Studying the influence of the sectoral structure of Russian regions on their economic development using artificial intelligence methods

Subject. The article explores the influence of the sectoral structure of economies of Russian regions on the level of their economic, innovation, and digital development in terms of achieving the national goal of "Technological Leadership" and ensuring the economic security of the country. Objectives. The study aims at clustering the Russian regions under artificial intelligence methods based on official statistics on the share of enlarged industries in the GRP of the region and the scale of region's economy; establishing links between identified cluster formations with average (for their constituent regions) indicators of economic development, innovation, and digital activity. Methods. We performed cluster analysis using the machine learning method being one of the most important parts of artificial intelligence. We compared the results of two main clustering procedures: hierarchical cluster analysis and the K-means method. Regions’ ratings for three groups of indicators were formed using the methods of normalization (Z-counting method) and aggregation of partial indicators. Results. The paper described the architecture of each cluster formation, calculated average values of considered indicators in clusters. Six out of seven identified clusters have a pronounced industry specialization. The seventh cluster was the most diversified, its sectoral structure is close to that of the Russian economy as a whole. For each group of development indicators (economic, innovative, digital), we identified leader and outsider clusters. Conclusions. The study confirmed the hypothesis about connection of the sectoral structure of Russian regions’ economies with the level of their economic development, innovation, and digital activity. The obtained clustering results can be used in the formation of digital twins of Russian regions to establish their general characteristics of development and forecast socio-economic indicators.