Datasets:

Modalities:
Tabular
Text
Formats:
csv
Languages:
Bulgarian
Tags:
Not-For-All-Audiences
medical
Libraries:
Datasets
pandas
License:
toxic-data-bg / Readme.md
melaniab's picture
Create Readme.md
c537908 verified

The toxic-data-bg dataset consists of 4,384 manually annotated sentences across four categories: toxic language, medical terminology, non-toxic language, and terms related to minority communities.