Abstract
The popularity of online social networks like Facebook and Twitter has become the regular way of communication and interaction. Due to the popularity of such networks, the attackers try to reveal suspicious behaviour in the form of fake profile. To stop fake profile, various approaches are proposed in the recent years. The focus of recent work is to implement a machine learning technique to detect fake profile on Facebook platform by analysing public as well as private features. In this paper, a machine learning-based approach is proposed for detecting suspicious profiles for tapping and tainting multimedia big data on Facebook. Multimedia big data is a type of dataset in which the data is heterogeneous, human centric and has more media related contents with huge volumes like text, audio and video generated in different online social network. The experimental result of our work using content-based and profile-based features delivers first rate performance as compared to other approaches.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have