Fast Linear Model for Knowledge Graph Embeddings
This provides a fast and effective method for knowledge graph tasks, though it is incremental as it builds on existing supervised classification approaches.
The paper tackles knowledge graph embedding by using a Bag-of-Words representation as a simple baseline, achieving state-of-the-art performance in knowledge base completion and question answering with training times of a few minutes.
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.