Epidemiological clustering of Russian regions for the socio-economic forecast of Covid-19 rates

Abstract

The paper analyzes 3 clusters that differ in the growth rate of Covid-19 from the point of view of the socio-economic structure of the regions of the Russian Federation. In addition, the database also contains clinical indicators characterizing morbidity in the regions, indicators of nosocomial infection, investment parameters and the state of the transport system. Cluster analysis methods was carried out to identify the relationship between socio-economic characteristics of regions. The first cluster is more densely populated, and the regions assigned to the second cluster are removed from each other. Perhaps for this reason, the indicators of the transport system turned out to be less important than socio-economic ones for the spread of infection. The analysis was carried out using machine learning methods based on original methods of optimally reliable partitions and statistically weighted syndromes. The results of comparing the dynamics of Covid-19 spread in clusters 1 and 3, 2 and 3 strongly indicate the importance of studying traffic flows, especially in cities with high population density. The mathematical methods used are an effective tool for the purposes of not only epidemiological analysis, but also for a systematic analysis of the functioning of the socio-economic activity of the population of interacting regions, as well as the role of transport in this process.