型研究[J]. 系统科学与数学, 2020, 40(11): 2136-2150.
[7] 王溪,明扬,洪奕光. 基于群论的深度卷积网络分析[J]. 系统科学与数学, 2019, 39(2): 219-227.
阅读次数
全文


摘要