The dissertation is devoted to the decision of an actual scientific and applied problem of development of new methods and means of formation of scenarios of information influences in systems of support of acceptance of administrative decisions on the basis of the analysis of semantic models and the analysis of the maintenance of global computer networks. The work is done with an introduction, four sections, bran, a list of sources used and two appendices. The work on the use of mathematical modeling, development of the theory of constructed computer computers, separate programs, as well as system, methods and means of modeling, including their structure and algorithmic organization, information technology and their use in proven research. The effective functioning of management decision support systems involves the rapid formation of reliable scenarios of information influences. An integral part of modern management decision support systems are tools for scenario modeling of information influences, which allows to investigate how significant the impact of a factor, to model the situation and predict the consequences of management influences with timely response to threats. But in the rapid development of the information society, characterized by the emergence of new forms of digital communication, increasing data and bandwidth of transmission channels, the emergence of new ways of information impact, traditional approaches to scenario modeling do not provide quality support for management decisions. The relevance of the dissertation is determined by the need to create modern information technologies for scenario modeling, which include the formation of arrays of input and output data, automatic modeling of the subject area based on scanning text resources and creating a network of concepts, integration of tools for research graphs (networks) and ontologies text information spaces. That is why the dissertation is devoted to the development and research of models, methods and means of forming scenarios of informational influence of some essences of information space on others, on the basis of automated construction of semantic networks. The concept of scenario refers to a sequence of informational influences on key concepts of the subject area in order to cause positive or negative consequences. Transitions in the scenario from one to another occur during the transition from one entity of the subject area to another, provided that there is a positive or negative relationship between these entities, pre-defined in the information space. On the basis of the analysis in the dissertation the necessity of reduction of time of formation of the scenario of information influence for the distributed information space is substantiated. To solve the problem, a set of methods is proposed, which are based on a combination of modern software and highly efficient algorithms. analysis of impact factors, calculation of the ratios of key indicators that determine information processes, setting characteristics (variables and constants) of information impact processes over time. A mathematical model for the task of finding the optimal information impact on the target objects of the subject area, which, in contrast to the existing ones, allows to investigate the magnitude of the impact for any individual node. In order to minimize the time of preparation of data for scenario modeling, a method of constructing a model of the subject area in the form of a semantic graph formed from computer network monitoring data, by identifying the most important concepts and relationships between them. The method of formation of optimal scenarios of information influences on target objects of the subject area on the basis of finding of set of routes of distribution of influence that allows to use at detection and planning of information influences is offered. Software and algorithmic tools for generating information impact scenarios for the subject area have been developed, including tools for integration with popular free data analysis programs, which allows for empowerment of analysts in management decision support systems. The offered methods and means with use of real data are investigated and their efficiency is shown. The graphs of ontologies formed during the formation of scenarios of information influences for different subject areas can be used to increase the efficiency of machine learning. For the first time, it is proposed to improve the method of forming a semantic model of the information space, which for the first time uses the extraction of the most important concepts, which reduces the modeling time. For the first time, a method of constructing optimal scenarios of impact on key concepts of the subject area is proposed, which for the first time is based on modified behavioral models of ant colonies, which reduces the computational complexity of the method and provides guaranteed convergence of scenario search algorithms.
