Back to Work
Work GNN + GAT: Knowledge Representation Research

GNN + GAT: Knowledge Representation Research

Category

RESEARCH · IEEE BEST PAPER

Role

Lead researcher and model developer

Timeline

Final year project → GC4T 2025

Published

IEEE · GC4T 2025

86.4%Accuracy
85.1%F1-Score
Best PaperGC4T 2025
3 datasetsCora / PubMed / DBLP
01SIGNAL

Existing GNN architectures struggle with multi-relational data and explainability in healthcare knowledge graphs — a gap between graph representation research and clinical application.

02ARCHITECTURE
  • Designed MetaGAT model architecture, ran benchmarks across Cora, PubMed, DBLP datasets, wrote and submitted the paper, presented at conference.
03OUTCOME

86.4% accuracy, 85.1% F1-score. Best Paper Award. Applied framework to disease prediction and recommender systems. Published under IEEE.

Shipped
04EDGE

Prioritised benchmarkable accuracy on standard datasets over clinical validation (which would require IRB-level access). Explainability layer is v2 research.

The MetaGAT architecture addressed a problem the community has been circling for a while — multi-relational graph learning with interpretability built into the design rather than added after. The benchmark results held up across all three datasets.

Review Panel

IEEE GC4T 2025 · Best Paper · Pune Section

The Outcome

Awarded Best Paper at GC4T 2025 by IEEE Pune Section. The MetaGAT framework demonstrated substantial leaps in predicting relationships across complex multi-relational graphical data, laying foundational architecture for subsequent healthcare recommender models and disease prediction pipelines.