Link Prediction

460 papers with code • 69 benchmarks • 45 datasets

Link prediction is a task to estimate the probability of links between nodes in a graph.

( Image credit: Inductive Representation Learning on Large Graphs )

Benchmarks

Add a Result

Dataset	Best Model	Paper Title	Compare
FB15k-237	GAATs	Knowledge Graph Embedding via Graph Attenuated Attention Networks	See all
WN18RR	StAR	Structure-Augmented Text Representation Learning for Efficient Knowledge Graph Completion	See all
WN18	Inverse Model	Convolutional 2D Knowledge Graph Embeddings	See all
FB15k	AutoKGE	AutoSF: Searching Scoring Functions for Knowledge Graph Embedding	See all
YAGO3-10	ComplEx-N3 (reciprocal)	Canonical Tensor Decomposition for Knowledge Base Completion	See all
Cora	Walkpooling	Neural Link Prediction with Walk Pooling	See all
Citeseer	Graph InfoClust (GIC)	Graph InfoClust: Leveraging cluster-level node information for unsupervised graph representation learning	See all
Yelp	PEAGAT	Metapath- and Entity-aware Graph Neural Network for Recommendation	See all
Pubmed	Walkpooling	Neural Link Prediction with Walk Pooling	See all
FB15k	MEI (small)	Multi-Partition Embedding Interaction with Block Term Format for Knowledge Graph Completion	See all
MovieLens 25M	PEAGAT	Metapath- and Entity-aware Graph Neural Network for Recommendation	See all
CoDEx Small	ComplEx-N3-RP	Relation Prediction as an Auxiliary Training Objective for Improving Multi-Relational Graph Representations	See all
CoDEx Medium	ComplEx-N3-RP	Relation Prediction as an Auxiliary Training Objective for Improving Multi-Relational Graph Representations	See all
CoDEx Large	ComplEx-N3-RP	Relation Prediction as an Auxiliary Training Objective for Improving Multi-Relational Graph Representations	See all
WordNet	Hyperbolic Entailment Cones	Hyperbolic Entailment Cones for Learning Hierarchical Embeddings	See all
FB122	Prob-CBR	Probabilistic Case-based Reasoning for Open-World Knowledge Graph Completion	See all
NELL-995	Prob-CBR	Probabilistic Case-based Reasoning for Open-World Knowledge Graph Completion	See all
DBLP	GLACE	Gaussian Embedding of Large-scale Attributed Graphs	See all
ICEWS05-15	TNTComplEx (x10)	Tensor Decompositions for temporal knowledge base completion	See all
ICEWS14	TNTComplEx (x10)	Tensor Decompositions for temporal knowledge base completion	See all
KG20C	MEI (small)	Multi-Partition Embedding Interaction with Block Term Format for Knowledge Graph Completion	See all
YAGO37	SEEK	SEEK: Segmented Embedding of Knowledge Graphs	See all
YouTube	GATNE-T	Representation Learning for Attributed Multiplex Heterogeneous Network	See all
LiveJournal	PBG (1 partition)	PyTorch-BigGraph: A Large-scale Graph Embedding System	See all
Douban	HSRL (DW)	Learning Topological Representation for Networks via Hierarchical Sampling	See all
YAGO15k	TNTComplEx (x10)	Tensor Decompositions for temporal knowledge base completion	See all
Cora (biased evaluation)	GraphStar (double weight on positive examples)	Graph Star Net for Generalized Multi-Task Learning	See all
Citeseer (biased evaluation)	GraphStar (double weight on positive examples)	Graph Star Net for Generalized Multi-Task Learning	See all
Pubmed (biased evaluation)	GraphStar (double weight on positive examples)	Graph Star Net for Generalized Multi-Task Learning	See all
JF17K	StarE (H) + Transformer (H)	Message Passing for Hyper-Relational Knowledge Graphs	See all
Decagon	Decagon	Modeling polypharmacy side effects with graph convolutional networks	See all
TSP/HCP Benchmark set	EGT	Edge-augmented Graph Transformers: Global Self-attention is Enough for Graphs	See all
USAir	SEAL	Link Prediction Based on Graph Neural Networks	See all
Wiki	BANE	Binarized Attributed Network Embedding	See all
PPI	HGCN	Hyperbolic Graph Convolutional Neural Networks	See all
YAGO39K	TransC (bern)	Differentiating Concepts and Instances for Knowledge Graph Embedding	See all
AKSW-bib	KG2Vec LSTM	Expeditious Generation of Knowledge Graph Embeddings	See all
Amazon	GATNE-T	Representation Learning for Attributed Multiplex Heterogeneous Network	See all
Twitter	GATNE-T	Representation Learning for Attributed Multiplex Heterogeneous Network	See all
Alibaba-S	GATNE-T	Representation Learning for Attributed Multiplex Heterogeneous Network	See all
Alibaba	GATNE-I	Representation Learning for Attributed Multiplex Heterogeneous Network	See all
ACM	GLACE	Gaussian Embedding of Large-scale Attributed Graphs	See all
Last.FM	MAGNN	MAGNN: Metapath Aggregated Graph Neural Network for Heterogeneous Graph Embedding	See all
Wiki-Vote	Asymmetric Transitivity Preservation	ATP: Directed Graph Embedding with Asymmetric Transitivity Preservation	See all
Cit-HepPH	Asymmetric Transitivity Preservation	ATP: Directed Graph Embedding with Asymmetric Transitivity Preservation	See all
Gnutella	Asymmetric Transitivity Preservation	ATP: Directed Graph Embedding with Asymmetric Transitivity Preservation	See all
WN18 (filtered)	ParTransH	Efficient Parallel Translating Embedding For Knowledge Graphs	See all
FB15k (filtered)	ParTransH	Efficient Parallel Translating Embedding For Knowledge Graphs	See all
IMDb	Event2vec	Representation Learning for Heterogeneous Information Networks via Embedding Events	See all
MIT	HSRL (DW)	Learning Topological Representation for Networks via Hierarchical Sampling	See all
Cora (nonstandard variant)	GLACE	Gaussian Embedding of Large-scale Attributed Graphs	See all
Citeseer (nonstandard variant)	GLACE	Gaussian Embedding of Large-scale Attributed Graphs	See all
Pubmed (nonstandard variant)	GLACE	Gaussian Embedding of Large-scale Attributed Graphs	See all
WD50K	StarE (H) + Transformer (H)	Message Passing for Hyper-Relational Knowledge Graphs	See all
Wikidata12k	TimePlex	Temporal Knowledge Base Completion: New Algorithms and Evaluation Protocols	See all
Yago11k	TimePlex	Temporal Knowledge Base Completion: New Algorithms and Evaluation Protocols	See all
Drug-target interactions	HOGCN	Predicting Biomedical Interactions with Higher-Order Graph Convolutional Networks	See all
Drug-Drug Interactions	HOGCN	Predicting Biomedical Interactions with Higher-Order Graph Convolutional Networks	See all
protein-protein interactions	HOGCN	Predicting Biomedical Interactions with Higher-Order Graph Convolutional Networks	See all
Gene-disease interactions	HOGCN	Predicting Biomedical Interactions with Higher-Order Graph Convolutional Networks	See all
FB-AUTO	BoxE	BoxE: A Box Embedding Model for Knowledge Base Completion	See all
AbstRCT - Neoplasm	ResAttArg	Multi-Task Attentive Residual Networks for Argument Mining	See all
CDCP	ResAttArg	Multi-Task Attentive Residual Networks for Argument Mining	See all
DRI Corpus	ResAttArg	Multi-Task Attentive Residual Networks for Argument Mining	See all
COLLAB	GatedGCN-PE	Benchmarking Graph Neural Networks	See all
Aristo-v4	ComplEx-N3-RP	Relation Prediction as an Auxiliary Training Objective for Improving Multi-Relational Graph Representations	See all
DDB14	ConE	Modeling Heterogeneous Hierarchies with Relation-specific Hyperbolic Cones	See all
GO21	ConE	Modeling Heterogeneous Hierarchies with Relation-specific Hyperbolic Cones	See all
OpenBioLink	DistMult		See all

