Fast Linear Model for Knowledge Graph Embeddings
Abstract
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.
- Publication:
-
arXiv e-prints
- Pub Date:
- October 2017
- DOI:
- 10.48550/arXiv.1710.10881
- arXiv:
- arXiv:1710.10881
- Bibcode:
- 2017arXiv171010881J
- Keywords:
-
- Statistics - Machine Learning;
- Computer Science - Machine Learning
- E-Print:
- Submitted AKBC 2017