The paper discusses the problem of determining the most influential people on social media while taking into consideration the possibilities of forming coalitions with their sets. The authors analyze social media networks created by criminal elements. Modern Russia has a practice of criminal and administrative prosecution of social media users who post information that violates Russian legislation. Today, there is an opportunity to study various aspects of society and obtain new research results using social media, which often play a key role. The authors suggest that the methodology of identifying social media users who are sources of extremist information should be examined and streamlined with the goal of improving the law enforcement practice of counteracting the spread of socially dangerous information on the Internet. They use mathematical methods to show that the problem of finding the maximally influential group consisting of k number of participants cannot have an optimal solution if this group includes k most influential participants individually. To solve this problem, the authors suggest using the game theory concept of Shapley vector, which makes it possible to evaluate the individual contribution of each participant into the formed group, i.e., to find out to what degree each of them could be useful as a team player. Law enforcement bodies already have certain experience of using software to upload data and analyze social networking sites. The presented method of calculating the significance of network users gives an opportunity to widen and improve this experience and to move from the ad hoc selective prosecution of separate users to the methodical work of suppressing and preventing crimes connected with the dissemination of information that constitutes an offence. The research area selected by the authors will, in its turn, contribute to improving the effectiveness of suppressing and preventing crimes connected with spreading information that contains constituting elements of different offences.
Read full abstract