Detecting-Twitter-Bots-using-Machine-Learning
Twitter is one of the most popular social networking sites currently on the market. Released in 2006 it has been steadily growing since its creation and currently sits at the 10th largest website in the world . Its main purpose is microblogging which are called tweets. With a popularity, such as this, there is inevitability that it would become a target for spammers and scammers. A lot of these Twitter accounts are in the form of bots, automated programs that follow vast amounts of people and can spam normal Twitter accounts with links to nefarious websites. In this project, the objective is to create and teach a machine learning program to detect these bot accounts, to help expose them from normal Twitter accounts. There is a web based interface where a user can input a URL for a Twitter account to determine whether they are a real or not. The project is a web based system developed in the Django framework, also including front-end technologies such as JavaScript, Bootstrap, HTML, and CSS. It uses a MySQL database to store and query a user’s Twitter account and display the likelihood of that user being a bot. It uses Scikit Learn Machine learning libraries to create the models used for evaluating a user.
