Hate Speech Nl

🚀 Hate Speech Classifier for Social Media Content in Dutch

A monolingual model for hate speech classification of Dutch social media content, leveraging the pre-trained BERTje model.

🚀 Quick Start

This is a monolingual model designed for hate speech classification of Dutch social media content. It was trained on 20,000 social media posts from platforms like YouTube, Twitter, and Facebook, and tested on an independent test set of 2,000 posts. The model is based on the pre-trained language model BERTje.

✨ Features

• Monolingual Focus: Specialized in classifying hate speech in Dutch social media content.
• Diverse Training Data: Trained on a large dataset from multiple social media platforms.
• Based on BERTje: Utilizes the pre-trained BERTje model for better performance.

📦 Installation

No installation steps were provided in the original document, so this section is skipped.

💻 Usage Examples

Basic Usage

During training, the text was preprocessed using the BERTje tokenizer. It is recommended to use the same tokenizer for inference.

Advanced Usage

No advanced usage examples were provided in the original document, so this part is skipped.

📚 Documentation

Tokenizer

During training, the text was preprocessed using the BERTje 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

🔧 Technical Details

The model is trained on 20,000 social media posts from YouTube, Twitter, and Facebook, and tested on an independent test set of 2,000 posts. It is based on the pre-trained language model BERTje.

📄 License

This project is licensed under the MIT License.