Certificate: View Certificate
Published Paper PDF: PDF
Confirmation Letter: View
DOI: https://doi.org/10.63345/ijrmeet.org.v12.i10.9
John Smith
Independent Researcher
Canada
Abstract— The increasing sophistication and volume of digital fraud, particularly in cloud-based financial platforms, renders traditional rule-based and static machine learning approaches inadequate. The relational nature of fraud—where illicit activities are orchestrated across networks of users, devices, and accounts—necessitates a paradigm shift toward graph-based methods. This paper proposes a unified framework that integrates Temporal Graph Neural Networks (TGNs) for real-time fraud pattern detection with an automated security response system inspired by recent patent developments. We demonstrate that TGNs, with their capacity to capture complex, dynamic relational patterns, significantly outperform conventional methods, achieving 12–25% AUROC improvement. Furthermore, we introduce a novel “contextual response module” that leverages the structural and temporal insights from the TGN to automatically deploy security policies, isolating threats and preventing propagation without manual intervention. Our framework addresses key challenges in the field, including class imbalance, scalability to billion-edge graphs, and the need for model interpretability through counterfactual subgraph explanations. By combining high-accuracy detection with immediate, context-aware automated response, this unified framework offers a robust, scalable, and practical solution for next-generation cloud security, validated by real-world deployment scenarios.
Keywords— Graph Neural Networks, Fraud Detection, Temporal Networks, Automated Response, Cloud Security, Financial Fraud, DevSecOps, Anomaly Detection, Explainable AI.
REFERENCES
[1] D. Cheng et al., “Graph neural networks for financial fraud detection: A review,” Frontiers of Computer Science, 2023.
[2] “Fraud detection using graph neural networks: A survey,” in Bioinspired Intelligent Systems, 2024.
[3] “Causal-guided dual-timescale graph neural networks for explainable fraud detection in mobile payment systems,” Springer, 2022.
[4] “Advancements and challenges in fraud detection and fairness-aware machine learning: A comprehensive review,” IEEE Xplore, 2023.
[5] E. Li et al., “Graph learning-empowered financial fraud detection: Progress and future directions,” Intelligent Computing, 2024.
[6] D. Salda√±a-Ulloa, G. De Ita Luna, and J. R. Marcial-Romero, “A temporal graph network algorithm for detecting fraudulent transactions on online payment platforms,” Algorithms, vol. 17, no. 12, p. 552, 2024.
[7] M. Z. Hossain George et al., “Machine learning for fraud detection in digital banking: A systematic literature review,” arXiv preprint, 2024.
[8] S. Singh, “Intelligent apparatus to monitor and auto deploy security policy rules on container based cloud infrastructure leveraging NFT and quantum knowledge graph,” U.S. Patent Appl. 2024/0273211 A1, Aug. 15, 2024.
[9] “Identifying anomalous activities in a cloud computing environment,” U.S. Patent Appl. 2024/0007492 A1, 2024.
[10] T. N. Kipf and M. Welling, “Semi-supervised classification with graph convolutional networks,” in Proc. Int. Conf. Learning Representations (ICLR), Toulon, France, 2017.
[11] P. Veliƒçkoviƒá, G. Cucurull, A. Casanova, A. Romero, P. Li√≤, and Y. Bengio, “Graph attention networks,” in Proc. Int. Conf. Learning Representations (ICLR), Vancouver, Canada, 2018.
[12] E. Rossi, B. Chamberlain, F. Frasca, D. Eynard, F. Monti, and M. Bronstein, “Temporal graph networks for deep learning on dynamic graphs,” arXiv:2006.10637, 2020. [Online]. Available: https://doi.org/10.48550/arXiv.2006.10637
[13] H. Mistry, A. Goswami, and C. Mavani, “Automated anomaly detection and response system for enhancing cloud security,” Zenodo, 2024. [Online]. Available: https://doi.org/10.5281/zenodo.18778285
