计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (10): 218-226.doi: 10.11896/jsjkx.230900145

• 数据库&大数据&数据科学 • 上一篇    下一篇

基于邻居采样和图注意力机制的产业链风险评估模型

孙鹏钊1, 毕可骏2, 唐潮3, 李冬芬4, 应时5, 王瑞锦1   

  1. 1 电子科技大学信息与软件工程学院 成都 610054
    2 四川大学计算机学院 成都 610065
    3 四川长虹电子控股集团有限公司 四川 绵阳 621000
    4 成都理工大学计算机与网络安全学院 成都 610059
    5 武汉大学计算机学院 武汉 430072
  • 收稿日期:2023-09-26 修回日期:2024-03-08 出版日期:2024-10-15 发布日期:2024-10-11
  • 通讯作者: 王瑞锦(ruijinwang@uestc.edu.cn)
  • 作者简介:(202252090717@std.uestc.edu.cn)
  • 基金资助:
    国家重点研发计划(2022YFB3304300,2022YFB3304303)

Risk Assessment Model for Industrial Chain Based on Neighbor Sampling and GraphAttention Mechanism

SUN Pengzhao1, BI Kejun2, TANG Chao3, LI Dongfen4, YING Shi5, WANG Ruijin1   

  1. 1 School of Information and Software Engineering,University of Electronic Science and Technology of China,Chengdu 610054,China
    2 School of Computer Science,Sichuan University,Chengdu 610065,China
    3 Sichuan Changhong Electronic Holding Group Co.Ltd,Mianyang,Sichuan 621000,China
    4 College of Computer Science and Cyber Security,Chengdu University of Technology,Chengdu 610059,China
    5 School of Computer Science,Wuhan University,Wuhan 430072,China
  • Received:2023-09-26 Revised:2024-03-08 Online:2024-10-15 Published:2024-10-11
  • About author:SUN Pengzhao,born in 1999,postgra-duate,is a member of CCF(No.67770G).His main research interests include graph neural network and artificial intelligence.
    WANG Ruijin,born in 1980,Ph.D,associate professor,is a member of CCF(No.22983S).Hismain research intere-sts include edge computing,artificial intelligence and information security.
  • Supported by:
    National Key R&D Program of China(2022YFB3304300,2022YFB3304303).

PDF (PC)

219

摘要： 风险评估是提高产业链韧性的重要途径,也是降低产业链不稳定性的有效手段。然而,现有风险评估的研究基于供应链结构,忽略了其他因素,无法准确地刻画产业链上下游各节点的关联关系,导致评估效果存在偏差。针对上述问题,考虑到产业链内部各节点相互关联、风险状况多样、存在风险传递的特性,提出了基于邻居采样和图注意力机制的产业链风险评估模型GANS。首先,构建了产业链的异质图,利用“产品-公司”“产品-产品”刻画了产业链节点之间的关联关系,并从产业链中提取财务数据等作为节点的数据特征;其次,提出了基于元路径和公司投融资关联规则的公司关系图生成模块,实现对产业链中公司节点关系的高效转化和结构特征的高效学习;接着,针对生成的多种公司关系图,设计了结合邻居采样和图注意力机制的产业链风险评估模块,对节点邻域特征进行随机采样和聚合,同时结合注意力机制对基于多种公司图的节点特征进行自适应聚合,并通过分类器实现节点级风险评估;最后,依据节点风险等级与节点的结构特征对产业链进行风险评估。实验表明,在真实产业链数据集上,GANS在准确性、F1分数等方面均优于现有的模型,证明了GANS实现产业链风险评估的准确性和有效性。

