We computed both Word and Sub-word Embeddings using FastText. For Sub-word embeddings we selected Byte Pair Encoding (BPE) algorithm to represent the sub-words. We evaluated the Biomedical Word Embeddings obtaining better results than previous versions showing the implication that with more data, we obtain better representations.
Cite
@article{arxiv.2102.12843,
title = {Spanish Biomedical and Clinical Language Embeddings},
author = {Asier Gutiérrez-Fandiño and Jordi Armengol-Estapé and Casimiro Pio Carrino and Ona De Gibert and Aitor Gonzalez-Agirre and Marta Villegas},
journal= {arXiv preprint arXiv:2102.12843},
year = {2021}
}