Show all 69 benchmarks

Collapse benchmarks

Datasets

Subtasks

Greatest papers with code

Without code Latest Greatest

Neural Collaborative Filtering vs. Matrix Factorization Revisited

google-research/google-research • • 19 May 2020

This approach is often referred to as neural collaborative filtering (NCF).

Ranked #6 on Link Prediction on Yelp

Collaborative Filtering Link Prediction

20,569

Paper
Code

Cluster-GCN: An Efficient Algorithm for Training Deep and Large Graph Convolutional Networks

google-research/google-research • • KDD 2019

Furthermore, Cluster-GCN allows us to train much deeper GCN without much time and memory overhead, which leads to improved prediction accuracy---using a 5-layer Cluster-GCN, we achieve state-of-the-art test F1 score 99. 36 on the PPI dataset, while the previous best result was 98. 71 by [16].

Ranked #1 on Node Classification on Amazon2M

Graph Clustering Link Prediction +1

20,569

Paper
Code

Is a Single Embedding Enough? Learning Node Representations that Capture Multiple Social Contexts

google-research/google-research • • 6 May 2019

Recent interest in graph embedding methods has focused on learning a single representation for each node in the graph.

Graph Embedding Link Prediction

20,569

Paper
Code

How Attentive are Graph Attention Networks?

rusty1s/pytorch_geometric • • 30 May 2021

We formally define this restricted kind of attention as static attention and distinguish it from a strictly more expressive dynamic attention.

Graph Attention Graph Property Prediction +3

13,175

Paper
Code

Recurrent Event Network: Autoregressive Structure Inference over Temporal Knowledge Graphs

dmlc/dgl • • 11 Apr 2019

The task becomes more challenging on temporal knowledge graphs, where each fact is associated with a timestamp.

Knowledge Graphs Link Prediction

8,512

Paper
Code

CoKE: Contextualized Knowledge Graph Embedding

PaddlePaddle/models • • 6 Nov 2019

This work presents Contextualized Knowledge Graph Embedding (CoKE), a novel paradigm that takes into account such contextual nature, and learns dynamic, flexible, and fully contextualized entity and relation embeddings.

Knowledge Graph Embedding Link Prediction

6,267

Paper
Code

Neural Factorization Machines for Sparse Predictive Analytics

shenweichen/DeepCTR • • 16 Aug 2017

However, FM models feature interactions in a linear way, which can be insufficient for capturing the non-linear and complex inherent structure of real-world data.

Ranked #2 on Link Prediction on MovieLens 25M

Link Prediction

5,611

Paper
Code

Graph Attention Networks

labmlai/annotated_deep_learning_paper_implementations • • ICLR 2018

We present graph attention networks (GATs), novel neural network architectures that operate on graph-structured data, leveraging masked self-attentional layers to address the shortcomings of prior methods based on graph convolutions or their approximations.

Ranked #2 on Skeleton Based Action Recognition on J-HMBD Early Action