Accuracy Improvement For NLP Classification To Identify Messageson Social Media

Abstract

Terrorist groups like ISIS and Jihadist are spreading their propaganda online using various forms of social networks like Facebook, Twitter, and YouTube. Best way to stop these groups is to suspend accounts that spread propaganda to harm living being and hurt any religious place or monuments. This can be done by analysts manually read and analyze an enormous amount of information on social media. Our proposed methodology is to attempt an automatically detect radical content that is released by jihadist and Terrorist groups on Twitter. We are using a branch of Artificial Intelligence i.e. Machine learning to classify a tweet as radical or non-radical and generate results which indicate that an automated approach to aid analysts detecting radical content on social media.