You can deploy this model and start earning money today!

sofia-uni/
toxic-data-bg

text-classification

apache-2.0

1K<n<10K

not-for-all-audiences

medical

Views

No views yet

Warning: This dataset contains content that includes toxic, offensive, or otherwise inappropriate language.

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.

The dataset is an extention of the "Hate speech detection in Bulgarian" dataset and consist of sentences of:

BG-Jargon BG-Nationalisti forum BG-Mamma forum Proud.bg forum… See the full description on the dataset page: https://huggingface.co/datasets/sofia-uni/toxic-data-bg.