Graph Reduction Techniques for Instance Selection: Comparative and Empirical Study
Artificial Intelligence Review · 2025
Zahiriddin Rustamov
PhD Student · UAEU & KU Leuven
Graph-based data selection.
Making machine learning efficient.
Drag a node to connect graphs
About
I'm a PhD student at UAEU and KU Leuven, researching graph-based approaches to data selection in machine learning. My work focuses on making ML more efficient by intelligently selecting the most informative training instances.
My research interests include instance selection, graph neural networks, attention mechanisms, and scalable machine learning methods.
Publications
GAIS: A Novel Approach to Instance Selection with Graph Attention Networks
arXiv · 2024
Scalable Graph Attention-based Instance Selection via Mini-Batch Sampling and Hierarchical Hashing
ResearchGate · 2024
GAT-RWOS: Graph Attention-Guided Random Walk Oversampling for Imbalanced Data Classification
ICKG 2024 · 2024
1 / 2
News
2025
Graph Reduction paper published in Artificial Intelligence Review
Our comparative study on graph reduction techniques for instance selection is now published.
Read more →2024
GAIS paper available on arXiv
Released our paper on Graph Attention-based Instance Selection with code.
Read more →GAT-RWOS presented at ICKG 2024
Presented our work on graph attention-guided oversampling for imbalanced data classification.