Evaluation of multiple graph neural network models—GCN, GAT, GraphSAGE, MPNN and DGI—for node classification on graph-structured data. Preprocessing includes feature normalization and adjacency-matrix regularization, and an ensemble of model predictions boosts performance. The best ensemble achieves 83.47% test accuracy.
deep-neural-networks deep-learning ensemble-learning regularization loss-functions relu-layer node-classification graphsage graph-neural-networks pytorch-implementation deep-infomax deep-graph-infomax gcn-architecture feature-normalization mpnn-model pytorch-tensors graph-attention-network softmax-function dense-array
-
Updated
May 12, 2025 - Jupyter Notebook