关键词: 产业链, 风险评估, 异质图, 邻域采样, 图注意力机制

Abstract: Risk assessment is an important way to improve the resilience of the industrial chain and also an effective method to reduce the instability of the industrial chain.However,existing research on risk assessment is based on supply chain structure and neglects other factors,which can not accurately depict the correlation between upstream and downstream nodes in the industrial chain,resulting in biased evaluation results.In response to the above issues,considering the interconnected nature of various nodes within the industry chain,diverse risk situations,and the existence of risk transmission,this paper proposes an industry chain risk assessment model based on graph attention mechanism and neighbor sampling(GANS).Firstly,a heterogeneous graph of the industrial chain is constructed,using “product-company” and “product-product” to depict the correlation between nodes in the industrial chain,and financial data and other data features are extracted from the industrial chain as nodes' data features.Se-condly,a company relationship graph generation module based on meta paths and company investment and financing association rules is proposed to achieve efficient transformation of company node relationships and efficient learning of structural features in the industrial chain.Next,an industry chain risk assessment module based on neighbor sampling and graph attention mechanism is designed for various generated company graphs.The features of node neighbors are randomly sampled and aggregated,and attention mechanism is used to adaptively aggregate node features based on multiple company graphs.Through the classifier,node-level risk assessment is realized.Finally,risk assessment of the industrial chain is conducted based on the risk level and structural features of nodes.Experiments show that GANS outperforms existing models in terms of accuracy and F1 scores on real indu-strial chain datasets.These results demonstrate the accuracy and effectiveness of GANS in implementing industrial chain risk assessment.

Key words: Industry chain, Risk assessment, Heterogeneous graph, Neighbourhood sampling, Graph attention mechanism

中图分类号: 

