Selective task-aware attribute promotion into graph nodes improves GNN classification on relational and tabular data compared to schema-based and heuristic graph construction.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
From Features to Structure: Task-Aware Graph Construction for Relational and Tabular Learning with GNNs
Selective task-aware attribute promotion into graph nodes improves GNN classification on relational and tabular data compared to schema-based and heuristic graph construction.