Try out the prototype from kNOwHATE project designed to automatically detect online hate speech.
Developed by a team of researchers, the prototype utilizes a rigorously annotated dataset, which includes 24,739 comments from 88 YouTube videos and 29,846 tweets from 2,775 Twitter conversations. Trained annotators, such as social psychologists and linguists, carefully conducted these annotations.
To understand how this prototype functions check out a brief video tutorial below
While this prototype is still under development and may contain errors, it offers you the chance to test an automatic detection system and understand how it works. Delve into the possibilities of using AI to combat hate speech.
kNOwHATE encourages educators to explore this tool in their classrooms, prompting discussions regarding hate speech and the potential benefits and limitations of such technology.
Visit https://huggingface.co/knowhate to explore the dataset and test the prototype yourself.
Input a sentence like "volta para a tua terra" (go back to your country) into the provided box and click "Submit" to see if the prototype identifies it as hate speech. This initiative strives to cultivate understanding and awareness around the complexities of online hate speech detection.