Unmasking deception: detection of fake profiles in online social ecosystems

Online Social networks (OSNs) have become vital to people’s daily lives. They have permitted their users to share their information, interact with people, and influence other users. Consequently, a significant risk has emerged to OSN security and privacy, which is malicious fake profiles. Researchers have proposed various approaches to limit this problem. This study, a Systematic Literature Review (SLR), aims to provide insight into the current fake profile identification approaches and highlights the underlying key challenges for existing approaches. The review comprises three key categories of fake profiles that include sybils, sock puppets, and social bots. The results of this SLR include a refined classification of various approaches used in existing studies. In addition, a detailed comparative analysis has been provided to quantify the effectiveness of existing approaches. Further, it introduced several research challenges and future research directions in this domain. Taxonomy is created in the context of online social ecosystems to facilitate the clear organization of the diverse elements of fake profile types. Moreover, this study offers an effective fundamental framework based on socio-technical design to mitigate the dissemination of fake profiles over OSN. This study should assist researchers in comprehending the destructive potential of fake profiles and improve upon the existing defense mechanisms.