• Utilities.py contains all the neccessary functions, it is the backbone of the repo.

• DF_creation.ipynb is a workflow example of creating a dataframe to be fed into the final classifier.

• In order to load the news datasets correctly have this folder structure :

   utilities.py
   
   DF_creation.ipynb
   
   Financial_News
   
    |--- analyst_rating_processed.csv
    
    |--- raw_analyst_ratings.csv
    
    |--- raw_partner_headlines.csv
  

• The dataset is downloaded from https://www.kaggle.com/miguelaenlle/massive-stock-news-analysis-db-for-nlpbacktests

Find short blog tutorials:

1.) https://medium.com/@poulinakis.kon/stocks-news-sentiment-analysis-with-deep-learning-transformers-and-machine-learning-cdcdb827fc06

2.) https://medium.com/mlearning-ai/predict-sp500-stock-price-with-python-machine-learning-sentiment-analysis-a296dc276353