This paper shows that a simple baseline based on a Bag-of-Words (BoW) representation learns surprisingly good knowledge graph embeddings. By casting knowledge base completion and question answering as supervised classification problems, we observe that modeling co-occurences of entities and relations leads to state-of-the-art performance with a training time of a few minutes using the open sourced library fastText.
Cite
@article{arxiv.1710.10881,
title = {Fast Linear Model for Knowledge Graph Embeddings},
author = {Armand Joulin and Edouard Grave and Piotr Bojanowski and Maximilian Nickel and Tomas Mikolov},
journal= {arXiv preprint arXiv:1710.10881},
year = {2017}
}