An Approach to Exploratory Neural Network Analysis and Visualization of Economic Data in The Space Industry

The research relevance is provided by the need to develop and apply in practice new tools for analyzing and visualizing economic data in the space industry, since the infographic analysis methods and traditional multiparameter analysis require a lot of time operating with big data, without providing some hidden patterns. An approach to the exploratory neural network analysis and visualization of economic data using soft computing technologies (fuzzy interval logic, artificial neural networks) in management decision-making in the space industry has been proposed. The means of cognitive visualization and management decision support an exploratory (emergent) neural network map has been considered. The application of the self-organizing Kohonen map to searching for correlations and visual grouping of economic data has been analyzed. The authors' initial hypothesis that soft computing methods, in particular artificial neural networks, are suitable for the analyzing economic data in the space industry has been confirmed.