Analysis and Design of Efficient generalized Forensic framework for Detecting Twitter Spammers
Asocial networking web site could be a platform to make social networks or social relations among those who share interests, activities, backgrounds or real-life connections. Users pay a good deal of your time on known social networks(e.g.Facebook,Twitter, SinaWeibo, etc.), reading news, discussing events and posting their message. Unfortunately, this quality conjointly attracts a big quantity of spammers whoincessantly expose malicious behaviours (e.g. post messages containing commercial topics or URLs, following a bigger quantity of users, etc.), resulting in nice inconvenience on traditional users’ social activities. The propose system will workAIER (Artificial intelligence for emergency response) approach for detecting twitter spammer .We first collected and labelled a large dataset with 34 K trending topics and 20 million tweets then, construct a labelled dataset of users and manually classify users into spammers and non-spammers; after that, abstract a set of novel features from message content and users’ social behavior. Our experiments show that true positive rate of spammers and non-spammers could reach 99.1% and 99.9%.