PROBLEMS OF APPLICATION OF DECISION-MAKING MODELS IN THE REGIONAL ECONOMY UNDER CONDITIONS OF PARTIALLY RELIABLE INFORMATION

Abstract

In this study, attention is drawn to the under-explored area of strategic content analysis and the development of strategic vision for managers, with the supporting role of interpreting visualized big data to apply appropriate knowledge management strategies in regional companies. The study suggests improved models that can be used to process data and apply solutions to Big Data. The paper proposes a model of business processes in the region in the context of information clusters, which become the object of analysis in the conditions of active accumulation of big data about the external and internal environment. Research has shown that traditional econometric and data collection techniques cannot be directly applied to Big Data analysis due to computational volatility or computational complexity. The paper provides a brief description of the essence of the methods of associative and causal data analysis and the problems that complicate its application in Big Data. The scheme of accelerated search for a set of causal relationships is described. The use of semantically structured models, cause-effect models and the K-clustering method for decision making in big data is practical and ensures the adequacy of the results. The article explains the stages of applying these models in practice. In the course of the study, content analysis was carried out using the main methods of processing structured data on the example of the countries of the world using synthetic indicators showing the trends of Industry 4.0. When assessing Industry 4.0 technologies by region, the diversity of country grouping attributes should be considered. Therefore, during the analysis, the countries of the world were compared in two groups. The first group - the results for developed countries are presented in tabular form. For the second group, the results are presented in an explanatory form. In the process of assessing industrial 4.0 technologies, statistical indicators were used: "The share of medium and high-tech activities", "Competitiveness indicators", "Results in the field of knowledge and technology", "The share of medium and high-tech production in the total value added in the manufacturing industry", “Industrial Competitiveness Index (CIP score)”. As a result, the rating of the countries was determined based on the analysis of these indicators. . The reasons for the difficulties of calculations when processing Big Data are given in the concluding part of the article. Keywords: K - clustering method, causal links, data point, Euclidean distance