Real time scalable sybil account admission control in online social networks

Abstract

The detection of Sybils (fake accounts or social bots) across Online newlineSocial Networks (OSNs) has turned into a major challenge due to the current newlineadvancement of different social networks like Twitter, LinkedIn and newlineFacebook. The exploding volumes of user information from OSNs are newlineproviding the need for extraordinary computational power and distinctive newlinecomputing facilities. Productive information and computation in parallel with newlineefficient partitioning is essential in delivering a rapid and scalable resolution. newlineThe user profiles can be copied and imitated by Sybil s with various purposes, newlinemost current Sybil distinguishing proof resolutions concentrate on mining of newlineusers exposed profile features, which are breakable. Studies have endeavoured newlineto match user accounts with the location of information and the timing of the newlineuser post alongside the written style of the user. On the other hand, the newlinelocation information is meagre for the major part of Social Media Networks newline(SMNs), and the writing style is hard to determine from short sentences. newlineAdditionally, online SMNs are relatively symmetric, existing user newlinerecognizable solutions centered on network structure is not successful. newlineCurrent studies have used k-NN (k-Nearest Neighbor) for identifying the newlinematching identity across OSNs. But, the execution of the k-NN search is less newlinevalid for large volumes. Also, the task is computationally challenging. newline newline