Recognizing Fake identities in Online Social Networks based on a Finite Automaton approach

Abstract

Online Social Networks (OSNs) are a great venue for scammers to impersonate the identities of users via creating fake profiles. Fake profiles are a popular tool for the intruders which can be used to carry out malicious activities such as impersonation attacks and harming persons' reputation and privacy in (OSN). Hence, recognizing the identities of fake profiles is one of the critical security problems in OSNs. In this paper, we proposed a detection mechanism called Fake Profiles Recognizer (FPR) for recognizing and detecting Fake Profiles in OSNs. The detection methodology in FPR is based on the functionality of Regular Expression and Deterministic Finite Automaton (DFA) approaches for recognizing the identity of profiles. We evaluated our detection system on three popular types of Online Social Networks: Facebook, Google+, and Twitter. The results explored high accuracy, efficiency, and low False Positive Rate of FPR mechanism in detecting the identities of Fake Profiles. In addition, our proposed detection mechanism achieved strong competitive results compared with other detection mechanisms in the literature.