TESLA: TWITTER SPAMMER LEARNING
This is a group project for CSCE 670 Information Retrieval and Storage (Spring 2018). I gathered data through different sources, wrote crawler scripts to acquire latest data from Twitter API, performed feature engineering for the account model, and further fine tuned and picked the best model for account features. I also worked closely with the web designer as well as the deep learning master to ensure that the final product worked as expected.
TESLA (Twitter Spammer Learning) is able to tell if a user is considered spam or not based on account and text features. We were able to do online prediction with offline trained models. For more technical details, please refer to the about page as well as our code repository.