Hugging Face's logo

Datasets:
UMCU
/
DutchMedicalTermPairs

Tasks:
Token Classification
Modalities:
Text
Formats:
parquet
Languages:
Dutch
Size:
10M - 100M
Tags:
medical
Libraries:
Datasets
Dask
License:
Dataset card Data Studio Files
xet
Community
1
UMCU commited on
Commit
daa9907
·
verified ·
1 Parent(s): 41f1fd0

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -0
README.md CHANGED
@@ -19,6 +19,7 @@ and *soft* positive term pairs by
19
 
20
  We expand this with negative term pairs by random sampling, followed by selecting the top-N most dissimilar terms with (sim<0.25)
21
  * using [BioLORD 2023 M Dutch](https://huggingface.co/FremyCompany/BioLORD-2023-M-Dutch-InContext-v1)
 
22
  * using [RotatE graph embedding](https://arxiv.org/abs/1902.10197)
23
 
24
  We further expand these lists using LLMs.
 
19
 
20
  We expand this with negative term pairs by random sampling, followed by selecting the top-N most dissimilar terms with (sim<0.25)
21
  * using [BioLORD 2023 M Dutch](https://huggingface.co/FremyCompany/BioLORD-2023-M-Dutch-InContext-v1)
22
+ * using [sapMedRoBERTa.nl](https://huggingface.co/UMCU/sap_umls_medroberta.nl_meantoken)
23
  * using [RotatE graph embedding](https://arxiv.org/abs/1902.10197)
24
 
25
  We further expand these lists using LLMs.