Node Classification On Pubmed With Public

Métriques

Accuracy

Résultats

Résultats de performance de divers modèles sur ce benchmark

Nom du modèle	Accuracy	Paper Title	Repository
DAGNN (Ours)	80.5 ± 0.5	Towards Deeper Graph Neural Networks
GGNN	75.8%	Gated Graph Sequence Neural Networks
ChebyNet	69.8%	Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering
SuperGAT MX	81.7%	How to Find Your Friendly Neighborhood: Graph Attention Design with Self-Supervision	-
LinkDistMLP	72.41%	Distilling Self-Knowledge From Contrastive Links to Classify Graph Nodes Without Passing Messages
G-APPNP	80.95%	Pre-train and Learn: Preserve Global Information for Graph Neural Networks
Truncated Krylov	81.7%	Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networks
GraphSAGE	76.8%	Inductive Representation Learning on Large Graphs
SSP	80.06 ± 0.34%	Optimization of Graph Neural Networks with Natural Gradient Descent
GCN	81.12 ± 0.52	Classic GNNs are Strong Baselines: Reassessing GNNs for Node Classification
AIR-GCN	80%	GraphAIR: Graph Representation Learning with Neighborhood Aggregation and Interaction
OGC	83.4%	From Cluster Assumption to Graph Convolution: Graph-based Semi-Supervised Learning Revisited
LanczosNet	78.3 ± 0.3	LanczosNet: Multi-Scale Deep Graph Convolutional Networks
GCN(predicted-targets)	80.42%	GraphMix: Improved Training of GNNs for Semi-Supervised Learning
Snowball (linear)	79.10%	Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networks
GGCM	80.8%	From Cluster Assumption to Graph Convolution: Graph-based Semi-Supervised Learning Revisited
SSGC	80.4	Simple Spectral Graph Convolution
Graph-MLP	79.91	Graph Entropy Minimization for Semi-supervised Node Classification
GCN+DropEdge	79.60%	-	-
CoLinkDistMLP	75.41%	Distilling Self-Knowledge From Contrastive Links to Classify Graph Nodes Without Passing Messages

0 of 37 row(s) selected.