Industrial chain risk assessment and early warning model combining hierarchical graph neural network and long short-term memory

doi:10.11772/j.issn.1001-9081.2023101387

Journal of Computer Applications ›› 2024, Vol. 44 ›› Issue (10): 3223-3231.DOI: 10.11772/j.issn.1001-9081.2023101387

• Frontier and comprehensive applications • Previous Articles Next Articles

Industrial chain risk assessment and early warning model combining hierarchical graph neural network and long short-term memory

Xiaoyu HUA¹, Dongfen LI¹, You FU¹, Kejun BI², Shi YING³, Ruijin WANG⁴()

^1.College of Computer Science and Cyber Security （Pilot Software College），Chengdu University of Technology，Chengdu Sichuan 610059，China
^2.College of Computer Science，Sichuan University，Chengdu Sichuan 610065，China
^3.School of Computer Science，Wuhan University，Wuhan Hubei 430072，China
^4.School of Information and Software Engineering，University of Electronic Science and Technology of China，Chengdu Sichuan 610054，China

Received:2023-10-17 Revised:2024-01-29 Accepted:2024-02-05 Online:2024-10-15 Published:2024-10-10
Contact: Ruijin WANG
About author:HUA Xiaoyu， born in 1999， M. S. candidate. His research interests include graph neural network， industrial chain risk assessment and early warning.
LI Dongfen， born in 1979， Ph. D.， associate professor. Her research interests include artificial intelligence， risk assessment and early warning， quantum cryptographic communication.
FU You， born in 1999， M. S. candidate. His research interests include industrial chain risk assessment and early warning， quantum cryptographic communication.
BI Kejun， born in 1982， M. S.， senior engineer. His research interests include industrial Internet， artificial intelligence， construction of industrial chain knowledge graph.
YING Shi， born in 1965， Ph. D.， professor. His research interests include efficient processing and intelligent analysis of big data， machine learning， artificial intelligence， construction of industrial chain knowledge graph.
Supported by:
National Key R&D Program of China(2022YFB3304300)

结合层次图神经网络与长短期记忆的产业链风险评估预警模型

花晓雨¹, 李冬芬¹, 付优¹, 毕可骏², 应时³, 王瑞锦⁴()

^1.成都理工大学计算机与网络安全学院（示范性软件学院），成都 610059
^2.四川大学计算机学院，成都 610065
^3.武汉大学计算机学院，武汉 430072
^4.电子科技大学信息与软件工程学院，成都 610054

通讯作者: 王瑞锦
作者简介:花晓雨（1999—），男，河南信阳人，硕士研究生，CCF会员，主要研究方向：图神经网络、产业链风险评估预警
李冬芬（1979—），女，青海海东人，副教授，博士，CCF会员，主要研究方向：人工智能、风险评估与预警、量子密码通信
付优（1999—），男，四川成都人，硕士研究生，主要研究方向：产业链风险评估预警、量子密码通信
毕可骏（1982—），男，四川德阳人，高级工程师，硕士，主要研究方向：工业互联网、人工智能、产业链知识图谱构建
应时（1965—），男，湖北武汉人，教授，博士，CCF会员，主要研究方向：大数据高效处理与智能分析、机器学习、人工智能、产业链知识图谱构建
王瑞锦（1980—），男，甘肃天水人，副教授，博士，CCF会员，主要研究方向：边缘计算、人工智能、信息加密、产业链风险评估 ruijinwang@uestc.edu.cn。
基金资助:
国家重点研发计划项目(2022YFB3304300)

Abstract

Abstract:

Industrial chain risk assessment and early warning are essential measures to effectively protect the interests of upstream and downstream companies in the industrial chain and mitigate company risks. However， existing methods often need to pay more attention to the propagation effects between upstream and downstream companies in the industrial chain and the opacity of company information， resulting in inaccurate risk assessment of companies and the failure to perceive risks in advance for early warning. To address the above problems， the Hierarchical Graph Neural Network （HiGNN）， an industrial chain risk assessment and early warning model that combined Hierarchical Graph （HG） neural network and Long Short-Term Memory （LSTM）， was proposed. Firstly， an “industrial chain-investment” HG was constructed based on the relationships between upstream and downstream companies and investment activities. Then， a financial feature extraction module was utilized to extract features from multi-quarter financial data of companies， while an investment feature extraction module was utilized to extract features from the investment relationship graph. Finally， an attention mechanism was employed to integrate the financial features with the investment features， enabling risk classification of company nodes through graph representation learning methods. The experimental results on a real integrated circuit manufacturing dataset showed that compared with the Graph ATtention network （GAT） model and the Recurrent Neural Network （RNN） model， the accuracy of the proposed model increased by 14.87% and 22.10%， and the F1-score increased by 12.63% and 16.67% with the 60% training ratio. The proposed model can effectively capture the contagion effect in the industrial chain and improve risk identification capability， which is superior to traditional machine learning methods and graph neural network methods.

