Node Classification On Chameleon 60 20 20

평가 지표

1:1 Accuracy

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	1:1 Accuracy	Paper Title
GAT	63.9 ± 0.46	Graph Attention Networks
ACM-GCNII*	61.66 ± 2.29	Revisiting Heterophily For Graph Neural Networks
GCN	64.18 ± 2.62	Semi-Supervised Classification with Graph Convolutional Networks
ACMII-Snowball-3	67.53 ± 2.83	Revisiting Heterophily For Graph Neural Networks
Geom-GCN*	60.9	Geom-GCN: Geometric Graph Convolutional Networks
GCN+JK	64.68 ± 2.85	Revisiting Heterophily For Graph Neural Networks
ACMII-GCN++	75.93 ± 1.71	Revisiting Heterophily For Graph Neural Networks
ACM-GCNII	58.73 ± 2.52	Revisiting Heterophily For Graph Neural Networks
ACM-Snowball-2	68.51 ± 1.7	Revisiting Heterophily For Graph Neural Networks
ACMII-Snowball-2	67.83 ± 2.63	Revisiting Heterophily For Graph Neural Networks
GraphSAGE	62.15 ± 0.42	Inductive Representation Learning on Large Graphs
Snowball-3	65.49 ± 1.64	Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networks
MLP-2	46.72 ± 0.46	Revisiting Heterophily For Graph Neural Networks
Snowball-2	64.99 ± 2.39	Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networks
GCNII*	62.8 ± 2.87	Simple and Deep Graph Convolutional Networks
ACMII-GCN+	75.51 ± 1.58	Revisiting Heterophily For Graph Neural Networks
MixHop	36.28 ± 10.22	MixHop: Higher-Order Graph Convolutional Architectures via Sparsified Neighborhood Mixing
HH-GraphSAGE	62.98 ± 3.35	Half-Hop: A graph upsampling approach for slowing down message passing
ACM-Snowball-3	68.4 ± 2.05	Revisiting Heterophily For Graph Neural Networks
ACM-SGC-1	63.68 ± 1.62	Revisiting Heterophily For Graph Neural Networks

0 of 38 row(s) selected.