BaseModel class is the base class of all models in CogKGE. BaseModel class organizes code into three basic sections: (1) forward function for training, (2) embedding function for getting the embedding of entities and relations, (3) scoring function for computing the score of triples. Model module consists of four parts: translation distance models, semantic matching models, graph neural network-based models and transformer-based models. We summarize the models in the following table:
Category | Model | Conference | Paper |
---|---|---|---|
Translation Distance Models | TransE | NIPS 2013 | Translating embeddings for modeling multi-relational data |
TransH | AAAI 2014 | Knowledge Graph Embedding by Translating on Hyperplanes | |
TransR | AAAI 2015 | Learning Entity and Relation Embeddings for Knowledge Graph Completion | |
TransD | ACL 2015 | Knowledge Graph Embedding via Dynamic Mapping Matrix | |
TransA | AAAI 2015 | TransA: An Adaptive Approach for Knowledge Graph Embedding | |
BoxE | NIPS 2020 | BoxE: A Box Embedding Model for Knowledge Base Completion | |
PairRE | ACL 2021 | PairRE: Knowledge Graph Embeddings via Paired Relation Vectorss | |
Semantic Matching Models | RESCAL | ICML 2011 | A Three-Way Model for Collective Learning on Multi-Relational Data |
DistMult | ICLR 2015 | Embedding Entities and Relations for Learning and Inference in Knowledge Bases | |
SimpleIE | NIPS 2018 | SimplE Embedding for Link Prediction in Knowledge Graphs | |
TuckER | ACL 2019 | TuckER: Tensor Factorization for Knowledge Graph Completion | |
RotatE | ICLR 2019 | RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Space | |
Graph Neural Network-based Models | R-GCN | ESWC 2018 | Modeling Relational Data with Graph Convolutional Networks |
CompGCN | ICLR 2020 | Composition-based Multi-Relational Graph Convolutional Networks | |
Transformer-based Models | HittER | EMNLP 2021 | HittER: Hierarchical Transformers for Knowledge Graph Embeddings |
KEPLER | TACL 2021 | KEPLER: A Unified Model for Knowledge Embedding and Pre-trained Language Representation |