Try the spacy large model to see if performance improves

umcu / negation-detection

Negation detection in Dutch clinical text.

GNU General Public License v3.0

3 stars 0 forks source link

Open lcreteig opened 2 years ago

lcreteig commented 2 years ago

Currently using the small model. Might help a bit with tokenisation / sentence splitting.