Hate Speech It

🚀 Hate Speech Classifier for Social Media Content in Italian Language

A monolingual model for hate speech classification of Italian social media content, trained on YouTube comments and based on the Italian ALBERTO pre - trained model.

🚀 Quick Start

This is a monolingual model designed for hate - speech classification of social media content in the Italian language. It was trained on 119,670 YouTube comments and tested on an independent test set of 21,072 YouTube comments. The model is built upon the Italian ALBERTO pre - trained language model.

📚 Documentation

Please cite

Kralj Novak, P., Scantamburlo, T., Pelicon, A., Cinelli, M., Mozetič, I., & Zollo, F. (2022, July). Handling disagreement in hate speech modelling. In International Conference on Information Processing and Management of Uncertainty in Knowledge - Based Systems (pp. 681 - 695). Cham: Springer International Publishing. https://link.springer.com/chapter/10.1007/978 - 3 - 031 - 08974 - 9_54

Tokenizer

During training, the text was preprocessed using the original Italian ALBERTO tokenizer. We suggest using the same tokenizer for inference.

Model output

The model classifies each input into one of four distinct classes:

• 0 - acceptable
• 1 - inappropriate
• 2 - offensive
• 3 - violent

📄 License

This project is under the MIT license.