Emtract Distilbert Base Uncased Emotion

🚀 EmTract (DistilBERT-Base-Uncased)

A specialized model for emotion detection in financial social media content

🚀 Quick Start

For steps on how to use the model for inference, please refer to the accompanying Inference.ipynb notebook.

✨ Features

• Specialized for emotion detection in financial social media content like that on StockTwits.
• Finetuned on a combination of large-scale emotion datasets and a smaller set of hand - tagged financial messages.

📦 Installation

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

💻 Usage Examples

No code examples are provided in the original document, so this section is skipped.

📚 Documentation

Model Description

emtract-distilbert-base-uncased-emotion is a specialized model finetuned on a combination of unify-emotion-datasets, containing around 250K texts labeled across seven emotion categories: neutral, happy, sad, anger, disgust, surprise, and fear. This model was later adapted to a smaller set of 10K hand - tagged messages from StockTwits. The model is designed to excel at emotion detection in financial social media content such as that found on StockTwits.

Model parameters were as follows: sequence length of 64, learning rate of 2e - 5, batch size of 128, trained for 8 epochs.

Training Data

The first part of the training data was obtained from the Unify Emotion Datasets available at here. The second part was obtained from social media and hand - tagged. It is available here.

Evaluation Metrics

The model was evaluated using the following metrics:

• Accuracy
• Precision
• Recall
• F1 - score

Research

The underlying research for emotion extraction from financial social media can be found on: arxiv and SSRN.

Citation

Please cite the following if you use this model:

Vamossy, Domonkos F., and Rolf Skog. "EmTract: Extracting Emotions from Social Media." Available at SSRN 3975884 (2023).

BibTex citation:

@article{vamossy2023emtract,
  title={EmTract: Extracting Emotions from Social Media},
  author={Vamossy, Domonkos F and Skog, Rolf},
  journal={Available at SSRN 3975884},
  year={2023}
}

Research using EmTract

Social Media Emotions and IPO Returns

Investor Emotions and Earnings Announcements

🔧 Technical Details

Model parameters were as follows: sequence length of 64, learning rate of 2e - 5, batch size of 128, trained for 8 epochs.

📄 License

This project is licensed under the terms of the MIT license.

Information Table