Numerical Claim Detection in Finance: A New Financial Dataset, Weak-Supervision Model, and Market Analysis
This repository contains the code for the paper "Numerical Claim Detection in Finance: A New Financial Dataset, Weak-Supervision Model, and Market Analysis".
The dataset and fine-tuned models are available on HuggingFace pages listed below. More information is available in the paper.
- Earnings call annotated data: HuggingFace Data
- RoBERTa-base model fine-tuned on earnings call data: HuggingFace Model
- RoBERTa-large model fine-tuned on analysts' reports data: HuggingFace Model
If you find this repository useful, please cite our work.
@article{shah2024numerical,
title={Numerical Claim Detection in Finance: A New Financial Dataset, Weak-Supervision Model, and Market Analysis},
author={Shah, Agam and Hiray, Arnav and Shah, Pratvi and Banerjee, Arkaprabha and Singh, Anushka and Eidnani, Dheeraj and Chava, Sahasra and Chaudhury, Bhaskar and Chava, Sudheer},
journal={arXiv preprint arXiv:2402.11728},
year={2024}
}Please raise issue on GitHub or contact Agam Shah (ashah482[at]gatech[dot]edu) for any issues and questions.
GitHub: @shahagam4