UMCU's picture
Update README.md
ec8ecd1
|
raw
history blame
1.7 kB
metadata
language: nl
license: mit

MedRoBERTa.nl finetuned for negation

Description

This model is a finetuned RoBERTa-based model called RobBERT, this model is pre-trained on the Dutch section of OSCAR. All code used for the creation of RobBERT can be found here https://github.com/iPieter/RobBERT. The publication associated with the negation detection task can be found at https://arxiv.org/abs/2209.00470. The code for finetuning the model can be found at https://github.com/umcu/negation-detection.

Intended use

The model is finetuned for negation detection on Dutch clinical text. Since it is a domain-specific model trained on medical data, it is meant to be used on medical NLP tasks for Dutch. This particular model is trained on a 32-max token windows surrounding the concept-to-be negated.

Data

The pre-trained model was trained the Dutch section of OSCAR (about 39GB), and is described here: http://dx.doi.org/10.18653/v1/2020.findings-emnlp.292.

Authors

RobBERT: Pieter Delobelle, Thomas Winters, Bettina Berendt, Finetuning: Bram van Es, Sebastiaan Arends.

Usage

If you use the model in your work please refer either to https://doi.org/10.5281/zenodo.6980076 or https://doi.org/10.48550/arXiv.2209.00470

References

Paper: Pieter Delobelle, Thomas Winters, Bettina Berendt (2020), RobBERT: a Dutch RoBERTa-based Language Model, Findings of the Association for Computational Linguistics: EMNLP 2020

Paper: Bram van Es, Leon C. Reteig, Sander C. Tan, Marijn Schraagen, Myrthe M. Hemker, Sebastiaan R.S. Arends, Miguel A.R. Rios, Saskia Haitjema (2022): Negation detection in Dutch clinical texts: an evaluation of rule-based and machine learning methods, Arxiv