Neural network analysis of the main challenges and threats to the economic security of the Russian Federation

Abstract

Subject. The article addresses regional economy development from the standpoint of production, innovation, investment, and research activities in the regions of the Russian Federation, in the perspective of technological sovereignty and economic security of the country. Objectives. The aim is to solve a multifactorial problem describing the state of the economy in Russian regions by a new promising method, i.e. cluster analysis based on neural network modeling. Methods. To examine multidimensional statistical data, we employed the cluster analysis based on neural networks being an important branch of artificial intelligence. Results. We examined 85 regions and assessed them, using 12 indicators that we selected from the Rosstat website. The tools of artificial neural networks enabled the clustering of heterogeneous data. As a result, the Russian regions were located in six cluster formations. The paper assesses the influence of each indicator on cluster formation, demonstrates a significant inequality in the number of regions in clusters, highlights different levels of development of the regional economy, according to the totality of the considered indicators on cluster scale. Conclusions. The findings may help work out and implement strategies aimed at improving the balance of regional economy development, in the focus of technological sovereignty, under new threats and challenges to economic security.