Dehatebert Mono Italian

🚀 Italian Monolingual Hate Speech Detection Model

This model is designed to detect hate speech in Italian. The "mono" in its name indicates a monolingual setup, where the model is trained solely on Italian language data. It is fine - tuned on a multilingual BERT model.

🚀 Quick Start

This model can be used to detect hate speech in Italian. You can find the training code at this url.

✨ Features

• Monolingual Focus: Trained specifically for Italian hate speech detection.
• Fine - Tuned on BERT: Leverages the power of a multilingual BERT model.
• Optimized Learning Rate: Achieved a validation score of 0.837288 with a learning rate of 3e - 5.

🔧 Technical Details

The model is trained with different learning rates. After extensive experimentation, the best validation score of 0.837288 was achieved when using a learning rate of 3e - 5.

📄 License

This project is licensed under the Apache 2.0 license.

📚 Documentation

For more details about our paper

Sai Saketh Aluru, Binny Mathew, Punyajoy Saha and Animesh Mukherjee. "Deep Learning Models for Multilingual Hate Speech Detection". Accepted at ECML - PKDD 2020.

Please cite our paper in any published work that uses any of these resources.

@article{aluru2020deep,
  title={Deep Learning Models for Multilingual Hate Speech Detection},
  author={Aluru, Sai Saket and Mathew, Binny and Saha, Punyajoy and Mukherjee, Animesh},
  journal={arXiv preprint arXiv:2004.06465},
  year={2020}
}