UCL Data Science Challenge

For the purpose of this challenge we have been exploring the correlation between news articles and prices.

Machine Learning Models

We have used Decision Forest Regression and Linear Regression to create a model comparison between sentiment and count based models.

Notebooks

• Comparing news sentiment and stock price (all - microsoft).ipynb - Correlates the sentiment of news with stock prices for Microsoft
• Events and stock price correlation.ipynb - Correlates Microsofts' events and their closest stock prices changes
• Microsoft Enquity.ipynb - Allows to look into the formatting of the enquities data set

Data Sets Used

• the main data set we had used was provided by Bloomberg and it contained equities, indexes, etc.
• https://github.com/philipperemy/financial-news-dataset