A hybrid approach for spam detection for Twitter

Online social networks (OSNs) are becoming extremely popular among Internet users as they spend significant amount of time on popular social networking sites like Facebook, Twitter and Google+. These sites are turning out to be fundamentally pervasive and are developing a communication channel for billions of users. Online community use them to find new friends, update their existing friends list with their latest thoughts and activities. Huge information available on these sites attracts the interest of cyber criminals who misuse these sites to exploit vulnerabilities for their illicit benefits such as advertising some product or to attract victims to click on malicious links or infecting users system just for the purpose of making money. Spam detection is one of the major problems these days in social networking sites such as twitter. Most previous techniques use different set of features to classify spam and non-spam users. In this paper, we proposed a hybrid technique which uses content-based as well as graph-based features for identification of spammers on twitter platform. We have analysed the proposed technique on real Twitter dataset with 11k uses and more than 400k tweets approximately. Our results show that the detection rate of our proposed technique is much higher than any of the existing techniques.