This repository contains a notebook that applies LSTM and BERT models to analyze financial tweets and predict the market state. The project explores the relationship between social media sentiment and financial market trends based on this https://huggingface.co/datasets/zeroshot/twitter-financial-news-sentiment dataset.
The code in this notebook was developed by me independently from scratch in Python, using PyTorch and Hugging Face libraries, as part of a group project. The goal was to compare BERT with LSTM for market prediction.
| File | Description |
|---|---|
financial-tweets-bert-lstm-analysis.ipynb |
Jupyter notebook containing data preprocessing, model implementation, training, and evaluation pipeline. |
This project is licensed under the MIT license.