Detection of Coordinated Cyber Attacks in Financial Networks Using Multi Layer Graph Embeddings

Alina D Hedrick

Detection of Coordinated Cyber Attacks in Financial Networks Using Multi Layer Graph Embeddings

Authors

Alina D Hedrick

Cybersecurity Data Scientist , Austria

Keywords:

Cybersecurity, Financial Networks, Graph Embeddings, Multi-layer Graphs, Anomaly Detection, Coordinated Attacks, FinTech Security

Synopsis

Coordinated cyberattacks on financial networks pose significant threats, often exploiting weak links across systems and actors. Traditional anomaly detection fails to capture complex relational behaviors spanning multiple financial layers. This paper proposes using multi-layer graph embeddings to model interdependencies and identify coordinated malicious activity. We explore graph-based machine learning methods, evaluate current detection models, and propose an improved architecture integrating cross-entity relationships, temporal dynamics, and edge semantics.

References

(1) Jin, R., Zhang, Y., & Liu, B. (2020). Detecting threat propagation in financial networks via dynamic multilayer graphs. IEEE Transactions on Cybernetics, 50(12), 5004–5017.

(2) Wang, H., & Liu, Q. (2019). Meta-path-based learning for anomaly detection in heterogeneous financial graphs. Journal of Financial Data Science, 1(3), 58–73.

(3) Zhang, L., Wang, M., & Li, S. (2022). Temporal graph embeddings for sequential fraud detection in digital payment systems. Expert Systems with Applications, 193, 116366.

(4) Li, Y., & Tan, W. (2020). Multi-relational graph attention networks for cyber threat intelligence in financial APIs. ACM Computing Surveys, 53(6), 1–26.

(5) Yang, D., Chen, F., & Zhao, J. (2022). Graph-based intrusion detection in decentralized finance networks. International Journal of Cybersecurity Intelligence and Cybercrime, 5(1), 23–38.

(6) Anugula Sethupathy, U.K. (2018). Maturity assessment frameworks for DevOps and Agile teams: Tools and metrics. International Journal of Engineering Technology Research & Management (IJETRM), 2(10), 66–76. https://zenodo.org/records/16948086

(7) Tang, J., Qu, M., Wang, M., Zhang, M., Yan, J., & Mei, Q. (2015). LINE: Large-scale information network embedding. Proceedings of the 24th International Conference on World Wide Web, 1067–1077.

(8) Perozzi, B., Al-Rfou, R., & Skiena, S. (2014). DeepWalk: Online learning of social representations. Proceedings of the 20th ACM SIGKDD, 701–710.

(9) Zhang, Z., Li, J., & Sun, Y. (2020). Heterogeneous graph neural network for fraud detection. Proceedings of the Web Conference (WWW), 689–698.

(10) Anugula Sethupathy, U.K. (2019). Real-time inventory visibility using event streaming and analytics in retail systems. International Journal of Novel Research and Development, 4(4), 23–33. https://doi.org/10.56975/ijnrd.v4i4.309064

(11) Jin, R., Zhang, Y., & Liu, B. (2020). Detecting threat propagation in financial networks via dynamic multilayer graphs. IEEE Transactions on Cybernetics, 50(12), 5004–5017.

(12) Jin, R., Zhang, Y., & Liu, B. (2020). Detecting threat propagation in financial networks via dynamic multilayer graphs. IEEE Transactions on Cybernetics, 50(12), 5004–5017.

(13) Wang, H., & Liu, Q. (2019). Meta-path-based learning for anomaly detection in heterogeneous financial graphs. Journal of Financial Data Science, 1(3), 58–73.

(14) Tang, J., Qu, M., Wang, M., Zhang, M., Yan, J., & Mei, Q. (2015). LINE: Large-scale information network embedding. Proceedings of the 24th International Conference on World Wide Web, 1067–1077.

(15) Perozzi, B., Al-Rfou, R., & Skiena, S. (2014). DeepWalk: Online learning of social representations. Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 701–710.

Published

February 27, 2021

Series

Research paper

License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.