Methods of real-time detecting manipulation in online communities

Abstract

Online communities are popular platforms for social interaction. To a large extent they substituted traditional means of communication. Ensued from the different environment, new means of communication caused transformation in the communication process. Aims of communication retained the same as in the traditional environment. Among them is altering people's outlooks and provoking actions. Manipulation is harmful for individual recipients, adverse for national security and undesirable for managerial branch of online communities. To solve the pressing problem of preventing and combating manipulation criteria for detecting manipulation on early stages are presented. These criteria underlie methods and algorithms for detecting manipulation. To formalize manipulation tactics Moore machine as well as piecewise linear aggregates are used.