Analyze financial news with FinBERT

This is an small application that showcases how to get financial news sentiment from Tiingo API and NLP pretrained models.

Project description

• Download news from Tiingo API within the desired dates range for the selected tickers.
• Load the FinBERT pretrained model from Hugging Face model hub and use it to score each piece of news.
• Classify those news in positive, negative or neutral based on a simple heuristic.
• This feature could be used later on a bigger model to try to predict stocks direction.
• This app could be modified to store the news and sentiment into a database with the proper format.

How to run it

• Clone the repository
• Install pipenv if needed: pip install pipenv --user.
• Install required libraries. Go to the cloned directory and run: pipenv install which will install dependencies based on my Pipfile.lock.
• You'll need your own Tiingo APIKEY. Create a .env file and write TIINGO_APIKEY="your-key-here".
• Define tickers and dates of interest an just run it.

FinBERT: Financial Sentiment Analysis with BERT

FinBERT sentiment analysis model is available on Hugging Face model hub. Check out their repo.

FinBERT is a pre-trained NLP model to analyze sentiment of financial text. It is built by further training the BERT language model in the finance domain, using a large financial corpus and thereby fine-tuning it for financial sentiment classification. For the details, please see FinBERT: Financial Sentiment Analysis with Pre-trained Language Models.