This repository implements a framework for agile and fair Enterprise Architecture (EA) analytics in decentralized environments. EA is modeled as a federated graph of organizational entities, where Graph Attention Networks (GATs) capture structural dependencies, contrastive representation learning produces robust cross-silo embeddings, and Nash Bargaining guarantees fairness and stability during federated aggregation. The system effectively functions as a digital twin of EA, enabling privacy-preserving, collaborative analytics that improve governance, accountability, and strategic decision-making.
Paper link (DOI): https://doi.org/10.1109/ICAEA69058.2025.11301549
The full paper is available on my ResearchGate profile. If you find this code useful in your work, please cite the following paper:
S. Shafaati, M. Rahmani and S. H. Erfani, “Agile and Fair Enterprise Architecture Analytics with Federated Graph Contrastive Learning and Nash Bargaining,” Proceedings of the 9th Iranian Conference on Advances in Enterprise Architecture (ICAEA), Tehran, Iran, 2025, pp. 1–6. doi: 10.1109/ICAEA69058.2025.11301549