LINKX Benchmarks
twitch-gamers
Structure
1Graphs
168,114Nodes
6,797,557Edges
2Classes
7Node Feature Dim
0.545Homophily
Coverage
56Results
1Variants
8Papers
- NeurIPS 2024: on Large Graphs using Intersecting Communities on Large Graphs using Intersecting Communities NeurIPS 2024
- Smoothed Graph Contrastive Learning via Seamless Proximity Integration Smoothed Graph Contrastive Learning via Seamless Proximity Integration
- ICLR 2023: Mixture of Weak & Strong Experts on Graphs Mixture of Weak & Strong Experts on Graphs ICLR 2023
- ICML 2023: Graph Neural Networks with Learnable and Optimal Polynomial Bases Graph Neural Networks with Learnable and Optimal Polynomial Bases ICML 2023
- ICML 2022: TIDE: Time Derivative Diffusion for Deep Learning on Graphs TIDE: Time Derivative Diffusion for Deep Learning on Graphs ICML 2022
- Deformable Graph Transformer Deformable Graph Transformer
- Augmentation-Free Graph Contrastive Learning with Performance Guarantee Augmentation-Free Graph Contrastive Learning with Performance Guarantee
- NeurIPS 2022: Convolutional Neural Networks on Graphs with Chebyshev Approximation, Revisited Convolutional Neural Networks on Graphs with Chebyshev Approximation, Revisited NeurIPS 2022
Binary node classification Twitch streamer network binary node classification (streaming language). Uses the released sparse node features and a low-homophily graph. Feature-only baselines are essential because structure can be weakly predictive. Evaluated by Accuracy.
Dataset details
Task
Twitch streamer network binary node classification (streaming language). The primary catalog metric is Accuracy.
Data
Twitch streamer network binary node classification (streaming language). The cataloged artifact reports 1 graphs, 2 target classes or tasks.
Features
Each release supplies a sparse node-feature matrix, graph edges, and node labels. Feature meaning varies by source; structure and features are deliberately evaluated without assuming neighboring labels agree.
Splits and evaluation
The catalog records these protocols or variants: Standard split. Evaluation uses Accuracy (higher is better). Exact masks or folds must come from the cited release.
Quirks and pitfalls
These are transductive single-graph tasks with fixed random splits. Label homophily is low and class balance varies; feature-only and structure-only baselines are essential, and similarly named OGB/source datasets may use different labels or splits.
Sources
SOTA Progression
new SOTA
matches SOTA at that time
other proposed result
other baseline result
Leaderboard
| # | Model | Accuracy ↑ | Paper |
|---|---|---|---|
| 1 | DGT | 0.6609
±
0.0022 | Deformable Graph Transformer
no code |
| 2 | ICG_u-NN | 0.6608
±
0.0074 | |
| 3 | LINKX | 0.6606
±
0.0019 | |
| 4 | MoE-H2GCN | 0.6603
±
0.0016 | |
| 5 | H2GCN | 0.6570
±
0.0020 | |
| 6 | MixHop | 0.6564
±
0.0027 | |
| 7 | DGT-light | 0.6559
±
0.0025 | Deformable Graph Transformer
no code |
| 8 | ICG-NN | 0.6527
±
0.0082 | |
| 9 | MixHop | 0.6520
±
0.0012 | Deformable Graph Transformer
no code |
| 10 | OptBasisGNN | 0.6517
±
0.0016 | |
| 11 | JKNet | 0.6508
±
0.0007 | Deformable Graph Transformer
no code |
| 12 | ChebNetII | 0.6503
±
0.0027 | |
| 13 | MoE-GIN-skip | 0.6491
±
0.0022 | |
| 14 | LINK | 0.6485
±
0.0021 | |
| 15 | GraphSAGE | 0.6473
±
0.0011 | Deformable Graph Transformer
no code |
| 16 | MoE-SAGE | 0.6438
±
0.0014 | |
| 17 | GCN | 0.6434
±
0.0012 | Deformable Graph Transformer
no code |
| 18 | MoE-GIN | 0.6432
±
0.0034 | |
| 19 | BernNet | 0.6427
±
0.0031 | |
| 20 | FavardGNN | 0.6426
±
0.0012 | |
| 21 | GATv2 | 0.6415
±
0.0009 | Deformable Graph Transformer
no code |
| 22 | MoE-GCN | 0.6374
±
0.0023 | |
| 23 | GraphMoE-SAGE | 0.6342
±
0.0023 | |
| 24 | GCNII | 0.6339
±
0.0061 | |
| 25 | GT-sparse | 0.6309
±
0.0071 | Deformable Graph Transformer
no code |
| 26 | GAT | 0.6290
±
0.0022 | Deformable Graph Transformer
no code |
| 27 | GraphMoE-GCN | 0.6276
±
0.0022 | |
| 28 | GCN | 0.6274
±
0.0003 | |
| 29 | GPR-GNN | 0.6259
±
0.0038 | |
| 30 | ChebNet | 0.6231
±
0.0037 | |
| 31 | GCN | 0.6218
±
0.0026 | |
| 32 | AF-GCL | 0.6204
±
0.0017 | |
| 33 | GraphSAGE | 0.6199
±
0.0030 | |
| 34 | GPR-GNN | 0.6189
±
0.0029 | |
| 35 | GIN | 0.6176
±
0.0060 | |
| 36 | DeepWalk | 0.6171
±
0.0041 | |
| 37 | DGI | 0.6147
±
0.0056 | |
| 38 | MLP | 0.6114
±
0.0006 | Deformable Graph Transformer
no code |
| 39 | Node2cec | 0.6112
±
0.0029 | |
| 40 | GCN-skip | 0.6105
±
0.0023 | |
| 41 | AdaGCN | 0.6102
±
0.0014 | |
| 42 | BGRL | 0.6093
±
0.0032 | |
| 43 | MLP | 0.6092
±
0.0007 | |
| 44 | TIDE-m | 0.6081
±
0.0004 | |
| 45 | VGAE | 0.6070
±
0.0061 | |
| 46 | GCN | 0.6060
±
0.0019 | |
| 47 | SGC | 0.6047
±
0.0014 | Deformable Graph Transformer
no code |
| 48 | TIDE-s | 0.6040
±
0.0013 | |
| 49 | GRAND | 0.5985
±
0.0003 | |
| 50 | SGCL-B | 0.5938
±
0.0088 |
SOTA performance
set a new SOTA at the time
matched the SOTA at the time