Detecting Twitter Bots

Time: Oct’19 - Dec’19

Course:24-787 Machine Learning and AI for Engineers

Source Code: Github Repository

Worked as a group of three members to implement multiple classification algorithms to distinguish Twitter bots from human accounts. These algorithms, namely logistic regression, KNN, MLP, SVM, Random Forest and XGBoost, are tested on continuous features including account age, number of followers, etc. To access individual user information from Twitter, we used the Python library Tweepy, which provides access to the twitter API.