Students Design Twitter Spam Detective

The rapid growth of the social media platform Twitter has triggered a dramatic increase in spam volume. Twitter data is valuable for a wide range of applications such as search engines and recommendation systems that demand high quality information to function properly. However, the existence of spammers on Twitter poses challenges to maintaining an acceptable level of data quality.

Aiming to eliminate this challenge, students Reem Jazi, Yara Ghawadreh and Dania Shilleh from the Faculty of Electrical and Computer Engineering designed an automated unsupervised method that reads the tweets of users and filters them out, sorting spam data from others in large-scale trending topics.

The students, supervised by electrical and computer engineering professor Mohammad Hussein and PhD student at the University of Toulouse, describe how their project is innovative. Most projects lean on learning approaches to produce classification models, while the young engineers’ notion is to use simple data (user name and tweet content) to perform the classification, which consumes less time and energy.

The students pointed out that their method proved a success with trial and error. They experimented it by filtering more than six million tweets posted in 50 trending topics.