Encoding Models
This section focuses on encoding the intersections between entities and relations. There are seven encoding models as shown below.

Linear / Bilinear

Factorisation

Neural Networks (NNs)

Convolutional NNs

Recurrent NNs

Transformers

Graph NNs
Linear / Bilinear
This type of encoding model uses linear operation to encode interactions between entities and relations. Examples of such encoding model are:

SE

SME

DistMult

ComplEx

ANALOGY

TransE with L2 regularisation, where scoring function can be expanded to only have linear transformation
Empirically, the ensembles of multiple linear models have shown to improve the predictive capability.
Factorisation
This involves formulating KRL models as threeway tensor X decomposition. The general principle can be denoted as $X_{hrt} ≈ h^TM_rt$ with the composition function following the semantic matching pattern.
Neural Networks (NNs)
Generally, the NNs will take in the entities and relations and compute a semantic matching score. An example of a simple MLP is shown in the figure above.
CNNs
This type of encoding model is used to capture deep expressive features. ConvE can model semantic information using the nonlinear feature learning from multiple conv layers. These nonlinear features can be concatenated to increase the learning ability of latent features. ConvKB has been shown to have better experimental performance than ConvE due to its transitional characteristics.
RNNs
RNNs can capture longterm dependency in knowledge graphs. There are research work that uses RNN to capture the vector representation of the long relation path from one entity to another, with or without entitylevel information.
Transformers
Using transformers can boost results through contextualised representation. We could use contextualised representation to encode edges and path sequences. KGBERT (uses the idea of language model and pretraining) can be used as an encoder for entities and relations.
GNNs
GNNS are used to learn the connectivity structure under an encoderdecoder framework. RGCN uses relationspecific transformation to model the directed nature of knowledge graphs. Other research uses graph attention networks to capture the multihop neighbourhood features.