RAPS-Net: A Risk-Aware CNN-LSTM Framework for Cross-Domain Risk Prediction and Dynamic Security Control in Cloud Payment Supply Chains
DOI:
https://doi.org/10.71222/jzx1yv31Keywords:
Cloud Payment Security, Supply Chain Risk, Risk-Aware Access Control, CNN-LSTM, IAM Permission Risk, Dynamic Security ControlAbstract
Cloud payment systems are increasingly deployed on cloud platforms and tightly coupled with complex supply chain ecosystems, including suppliers, logistics services, and financial intermediaries. In such environments, security risks are not only driven by malicious payment behaviors, but are also significantly amplified by misconfigured Identity and Access Management (IAM) policies and supply chain disruptions. In particular, excessive authorization, permission drift, and abnormal access behaviors may propagate across payment supply chains under volatile operational conditions, posing substantial threats to payment availability and security. However, existing payment security mechanisms typically rely on static access control or isolated risk analysis, lacking predictive capability and adaptive security responses. To address these challenges, this paper proposes RAPS-Net, a risk-aware CNN-LSTM framework that integrates permission risk sensing, cross-domain risk prediction, and dynamic security control for cloud payment supply chains. RAPS-Net jointly models IAM permission risks, supply chain operational risks, and payment system states, and employs a hybrid CNN-LSTM architecture to capture short-term cross-domain risk coupling patterns as well as long-term risk evolution trends. Based on the predicted risk levels, a risk-aware access control mechanism is designed to dynamically adjust cloud permissions and proactively mitigate potential security threats. Experimental results on an integrated cloud payment supply chain dataset demonstrate that RAPS-Net consistently outperforms representative baseline models. Specifically, RAPS-Net reduces RMSE to 0.108, achieving approximately 19% improvement over LSTM and 11% over CNN-LSTM, while obtaining the highest F1-score of 0.87. These results validate the effectiveness of jointly modeling permission risks and supply chain dynamics for accurate risk prediction and adaptive security control in cloud payment environments.References
1. B. Liu, Q. Sun, and L. Wei, "Multimodal Forgery Recognition Algorithm and System Design for AI Frauds," In Proceedings of the 2nd International Symposium on Integrated Circuit Design and Integrated Systems, September, 2025, pp. 156-160. doi: 10.1145/3772326.3774725
2. S. O. Olabanji, O. O. Olaniyi, C. S. Adigwe, O. J. Okunleye, and T. O. Oladoyinbo, "AI for Identity and Access Management (IAM) in the cloud: Exploring the potential of artificial intelligence to improve user authentication, authorization, and access control within cloud-based systems," Authorization, and Access Control within Cloud-Based Systems (January 25, 2024), 2024. doi: 10.9734/ajrcos/2024/v17i3423
3. L. Golightly, P. Modesti, R. Garcia, and V. Chang, "Securing distributed systems: A survey on access control techniques for cloud, blockchain, IoT and SDN," Cyber Security and Applications, vol. 1, p. 100015, 2023. doi: 10.1016/j.csa.2023.100015
4. N. Alharbe, A. Aljohani, M. A. Rakrouki, and M. Khayyat, "An access control model based on system security risk for dynamic sensitive data storage in the cloud," Applied Sciences, vol. 13, no. 5, p. 3187, 2023. doi: 10.3390/app13053187
5. Z. Lin, and B. Wang, "Adaptive load balancing algorithms for cloud computing distributed systems," In IET Conference Proceedings CP952, October, 2025, pp. 1319-1325. doi: 10.1049/icp.2025.4664
6. M. Saqib, D. Mehta, F. Yashu, and S. Malhotra, "Adaptive security policy management in cloud environments using reinforcement learning," In 2025 International Conference on Metaverse and Current Trends in Computing (ICMCTC), April, 2025, pp. 1-10. doi: 10.1109/icmctc62214.2025.11196240
7. V. T. Madireddy, "Graph neural network based adaptive threat detection for cloud identity and access management logs," arXiv preprint arXiv:2512.10280, 2025.
8. M. M. Bassiouni, R. K. Chakrabortty, O. K. Hussain, and H. F. Rahman, "Advanced deep learning approaches to predict supply chain risks under COVID-19 restrictions," Expert Systems with Applications, vol. 211, p. 118604, 2023. doi: 10.1016/j.eswa.2022.118604
9. W. A. Zogaan, N. Ajabnoor, and A. A. Salamai, "Leveraging deep learning for risk prediction and resilience in supply chains: insights from critical industries," Journal of Big Data, vol. 12, no. 1, p. 94, 2025. doi: 10.1186/s40537-025-01143-4
10. W. Gamaleldin, O. Attayyib, M. M. Alnfiai, F. A. Alotaibi, and R. Ming, "A hybrid model based on CNN-LSTM for assessing the risk of increasing claims in insurance companies," PeerJ Computer Science, vol. 11, p. e2830, 2025. doi: 10.7717/peerj-cs.2830
11. A. Gupta, and S. Remella, "Privacy-Preserving Smart and Secure Contract Solutions for Digital Supply Chain Payments," International Journal of AI, BigData, Computational and Management Studies, vol. 6, no. 4, pp. 232-240, 2025. doi: 10.63282/3050-9416.ijaibdcms-v6i4p127
12. B. Su, G. Gui, S. Xu, and S. Shen, "Study on Real Estate Investment Risk Assessment and Decision Support System Driven by Fintech," In Proceedings of the 2nd International Symposium on Integrated Circuit Design and Integrated Systems, September, 2025, pp. 168-174. doi: 10.1145/3772326.3774727
