The dissertation to the solution of the actual scientific and technical problem of development of methods and means of software modeling and realization of various strategies of distribution of information influences in segments of social networks is devoted.
The relevance of the topic is determined by the increasing role of social networks in the processes of information and psychological influences, which may have economic, political and other motives. In particular, one of the aspects that determine the relevance of the study is countering the propaganda of enemy states in social networks.
The analysis of modern methods and models of information influences dissemination in social networks in the conditions of information confrontation is carried out, which showed that the vast majority of models and methods do not take into account individual node characteristics, namely behavior of separate node and strategy of dissemination information influences.
The mathematical model of information influence distribution in the social network segment has been developed, which allows the application of different behavioral strategies by the subjects of influence. In addition, on the basis of the proposed model developed algorithms for software simulation of the process of dissemination of information influences in the segments of the social network, as well as algorithms for generating the structure of segments of the social network. Basic behavioral strategies of subjects of influence in a social network during information confrontations are developed, which allow to effectively model different approaches to the choice of target nodes by subjects of influence and the process of dissemination of information-psychological influences. Algorithms for software simulation of various behavioral strategies of the subjects of influence in social networks have been developed and their effectiveness has been compared. Examples of basic behavioral strategies of subjects of influence in a social network during information confrontations are offered.
As a result of considering a number of proposed methods for generating the network structure, their disadvantages were revealed: a low level of clustering, the inability to generate a network with a given structure.
The method of generating the structure of the social network segment has been improved, which by choosing the structure of the social network from a set of parameterized subsets of clusters and choosing their structural features, allows you to choose the number and types of clusters in the network. The method of software simulation modeling of information influence distribution in the social network has been improved, which due to the application of the proposed mathematical model of information influence distribution in the social network segment makes it possible to model different behavioral strategies of information influence subjects. The method of choosing the behavioral strategies of the subject of influence in the social network based on analytic hierarchy process during information conflicts, which allows for less time to spread informational influence among the nodes of the social network segment.
The method of optimal selection of the target node for the attack during the spread of influences by the network entity based on the method of analysis of hierarchies and the application of behavioral strategies based on available data about the network during information conflicts has been further developed. Applying the behavioral strategies of the subjects of information influences based on the analysis of network data can significantly increase the speed of information dissemination by an average of 70% compared to the random selection of nodes to attack. Selecting the optimal target node allows for less time to spread informational influences among the nodes of the network segment.
The software simulation model of the social network and the processes of dissemination of information and psychological influences in it was developed using an object-oriented approach. All actions of network nodes are implemented as their methods, there is a separate class that monitors the iterative process, counts iterations and initiates the action of nodes. To simplify the perception of simulation results, as well as their analysis, the software simulation model contains a graphical display of the social network graph and shows changes in it over time, in particular, the state of network nodes in relation to information and psychological influences, as well as a dynamic graph involved in the influence of nodes on each iteration of the model.
Keywords: information security, information confrontation, social networks, information influences, dissemination strategies, simulation modeling
