Sentiment Roberta Large English 3 Classes

🚀 Sentiment Classification Model with RoBERTa

This RoBERTa-based model can classify the sentiment of English text into three classes: positive, neutral, and negative, offering high accuracy through fine - tuning on social media data.

🚀 Quick Start

This RoBERTa-based model is capable of classifying the sentiment of English language text into 3 classes:

• positive 😀
• neutral 😐
• negative 🙁

The model was fine-tuned on 5,304 manually annotated social media posts, achieving a hold-out accuracy of 86.1%. For details on the training approach, refer to Web Appendix F in Hartmann et al. (2021).

💻 Usage Examples

Basic Usage

from transformers import pipeline
classifier = pipeline("text-classification", model="j-hartmann/sentiment-roberta-large-english-3-classes", return_all_scores=True)
classifier("This is so nice!")

Advanced Usage

Output:
[[{'label': 'negative', 'score': 0.00016451838018838316},
  {'label': 'neutral', 'score': 0.000174045650055632},
  {'label': 'positive', 'score': 0.9996614456176758}]]

📚 Documentation

Please cite this paper when you use our model. Feel free to reach out to jochen.hartmann@tum.de with any questions or feedback you may have.

@article{hartmann2021,
  title={The Power of Brand Selfies},
  author={Hartmann, Jochen and Heitmann, Mark and Schamp, Christina and Netzer, Oded},
  journal={Journal of Marketing Research},
  year={2021}
}