  • TP391
[1]ZHAO Z Y,TIAN Y X,ZILLANTE G.Modeling and evaluationof the wind power industry chain:A China study[J].Renewable and Sustainable Energy Reviews,2014,31:397-406.
[2]LIU J,DU Z,MA L,et al.Identification and assessment of subway construction risk:an integration of AHP and experts gra-ding method[J].Advances in Civil Engineering,2021,2021:1-18.
[3]SAFAVIAN S R,LANDGREBE D.A survey of decision tree classifier methodology[J].IEEE Transactions on Systems,Man,Cnd cybernetics,1991,21(3):660-674.
[4]XU X.Risk factor analysis combined with deep learning in the risk assessment of overseas investment of enterprises[J].Plos One,2020,15(10):e0239635.
[5]SHARMA S K,BHAT A.Supply Chain Risk Assessment Tools and Techniques in the Automobile Industry:A Survey.[J].IUP Journal of Supply Chain Management,2014,11(1):67-76.
[6]TUMMALA R,SCHOENHERR T.Assessing and managingrisks using the supply chain risk management process(SCRMP)[J].Supply Chain Management:An International Journal,2011,16(6):474-483.
[7]ABDELGAWAD M,FAYEK A R.Fuzzy reliability analyzer:Quantitative assessment of risk events in the construction industry using fuzzy fault-tree analysis[J].Journal of Construction Engineering and Management,2011,137(4):294-302.
[8]MARKMANN C,DARKOW I L,VON DER GRACHT H.ADelphi-based risk analysis Identifying and assessing future challenges for supply chain security in a multi-stakeholder environment[J].Technological Forecasting and Social Change,2013,80(9):1815-1833.
[9]DONG Q,COOPER O.An orders-of-magnitude AHP supplychain risk assessment framework[J].International Journal of Production Economics,2016,182:144-156.
[10]ZHANG D,TANG Y,YAN X.Supply chain risk management of badminton supplies company using decision tree model assisted by fuzzy comprehensive evaluation[J].Expert Systems,2024,41(5):e13275.
[11]DELERIS L A,ERHUN F.Risk management in supply net-works using Monte-Carlo simulation[C]//Proceedings of the Winter Simulation Conference,2005.IEEE,2005.
[12]ZHANG L,HU H,ZHANG D.A credit risk assessment model based on SVM for small and medium enterprises in supply chain finance[J].Financial Innovation,2015,1(1):1-21.
[13]LU S.Enterprise supply chain risk assessment based on im-proved neural network algorithm and machine learning[J].Journal of Intelligent & Fuzzy Systems,2021,40(4):7013-7024.
[14]DUNDULIENE· I,ALZBUTAS R,BALTRUŠONE· K.Traderisk identification and prediction using experts' knowledge and machine learning[C]//2021 IEEE International Conference on Technology and Entrepreneurship(ICTE).IEEE,2021:1-5.
[15]WANG J,LIU S,SU J.Identifying vulnerable nodes in supply chain based on the risk transmission model[C]//2017 6th International Conference on Industrial Technology and Management(ICITM).IEEE,2017:1-5.
[16]HO W,ZHENG T,YILDIZ H,et al.Supply chain risk management:a literature review[J].International Journal of Production Research,2015,53(16):5031-5069.
[17]HE M,JI H,WANG Q,et al.Big data fueled process manage-ment of supply risks:sensing,prediction,evaluation and mitigation[C]//Proceedings of the Winter Simulation Conference 2014.IEEE,2014:1005-1013.
[18]GORI M,MONFARDINI G,SCARSELLI F.A new model for learning in graph domains[C]//Proceedings of the International Joint Conference on Neural Networks,2005.IEEE,2005:729-734.
[19]BRUNA J,ZAREMBA W,SZLAM A,et al.Spectral networks and locally connected networks on graphs[J].arXiv:1312.6203,2013.
[20]HENAFF M,BRUNA J,LECUN Y.Deep Convolutional Net-works on Graph-Structured Data[J].arXiv:1506.05163,2015.
[21]KIPF T N,WELLING M.Semi-supervised classification withgraph convolutional networks[J].arXiv:1609.02907,2016.
[22]DUVENAUD D,MACLAURIN D,AGUILERA-IPARRAGUIRRE J,et al.Convolutional Networks on Graphs for Learning Molecular Fingerprints[J].Advances in Neural Information Processing Systems,2015,28:2224-2232.
[23]ATWOOD J,TOWSLEY D.Diffusion-Convolutional NeuralNetworks[J].Advances in Neural Information Processing Systems,2016,9:2001-2009.
[24]HAMILTON W,YING Z,LESKOVEC J.Inductive representation learning on large graphs[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems(NIPS'17).San Diego:NIPS,2017:1025-1035.
[25]DONG Y,CHAWLA N V,SWAMI A.metapath2vec:ScalableRepresentation Learning for Heterogeneous Networks[C]//Proceedings of the 23rd ACM SIGKDD International Confe-rence on Knowledge Discovery and Data Mining.Halifax NS Canada:ACM,2017:135-144.
[26]SUN Y,HAN J,YAN X,et al.Pathsim:Meta path-based top-k similarity search in heterogeneous information networks[J].Proceedings of the VLDB Endowment,2011,4(11):992-1003.
[27]BI W,XU B,SUN X,et al.Company-as-Tribe:Company Financial Risk Assessment on Tribe-Style Graph with Hierarchical Graph Neural Networks[C]//Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining.2022:2712-2720.
[28]VASWANI A,SHAZEER N,PARMAR N,et al.Attention isall you need[C]//Proceedings of the 31st International Confe-rence on Neural Information Processing Systems(NIPS'17).San Diego:NIPS,2017:6000-6010.
[29]VELIČKOVIĆ P,CUCURULL G,CASANOVA A,et al.Graph attention networks[J].arXiv:1710.10903,2017.
[30]WANG X,JI H,SHI C,et al.Heterogeneous graph attention network[C]//The world wide web conference(WWW'19).New York:Association for Computing Machinery,2019:2022-2032.
[31]MA Y Y,ZHUANG X T,LI L X.Research on the relationships of the domestic mutual investment of China Based on the cross-shareholding networks of the listed companies[J].Physica A:Statistical Mechanics and its Applications,2011,390(4):749-759.
[32]MANNERS B J.Supply chain risk management:understandingemerging threats to global supply chains[M].Kogan Page Publishers,2017:12-14.
[33]WANG L J,CHEN J.Research on Risk Assessment of Equipment Manufacturing Industry Chain Based on Grey Cluster Analysis[J].Modern Business,2017,33:44-45.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发