A level project built to detect twitter bots
Based on a towards data science post by Daniel Faucher: https://towardsdatascience.com/python-detecting-twitter-bots-with-graphs-and-machine-learning-41269205ab07
Model's a twitter users "user space" by scraping their tweets and replies and producing a graph based on how frequently other users interact with the "main" user and eachother within this "user space". This allows you to determine the impact a user (node) has on a user space.
An XGBoost linear regression model was trained on these user spaces to attempt to categorise the difference between a humans user space and a bots user space.
No longer works :( - Cant scrape data from twitter anymore.