Hypergraph Benchmarks
AMAP
Structure
1Graphs
7,650Nodes
119,081Edges
8Classes
745Node Feature Dim
Coverage
113Results
1Variants
11Papers
- Aggregation-aware MLP: An Unsupervised Approach for Graph Message-passing Aggregation-aware MLP: An Unsupervised Approach for Graph Message-passing
- Non-Dissipative Graph Propagation for Non-Local Community Detection Non-Dissipative Graph Propagation for Non-Local Community Detection
- CDC: A Simple Framework for Complex Data Clustering CDC: A Simple Framework for Complex Data Clustering
- Provable Filter for Real-world Graph Clustering Provable Filter for Real-world Graph Clustering
- Reinforcement Graph Clustering with Unknown Cluster Number Reinforcement Graph Clustering with Unknown Cluster Number
- Self-Contrastive Graph Diffusion Network Self-Contrastive Graph Diffusion Network
- GANN: Graph Alignment Neural Network for Semi-Supervised Learning GANN: Graph Alignment Neural Network for Semi-Supervised Learning
- AAAI 2022: Hard Sample Aware Network for Contrastive Deep Graph Clustering [5] Hard Sample Aware Network for Contrastive Deep Graph Clustering [5] AAAI 2022
- GraphLearner: Graph Node Clustering with Fully Learnable Augmentation GraphLearner: Graph Node Clustering with Fully Learnable Augmentation
- Simple Contrastive Graph Clustering Simple Contrastive Graph Clustering
- Improved Dual Correlation Reduction Network Improved Dual Correlation Reduction Network
8-class node classification Amazon Photography co-purchase graph node classification (8 categories). Uses the PyG sparse product features and ordinary co-purchase links. Evaluated by Accuracy.
Dataset details
Task
Amazon Photography co-purchase graph node classification (8 categories). The primary catalog metric is Accuracy.
Data
Amazon Photography co-purchase graph node classification (8 categories). The cataloged artifact reports 1 graph and 8 target classes.
Features
Nodes are products, edges are co-purchase links, and node features are sparse bag-of-words review features from the PyG Amazon-Photo release.
Splits and evaluation
The catalog records these protocols or variants: Standard split. Evaluation uses F1 (higher is better). Exact masks or folds must come from the cited release.
Quirks and pitfalls
This is an ordinary graph in the PyG release, not an incidence hypergraph; treating each co-purchase edge as a hyperedge changes the dataset.
Sources
SOTA Progression
new SOTA
matches SOTA at that time
other proposed result
other baseline result
Leaderboard
| # | Model | F1 ↑ | ACC | ARI | Conductance | NMI | Paper |
|---|---|---|---|---|---|---|---|
| 1 | GANN | 0.9405
±
0.0037 | — | — | — | — | |
| 2 | GPRGNN | 0.9308
±
0.0026 | — | — | — | — | |
| 3 | GAT | 0.9238
±
0.0014 | — | — | — | — | |
| 4 | JK-Net | 0.9168
±
0.0082 | — | — | — | — | |
| 5 | GCN-Cheby | 0.9156
±
0.0075 | — | — | — | — | |
| 6 | GWNN | 0.9075
±
0.0126 | — | — | — | — | |
| 7 | SGC | 0.9017
±
0.0075 | — | — | — | — | |
| 8 | FAGCN | 0.9006
±
0.0104 | — | — | — | — | |
| 9 | MixupForGraph | 0.8728
±
0.0108 | — | — | — | — | |
| 10 | PPNP | 0.8627
±
0.0074 | — | — | — | — | |
| 11 | APPNP | 0.8602
±
0.0076 | — | — | — | — | |
| 12 | IDCRN | 0.8208
±
0.0018 | — | — | — | — | |
| 13 | V. GCN | 0.8141
±
0.0147 | — | — | — | — | |
| 14 | GCN | 0.8115
±
0.0105 | — | — | — | — | |
| 15 | DCRN | 0.7966
±
0.0025 | — | — | — | — | |
| 16 | DFCN | 0.7600
±
0.0080 | — | — | — | — | |
| 17 | SCGDN | 0.7527
±
0.0018 | — | — | — | — | |
| 18 | GraphLearner | 0.7302
±
0.0234 | — | — | — | — | |
| 19 | CDC | 0.7300 | — | — | — | — | |
| 20 | SCAGC | 0.7277
±
0.0016 | — | — | — | — | |
| 21 | SCGC | 0.7222
±
0.0097 | — | — | — | — | |
| 22 | HSAN | 0.7203
±
0.0046 | — | — | — | — | |
| 23 | AGE | 0.7174
±
0.0093 | — | — | — | — | |
| 24 | DFCN | 0.7125
±
0.0031 | — | — | — | — | |
| 25 | AGC-DRR | 0.7103
±
0.0064 | — | — | — | — | |
| 26 | DCRN | 0.7086
±
0.0018 | — | — | — | — | |
| 27 | AFGRL | 0.6999
±
0.0034 | — | — | — | — | |
| 28 | DEAGC | 0.6987
±
0.0054 | — | — | — | — | |
| 29 | DAEGC | 0.6987
±
0.0054 | — | — | — | — | |
| 30 | MCGC | 0.6860 | — | — | — | — | |
| 31 | GAE | 0.6808
±
0.0176 | — | — | — | — | |
| 32 | MGAE | 0.6808
±
0.0176 | — | — | — | — | |
| 33 | SDCN | 0.6805
±
0.0181 | — | — | — | — | |
| 34 | DMoN | 0.6700
±
0.1000 Conductance
0.6000
±
0.4900 NMI
0.6100
±
0.0800 | — | — | 0.6000
±
0.4900 | 0.6100
±
0.0800 | |
| 35 | SDCN_Q | 0.6574
±
0.0134 | — | — | — | — | |
| 36 | SUBLIME | 0.6500
±
0.0700 Conductance
0.2300
±
0.1800 NMI
0.6000
±
0.0400 | — | — | 0.2300
±
0.1800 | 0.6000
±
0.0400 | |
| 37 | ARGA | 0.6483
±
0.0059 | — | — | — | — | |
| 38 | MvAGC | 0.6400 | — | — | — | — | |
| 39 | DAEGC | 0.6205
±
0.0048 | — | — | — | — | |
| 40 | GAE | 0.6121
±
0.0122 | — | — | — | — | |
| 41 | VGAER | 0.6100
±
0.1600 Conductance
0.7000
±
0.1800 NMI
0.5500
±
0.1800 | — | — | 0.7000
±
0.1800 | 0.5500
±
0.1800 | |
| 42 | IDEC | 0.6031
±
0.0062 | — | — | — | — | |
| 43 | MCGC | 0.5892
±
0.0005 | — | — | — | — | |
| 44 | VGAE | 0.5859
±
0.0006 | — | — | — | — | |
| 45 | DEC | 0.5816
±
0.0056 | — | — | — | — | |
| 46 | AutoSSL | 0.5447
±
0.0083 | — | — | — | — | |
| 47 | ARVGA | 0.5441
±
0.0042 | — | — | — | — | |
| 48 | GCA | 0.5359
±
0.0057 | — | — | — | — | |
| 49 | SSGC | 0.5279
±
0.0001 | — | — | — | — | |
| 50 | AE | 0.5143
±
0.0035 | — | — | — | — |
SOTA performance
set a new SOTA at the time
matched the SOTA at the time