search

shaye059 / Machine-Learning-Stock-Prediction

Using NLP Pipeline and Machine learning to anticipate the effect of news stories on stock prices.
MIT License
0 stars 1 forks source link
machine-learning newsapi nlp nltk stock-market stock-price-prediction

readme

Machine-Learning-Stock-Prediction

The goal of this project is to use Natural Language Processing and Machine Learning to predict the effects of news stories on company stock prices.

The most up to date dataset is available from here: Processed_Articles.csv

To run the scraping you'll need the following libraries:

pip install pandas newsapi-python yfinance

You'll also need to sign up for an API key and set it as an environment variable named NEWS_API_KEY.

To view the current notebook you can open it in Colab here . This started out as just a small little Notebook experiment but seeing as the developer edition of the news API is limited to pulling the last 30 days I realized that to build a usable dataset I'd have to set up routine scrapers. The notebook itself is currently being broken down into modularized scripts and therefore is probably not runnable if you want to use the already scraped and processed data provided in the CSV.