Graph embedding models produce embedding vectors for entities and relations in Knowledge Graphs, often without taking literal properties into account. We show an initial idea based on the combination of global graph structure with additional information provided by textual information in properties. Our initial experiment shows that this approach might be useful, but does not clearly outperform earlier approaches when evaluated on machine learning tasks.
@article{arxiv.1807.11761,
title = {A First Experiment on Including Text Literals in KGloVe},
author = {Michael Cochez and Martina Garofalo and Jérôme Lenßen and Maria Angela Pellegrino},
journal= {arXiv preprint arXiv:1807.11761},
year = {2018}
}
Comments
Presented at the 4th Workshop on Semantic Deep Learning (SemDeep-4)