Link Property Prediction On Ogbl Citation2
평가 지표
Ext. data
Number of params
Test MRR
Validation MRR
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | Ext. data | Number of params | Test MRR | Validation MRR | Paper Title | Repository |
|---|---|---|---|---|---|---|
| GCN + Heuristic Encoding | No | 372674 | 0.8891 ± 0.0005 | 0.8892 ± 0.0005 | Can GNNs Learn Link Heuristics? A Concise Review and Evaluation of Link Prediction Methods | - |
| Matrix Factorization | No | 281113505 | 0.5186 ± 0.0443 | 0.5181 ± 0.0436 | Open Graph Benchmark: Datasets for Machine Learning on Graphs | |
| Common Neighbor | No | 0 | 0.5147 ± 0.0000 | 0.5119 ± 0.0000 | - | - |
| GraphGPT(SMTP) | No | 46784128 | 0.9055 ± 0.0016 | 0.9042 ± 0.0014 | GraphGPT: Graph Learning with Generative Pre-trained Transformers | |
| NeighborSampling (SAGE aggr) | No | 460289 | 0.8044 ± 0.0010 | 0.8054 ± 0.0009 | Inductive Representation Learning on Large Graphs | |
| Full-batch GraphSAGE | No | 460289 | 0.8260 ± 0.0036 | 0.8263 ± 0.0033 | Inductive Representation Learning on Large Graphs | |
| AGDN w/GraphSAINT | No | 306716 | 0.8549 ± 0.0029 | 0.8556 ± 0.0033 | Adaptive Graph Diffusion Networks | |
| NGNN + SEAL | No | 1134402 | 0.8891 ± 0.0022 | 0.8879 ± 0.0022 | Network In Graph Neural Network | - |
| Node2vec | No | 374911105 | 0.6141 ± 0.0011 | 0.6124 ± 0.0011 | node2vec: Scalable Feature Learning for Networks | |
| CFG | No | 686253 | 0.8997 ± 0.0015 | 0.8987 ± 0.0011 | Circle Feature Graphormer: Can Circle Features Stimulate Graph Transformer? | |
| SUREL | No | 79617 | 0.8883 ± 0.0018 | 0.8891 ± 0.0021 | Algorithm and System Co-design for Efficient Subgraph-based Graph Representation Learning | |
| MPLP | No | 749757283 | 0.9072 ± 0.0012 | 0.9074 ± 0.0011 | Pure Message Passing Can Estimate Common Neighbor for Link Prediction | |
| S3GRL (PoS Plus) | No | 142275001 | 0.8814 ± 0.0008 | 0.8809 ± 0.0074 | Simplifying Subgraph Representation Learning for Scalable Link Prediction | |
| GraphGPT(d1n30) | No | 133096832 | 0.9305 ± 0.0020 | 0.9295 ± 0.0022 | GraphGPT: Graph Learning with Generative Pre-trained Transformers | |
| GraphSAINT (GCN aggr) | No | 296449 | 0.7985 ± 0.0040 | 0.7975 ± 0.0039 | GraphSAINT: Graph Sampling Based Inductive Learning Method | |
| HPE - Pre-trained Initialized | No | 749558528 | 0.8432 ± 0.0003 | 0.8422 ± 0.0002 | - | - |
| Full-batch GCN | No | 296449 | 0.8474 ± 0.0021 | 0.8479 ± 0.0023 | Semi-Supervised Classification with Graph Convolutional Networks | |
| ClusterGCN (GCN aggr) | No | 296449 | 0.8004 ± 0.0025 | 0.7994 ± 0.0025 | Cluster-GCN: An Efficient Algorithm for Training Deep and Large Graph Convolutional Networks | |
| SIEG | No | 256802 | 0.8957 ± 0.0010 | 0.8948 ± 0.0008 | - | - |
| Adamic Adar | No | 0 | 0.5189 ± 0.0000 | 0.5167 ± 0.0000 | - | - |
0 of 23 row(s) selected.