IEA-GNN: Anchor-aware graph neural network fused with information entropy for node classification and link prediction

Peiliang Zhang, Jiatao Chen, Chao Che, Liang Zhang, Bo Jin, Yongjun Zhu

Abstract: Highlights•Enhancing graph representation learning with information entropy-based anchor selection.•Calculating nodes entropy with information contribution to avoid anchor point aggregation.•Introducing global location information to highlight the differences between nodes.•The performance of IEA-GNN is more stable than stochastic anchor point selection models.