Key words: industrial chain risk assessment, Hierarchical Graph (HG) neural network, Long Short-Term Memory (LSTM) network, financial feature extraction, investment feature extraction

摘要：

产业链风险评估预警是有效保护产业链上下游公司利益和减轻公司风险的重要措施。然而，现有方法由于忽视了产业链上下游公司之间的传播效应和公司信息的不透明性，无法准确评估公司风险，且忽略了动态财务数据对产业链的影响，无法提前感知风险，进行风险预警。针对以上问题，提出一种结合层次图（HG）神经网络与长短期记忆（LSTM）的产业链风险评估预警模型（HiGNN）。首先，利用产业链上下游关系和投融资关系构建“产业链-投资”HG；其次，利用财务特征提取模块提取公司多季度财务数据的特征；再次，利用投资特征提取模块提取投资关系图特征；最后，利用注意力机制融合财务特征和投资特征，通过图表示学习方法对公司节点进行风险分类。在真实的集成电路制造业数据集上的实验结果表明，与图注意力网络（GAT）模型、循环神经网络（RNN）模型相比，当训练比率为60%时，所提模型的准确率分别提升了14.87%、22.10%，F1值提升了12.63%、16.67%。所提模型能够有效捕捉产业链中的传染效应，提高风险识别能力，优于传统的机器学习方法和图神经网络方法。

关键词: 产业链风险评估, 层次图神经网络, 长短期记忆网络, 财务特征提取, 投资特征提取

CLC Number:

TP391.4

Xiaoyu HUA, Dongfen LI, You FU, Kejun BI, Shi YING, Ruijin WANG. Industrial chain risk assessment and early warning model combining hierarchical graph neural network and long short-term memory[J]. Journal of Computer Applications, 2024, 44(10): 3223-3231.

花晓雨, 李冬芬, 付优, 毕可骏, 应时, 王瑞锦. 结合层次图神经网络与长短期记忆的产业链风险评估预警模型[J]. 《计算机应用》唯一官方网站, 2024, 44(10): 3223-3231.

