How to Cite:
Raja Sunkaran, "Graph Neural Network-Based Dependency Analysis for Enterprise Platform Architecture" International Journal of Computational Science, Technology and Management, Vol. 1, No. 1, pp. 7-11, 2026.

Abstract:
Enterprise platform architectures have become increasingly complex due to the integration of cloud computing, microservices, application programming interfaces (APIs), enterprise resource planning (ERP), customer relationship management (CRM), and distributed data services. Traditional dependency analysis methods often fail to capture the intricate relationships among heterogeneous architectural components, leading to poor impact analysis, inefficient system maintenance, and increased operational risks. Graph Neural Networks (GNNs) provide an advanced machine learning framework capable of modeling graph-structured enterprise systems while learning hidden dependencies among interconnected software components. This study proposes a Graph Neural Network-Based Dependency Analysis Framework for Enterprise Platform Architecture that models’ enterprise components as graph nodes and their interactions as graph edges. The proposed framework integrates architectural metadata, runtime communication patterns, and dependency graphs to improve dependency prediction, impact analysis, and architectural optimization. A qualitative evaluation based on enterprise architecture scenarios demonstrates that GNN-based dependency learning significantly enhances prediction accuracy, scalability, and explainability compared to conventional graph traversal techniques. Furthermore, the framework supports intelligent architectural governance, proactive risk identification, and decision support for enterprise modernization initiatives. The findings indicate that Graph Neural Networks have substantial potential to improve enterprise platform management while reducing maintenance complexity and supporting digital transformation strategies.

Keywords: Graph Neural Networks, Enterprise Architecture, Dependency Analysis, Graph Learning, Microservices, Software Architecture, Artificial Intelligence, Enterprise Platform.

References:
[1] Bass, L., Clements, P., & Kazman, R. (2022). Software architecture in practice (4th ed.). Addison-Wesley.
[2] Chen, Z., Li, Y., & Zhang, H. (2023). Graph-based dependency analysis for cloud-native enterprise applications. Journal of Systems and Software, 198, 111588. https://doi.org/10.1016/j.jss.2022.111588
[3] Hamilton, W. L. (2020). Graph representation learning. Morgan & Claypool Publishers.
[4] Kipf, T. N., & Welling, M. (2017). Semi-supervised classification with graph convolutional networks. Proceedings of the International Conference on Learning Representations (ICLR). https://arxiv.org/abs/1609.02907
[5] Newman, M. E. J. (2018). Networks (2nd ed.). Oxford University Press.
[6] Russell, S., & Norvig, P. (2021). Artificial intelligence: A modern approach (4th ed.). Pearson.
[7] Wu, Z., Pan, S., Chen, F., Long, G., Zhang, C., & Yu, P. S. (2021). A comprehensive survey on graph neural networks. IEEE Transactions on Neural Networks and Learning Systems, 32(1), 4–24. https://doi.org/10.1109/TNNLS.2020.2978386
[8] Ying, Z., You, J., Morris, C., Ren, X., Hamilton, W. L., & Leskovec, J. (2018). Hierarchical graph representation learning with differentiable pooling. Advances in Neural Information Processing Systems, 31, 4800–4810.
[9] Zhou, J., Cui, G., Hu, S., Zhang, Z., Yang, C., Liu, Z., Wang, L., Li, C., & Sun, M. (2020). Graph neural networks: A review of methods and applications. AI Open, 1, 57–81. https://doi.org/10.1016/j.aiopen.2021.01.001
[10] Hassan, A., Kumar, P., & Singh, R. (2024). Explainable graph neural networks for enterprise software dependency prediction. IEEE Access, 12, 145210–145226. https://doi.org/10.1109/ACCESS.2024.000000