IDENTIFICATION AND PREVENTION OF FAKE IDENTITIES IN SOCIAL MEDIA

Abstract

Now Days online social networks such as Face book, Twitter, Google+, LinkedIn, have become extremely popular all over the world and play a significant role in people’s daily lives. Due to open and anonymous nature of social sites the vulnerabilities on this social network are increasing, such as fake users also called as Sybil users. These kinds of malicious users can fabricate many dummy identities to target systems. So here we proposed a system. Which is a scalable defense system, which leverages user level activities such as friend acceptance, rejection? In this survey, our aim to give a comprehensive review of research related to user behavior in OSNs from several perspectives. First, we discuss social connectivity and interaction among users, Acceptance and rejections ratio are also important for that. Also, we investigate traffic activity from a network perspective. Friends invitation interactions among users as a social graph. Based on this social graph we proposed two key methods in order to detect fake identities that are fake users. Time Required to Browse and the time Gap Between the posting the status, sending SMS etc. Method is a voting-based Sybil detection and second is Sybil community detection to find other colluding Sybil around identified Sybil. In the second method we are using the global acceptance rate in order to detect the fake users. We are going to show the results in the form of global acceptance ratio of a particular user.