Figures/Tables 10

References 27

1	H-C RYU， LEE C. A study on the effects of supply chain risk factors on supply chain capability strategies and corporate performance［J］. Korea International Commercial Review， 2022， 37（1）： 133-158.
2	SYED M W， LI J Z， JUNAID M， et al. An empirical examination of sustainable supply chain risk and integration practices： a performance-based evidence from Pakistan［J］. Sustainability， 2019， 11（19）： 5334.
3	SONG E， PARK M S， JEONG E， et al. Empirical study of supply chain risk management effectiveness［J］. Global Business Administration Review， 2021， 18（5）： 143-162.
4	李政，王思霓.产业链安全风险评估与预警监测体系构建研究［J］.创新科技， 2022， 22（1）：11-20.
	LI Z， WANG S N. Research on the construction of the risk monitoring and assessment and early warning system of industrial chain safety ［J］. Innovation Science and Technology， 2022， 22（1）：11-20.
5	ADHISTI A K， RAHADI R A. Financial risk early warning system for coal mining company in facing disruption of supply chain［J］. Asian Journal of Research in Business and Management， 2022， 4（4）： 98-113.
6	LIU S， CHEN H， HU Z. Research on risk assessment and early warning of supply chain based on extension in the context for new retailing［C］// Proceedings of the Sixth International Forum on Decision Sciences. Singapore： Springer， 2020： 1-14.
7	HE J， ZHANG H. Research on risk assessment of supply chain financing for big data enterprises［C］// Proceedings of the Sixth Symposium of Risk Analysis and Risk Management in Western China. ［S.l.］： Atlantis Press， 2019： 33-38.
8	YIN M， LI G. Supply chain financial default risk early warning system based on particle swarm optimization algorithm［J］. Mathematical Problems in Engineering， 2022， 2022： e7255967.
9	WANG L， HU X. Research on power supply chain risk early warning system based on ACO-SVM algorithm［C］// Proceedings of the 2023 International Symposium on Intelligent Robotics and Systems. Piscataway： IEEE， 2023： 101-105.
10	ZHANG X. Enterprise supply chain risk assessment based on the support vector machine algorithm and fuzzy model［J］. Security and Communication Networks， 2022， 2022： e3692628.
11	YIN L-L， QIN Y-W， HOU Y， et al. A convolutional neural network-based model for supply chain financial risk early warning［J］. Computational Intelligence and Neuroscience， 2022， 2022： 7825597.
12	WU Z， PAN S， CHEN F， et al. A comprehensive survey on graph neural networks［J］. IEEE Transactions on Neural Networks and Learning Systems， 2021， 32（1）： 4-24.
13	CHUNG J， GULCEHRE C， CHO K， et al. Empirical evaluation of gated recurrent neural networks on sequence modeling ［EB/OL］. ［2023-12-17］. .
14	RAFFEL C， SHAZEER N， ROBERTS A， et al. Exploring the limits of transfer learning with a unified text-to-text transformer ［EB/OL］. ［2023-12-17］. .
15	电子科技大学. 基于图神经网络的产业链风险评估模型、方法及介质： CN117236698A［P］. 2023-12-15.
	University of Electronic Science and Technology of China. Industrial chain risk assessment model， method， and medium based on graph neural network：CN117236698A［P］. 2023-12-15.
16	KOSASIH E E， MARGAROLI F， GELLI S， et al. Towards knowledge graph reasoning for supply chain risk management using graph neural networks［J］. International Journal of Production Research， 2024， 62（15）： 5596-5612.
17	DING C， SUN S， ZHAO J. MST-GAT： a multimodal spatial-temporal graph attention network for time series anomaly detection［J］. Information Fusion， 2023， 89： 527-536.
18	DUO L， ZHAO T， GAO S. Guaranteed network risk rating based on graph neural networks［C］// Proceedings of the 2022 2nd International Conference on Electronic Information Engineering and Computer Technology. Piscataway： IEEE， 2022： 391-400.
19	KOSASIH E E， BRINTRUP A. A machine learning approach for predicting hidden links in supply chain with graph neural networks［J］. International Journal of Production Research， 2022， 60（17）： 5380-5393.
20	ZHAO X. Enterprise risk early warning model based on recurrent neural network［J］. International Journal of Neural Network， 2022， 3（4）： 17-25.
21	SHI X， CHEN Z， WANG H， et al. Convolutional LSTM network： a machine learning approach for precipitation nowcasting ［EB/OL］. ［2023-12-17］. .
22	LI Z， SHEN X， JIAO Y， et al. Hierarchical bipartite graph neural networks： towards large-scale e-commerce applications［C］// Proceedings of the 2020 IEEE 36th International Conference on Data Engineering. Piscataway： IEEE， 2020： 1677-1688.
23	PENG S， SUGIYAMA K， MINE T. SVD-GCN： a simplified graph convolution paradigm for recommendation［C］// Proceedings of the 31st ACM International Conference on Information & Knowledge Management. New York： ACM， 2022： 1625-1634.
24	YAN K， LV H， GUO Y， et al. sAMPpred-GAT： prediction of antimicrobial peptide by graph attention network and predicted peptide structure［J］. Bioinformatics， 2023， 39（1）： btac715.
25	KHAN W， ZAKI N， AHMAD A， et al. Node embedding-based graph autoencoder outlier detection for adverse pregnancy outcomes［J］. Scientific Reports， 2023， 13（1）： 19817.
26	迟国泰，王珊珊.基于XGBoost的中国上市公司违约风险预测模型［J］. 系统管理学报， 2024， 33（3）： 735-754 .
	CHI G T， WANG S S. Default risk prediction model for Chinese listed companies based on XGBoost［J］. Journal of Systems & Management， 2024， 33（3）： 735-754.
27	WANG C， DONG Z， HU W， et al. Classification of IoT intrusion detection data based on WGAN-gp and E-GraphSAGE［C］// Proceedings of the Third International Conference on Green Communication， Network， and Internet of Things. ［S.l.］： SPIE， 2023： 1281416.

名称	值
层次图节点数	6 963
层次图边数	128 326
产业链图节点数	1 700
产业链图边数	83 052
投资图数	1 700
投资图节点数	45 263
财务指标数	18
财务指标季度数	49

名称	值
层次图节点数	6 963
层次图边数	128 326
产业链图节点数	1 700
产业链图边数	83 052
投资图数	1 700
投资图节点数	45 263
财务指标数	18
财务指标季度数	49

模型	训练比率/%	F1-score	AUC	Accuracy
XGBoost	60	0.554	0.657	0.521
	40	0.550	0.653	0.520
	20	0.548	0.647	0.514
GCN	60	0.582	0.694	0.562
	40	0.575	0.688	0.560
	20	0.559	0.691	0.558
GAT	60	0.578	0.672	0.558
	40	0.572	0.667	0.553
	20	0.568	0.653	0.547
GraphSAGE	60	0.585	0.690	0.568
	40	0.579	0.685	0.559
	20	0.563	0.682	0.558
RNN	60	0.558	0.652	0.525
	40	0.547	0.656	0.518
	20	0.535	0.653	0.501
Transformer	60	0.564	0.663	0.531
	40	0.549	0.658	0.514
	20	0.524	0.650	0.505
HiGNN	60	0.651	0.736	0.641
	40	0.650	0.729	0.631
	20	0.641	0.722	0.627

