Impersonation on Social Media: A Deep Neural Approach to Identify\n Ingenuine Content

Impersonators are playing an important role in the production and propagation\nof the content on Online Social Networks, notably on Instagram. These entities\nare nefarious fake accounts that intend to disguise a legitimate account by\nmaking similar profiles and then striking social media by fake content, which\nmakes it considerably harder to understand which posts are genuinely produced.\nIn this study, we focus on three important communities with legitimate verified\naccounts. Among them, we identify a collection of 2.2K impersonator profiles\nwith nearly 10k generated posts, 68K comments, and 90K likes. Then, based on\nprofile characteristics and user behaviours, we cluster them into two\ncollections of `bot' and `fan'. In order to separate the impersonator-generated\npost from genuine content, we propose a Deep Neural Network architecture that\nmeasures `profiles' and `posts' features to predict the content type:\n`bot-generated', 'fan-generated', or `genuine' content. Our study shed light\ninto this interesting phenomena and provides interesting observation on\nbot-generated content that can help us to understand the role of impersonators\nin the production of fake content on Instagram.\n