paper-with-me

Link Property Prediction 벤치마크

Link Property Prediction on ogbl-collab

34개 결과 · ⬇ CSV · JSON

Test Hits@50

0.3886 0.5293 0.6701 0.8108 0.9515 2014-03 2026-09 DeepWalk — 0.5037 (2014-03-26) Node2vec — 0.4888 (2016-07-03) GCN (val as input) — 0.4714 (2016-09-09) GCN — 0.4475 (2016-09-09) GraphSAGE (val as input) — 0.5463 (2017-06-07) GraphSAGE — 0.481 (2017-06-07) Matrix Factorization — 0.3886 (2020-05-02) DeeperGCN — 0.5273 (2020-06-13) PLNLP+ LRGA — 0.6909 (2020-06-14) LRGA + GCN — 0.5221 (2020-06-14) SEAL-nofeat (val as input) — 0.6474 (2020-10-30) SEAL-nofeat — 0.5471 (2020-10-30) Adamic Adar+Edge Proposal Set — 0.6548 (2021-06-30) VQ-GNN (SAGE-Mean) — 0.4673 (2021-10-27) VQ-GNN (GCN) — 0.4316 (2021-10-27) VQ-GNN (GAT) — 0.4102 (2021-10-27) NGNN + GraphSAGE — 0.5359 (2021-11-23) NGNN + GCN — 0.5348 (2021-11-23) PLNLP (random walk aug.) — 0.7059 (2021-12-06) PLNLP (val as input) — 0.6872 (2021-12-06) GIDN@YITU — 0.7096 (2022-10-04) S3GRL (PoS Plus) — 0.6683 (2023-01-29) E2N — 0.9515 (2023-11-06) Refined-GAE — 0.6816 (2024-11-06) Jaccard Index — 0.505 (2025-01-12) DeepWalk — 0.5037 (2014-03-26) GraphSAGE (val as input) — 0.5463 (2017-06-07) PLNLP+ LRGA — 0.6909 (2020-06-14) PLNLP (random walk aug.) — 0.7059 (2021-12-06) GIDN@YITU — 0.7096 (2022-10-04) E2N — 0.9515 (2023-11-06)
RankModel Test Hits@50Ext. dataValidation Hits@50Number of params PaperCodeYear
1 E2N 0.9515 ± 0.1410No0.9546 ± 0.1270526851 Edge2Node: Reducing Edge Prediction to Node Classification 2023
2 HyperFusion 0.7129 ± 0.0018No0.7385 ± 0.00991064446212
3 GIDN@YITU 0.7096 ± 0.0055No0.9620 ± 0.004060449025 GIDN: A Lightweight Graph Inception Diffusion Network for High-efficient Link Prediction 2022
4 PLNLP + SIGN 0.7087 ± 0.0033No1.0000 ± 0.000034980864
5 PLNLP (random walk aug.) 0.7059 ± 0.0029No1.0000 ± 0.000034980864 Pairwise Learning for Neural Link Prediction zhitao-wang/PLNLP · zhitao-wang/plnlp 2021
6 HOP-REC 0.7012 ± 0.0016No1.0000 ± 0.000030191104
7 PLNLP+ LRGA 0.6909 ± 0.0055No1.0000 ± 0.000035200656 Global Attention Improves Graph Networks Generalization omri1348/LRGA · omri1348/LRGA · chuanqichen/cs224w 2020
8 PLNLP (val as input) 0.6872 ± 0.0052No1.0000 ± 0.000035112192 Pairwise Learning for Neural Link Prediction zhitao-wang/PLNLP · zhitao-wang/plnlp 2021
9 Refined-GAE 0.6816 ± 0.0041No1.0000 ± 0.0000126669825 Reconsidering the Performance of GAE in Link Prediction GraphPKU/Refined-GAE · graphpku/refined-gae 2024
10 TopoLink 0.6792 ± 0.0074No0.6771 ± 0.0083483363845
11 S3GRL (PoS Plus) 0.6683 ± 0.0030No0.9861 ± 0.00065913025 Simplifying Subgraph Representation Learning for Scalable Link Prediction venomouscyanide/s3grl · venomouscyanide/s3grl_ogb · venomouscyanide/S3GRL_OGB 2023
12 ELPH 0.6636 ± 0.5876No0.6631 ± 0.00213284065
13 BUDDY 0.6572 ± 0.0053No0.6621 ± 0.00161184867
14 Adamic Adar+Edge Proposal Set 0.6548 ± 0.0000No0.9735 ± 0.00000 Edge Proposal Sets for Link Prediction sangyx/gtrick · CUAI/Edge-Proposal-Sets 2021
15 SEAL-nofeat (val as input) 0.6474 ± 0.0043No0.6495 ± 0.0043501570 Labeling Trick: A Theory of Using Graph Neural Networks for Multi-Node Representation Learning dmlc/dgl · facebookresearch/SEAL_OGB 2020
16 Adamic Adar 0.6417 ± 0.0000No0.6349 ± 0.00000
17 Common Neighbor 0.6137 ± 0.0000No0.6036 ± 0.00000
18 SEAL-nofeat 0.5471 ± 0.0049No0.6495 ± 0.0043501570 Labeling Trick: A Theory of Using Graph Neural Networks for Multi-Node Representation Learning dmlc/dgl · facebookresearch/SEAL_OGB 2020
19 GraphSAGE (val as input) 0.5463 ± 0.0112No0.5688 ± 0.0077460289 Inductive Representation Learning on Large Graphs pyg-team/pytorch_geometric · dmlc/dgl · dmlc/dgl · +17 2017
20 NGNN + GraphSAGE 0.5359 ± 0.0056No0.6281 ± 0.0046591873 Network In Graph Neural Network 2021
1–20 / 34 다음 → 페이지당 10 20 50 100