模型	训练比率/%	F1-score	AUC	Accuracy
XGBoost	60	0.554	0.657	0.521
	40	0.550	0.653	0.520
	20	0.548	0.647	0.514
GCN	60	0.582	0.694	0.562
	40	0.575	0.688	0.560
	20	0.559	0.691	0.558
GAT	60	0.578	0.672	0.558
	40	0.572	0.667	0.553
	20	0.568	0.653	0.547
GraphSAGE	60	0.585	0.690	0.568
	40	0.579	0.685	0.559
	20	0.563	0.682	0.558
RNN	60	0.558	0.652	0.525
	40	0.547	0.656	0.518
	20	0.535	0.653	0.501
Transformer	60	0.564	0.663	0.531
	40	0.549	0.658	0.514
	20	0.524	0.650	0.505
HiGNN	60	0.651	0.736	0.641
	40	0.650	0.729	0.631
	20	0.641	0.722	0.627

模型	F1-score	AUC	Accuracy
HiGNN\lstm	0.605	0.698	0.597
HiGNN\invest	0.621	0.702	0.612
HiGNN	0.650	0.721	0.627

Industrial chain risk assessment and early warning model combining hierarchical graph neural network and long short-term memory

结合层次图神经网络与长短期记忆的产业链风险评估预警模型

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 10

References 27

Related Articles 15

Recommended Articles

Metrics

[1]	Li’an CHEN, Yi GUO. Text sentiment analysis model based on individual bias information [J]. Journal of Computer Applications, 2024, 44(1): 145-151.
[2]	Hanxiao SHI, Leichun WANG. Short-term power load forecasting by graph convolutional network combining LSTM and self-attention mechanism [J]. Journal of Computer Applications, 2024, 44(1): 311-317.
[3]	Zhiping ZHU, Yan YANG, Jie WANG. Scene graph-aware cross-modal image captioning model [J]. Journal of Computer Applications, 2024, 44(1): 58-64.
[4]	Jiagao WU, Shiwen ZHANG, Yudong JIANG, Linfeng LIU. Social-interaction GAN for pedestrian trajectory prediction based on state-refinement long short-term memory and attention mechanism [J]. Journal of Computer Applications, 2023, 43(5): 1565-1570.
[5]	Haiyu YANG, Wenpu GUO, Kai KANG. Signal modulation recognition method based on convolutional long short-term deep neural network [J]. Journal of Computer Applications, 2023, 43(4): 1318-1322.
[6]	Chunyong YIN, Liwen ZHOU. Unsupervised time series anomaly detection model based on re-encoding [J]. Journal of Computer Applications, 2023, 43(3): 804-811.
[7]	Bennian YU, Yongzhao ZHAN, Qirong MAO, Wenlong DONG, Honglin LIU. Double complex convolution and attention aggregating recurrent network for speech enhancement [J]. Journal of Computer Applications, 2023, 43(10): 3217-3224.
[8]	ZHANG Jun, WU Pengli, SHI Lukui, SHI Jin, PAN Bin. Deep learning model for multi-station temperature prediction combined with MOD11A1 and surface meteorological station data [J]. Journal of Computer Applications, 2023, 43(1): 321-328.
[9]	LIU Hui, MA Xiang, ZHANG Linyu, HE Rujin. Aspect-based sentiment analysis model integrating match-LSTM network and grammatical distance [J]. Journal of Computer Applications, 2023, 43(1): 45-50.
[10]	Yinglü XUAN, Yuan WAN, Jiahui CHEN. Time series classification by LSTM based on multi-scale convolution and attention mechanism [J]. Journal of Computer Applications, 2022, 42(8): 2343-2352.
[11]	Yifei WANG, Lei YU, Fei TENG, Jiayu SONG, Yue YUAN. Resource load prediction model based on long-short time series feature fusion [J]. Journal of Computer Applications, 2022, 42(5): 1508-1515.
[12]	Chong CHEN, Zhu YAN, Jixuan ZHAO, Wei HE, Huaqing LIANG. Prediction of NOx emission from fluid catalytic cracking unit based on ensemble empirical mode decomposition and long short-term memory network [J]. Journal of Computer Applications, 2022, 42(3): 791-796.
[13]	An YANG, Qun JIANG, Gang SUN, Jie YIN, Ying LIU. Power data analysis based on financial technical indicators [J]. Journal of Computer Applications, 2022, 42(3): 904-910.
[14]	Yuli CHEN, Qiang TONG, Tongtong CHEN, Shoulu HOU, Xiulei LIU. Short-term trajectory prediction model of aircraft based on attention mechanism and generative adversarial network [J]. Journal of Computer Applications, 2022, 42(10): 3292-3299.
[15]	Rongkang XI, Manchun CAI, Tianliang LU, Yanlin LI. Tor website traffic analysis model based on self-attention mechanism and spatiotemporal features [J]. Journal of Computer Applications, 2022, 42(10): 3084-3090.