setfit-9classes-multi_label

Computational Analysis of Communicative Acts for Understanding Crisis News Comment Discourses

The official trained models for "Computational Analysis of Communicative Acts for Understanding Crisis News Comment Discourses".

This model is based on SetFit (SetFit: Efficient Few-Shot Learning Without Prompts) and uses the sentence-transformers/paraphrase-mpnet-base-v2 pretrained model. It has been fine-tuned on our crisis narratives dataset.

---

Model Information

• Architecture: SetFit with sentence-transformers/paraphrase-mpnet-base-v2
• Task: Multi-label classification for communicative act actions
• Classes:
  • informing statement
  • challenge
  • accusation
  • rejection
  • appreciation
  • request
  • question
  • acceptance
  • apology

---

How to Use the Model

You can find the code to fine-tune this model and detailed instructions in the following GitHub repository:
Acts in Crisis Narratives - SetFit Fine-Tuning Notebook

Steps to Load and Use the Model:

1. Install the SetFit library:

   pip install setfit
   

2. Load the model and run inference:

   from setfit import SetFitModel
   
   # Download from the 🤗 Hub
   model = SetFitModel.from_pretrained("CrisisNarratives/setfit-9classes-multi_label")
   
   # Run inference
   preds = model("I'm sorry.")
   

For detailed instructions, refer to the GitHub repository linked above.

---

Citation

If you use this model in your work, please cite:

Paakki, H., Ghorbanpour, F. (2025). Computational Analysis of Communicative Acts for Understanding Crisis News Comment Discourses. In: Aiello, L.M., Chakraborty, T., Gaito, S. (eds) Social Networks Analysis and Mining. ASONAM 2024. Lecture Notes in Computer Science, vol 15212. Springer, Cham. https://doi.org/10.1007/978-3-031-78538-2_20

Questions or Feedback?

For questions or feedback, please reach out via our contact form.