Detection of fake followers using feature ratio in self-organizing maps
Nitin T Simon, Susan Elias · 2017
Detection of Fake followers has been a challenging task for the social media research community. Fake followers adopt varied methods to accomplish the goal. Identifying and modeling the behavior of fake followers is an interesting ongoing active research field. In this paper we propose a measure based on a computed feature ratio value to effectively isolate fake follower accounts from genuine users. A twitter dataset comprising of fake follower information has been used for the analytics presented in this paper. The artificial neural network referred to as Self-Organizing Map has been used for training and analysis of the Twitter fake follower dataset. The analysis presented demonstrates that the proposed metric is an efficient measure to cluster fake followers.