Application of agglomerative clustering tools to identify degrading agricultural productions and determine key support measures

Abstract

Introduction. The relevance of the study is due to: the need for automated analysis of agricultural degradation (manual data processing for all regions and crops is ineffective, and agglomerative machine clustering allows regions with similar trends to be grouped, identifying the most problematic areas); risks to food security (reduced self-sufficiency with key products requires timely government support measures, clustering helps to rank regions by threat level and optimize resource allocation); lack of a unified methodology for assessing degradation; digitalization of agriculture. Purpose. The study aims to analyze the degradation of agricultural production in the regions of the Russian Federation, particularly to identify regions with the most pronounced decline in potato and other agricultural production, as well as to develop measures to improve the situation. Methods. Clustering methods were used for the analysis, specifically the agglomerative clustering algorithm, which allowed grouping regions based on the similarity of agricultural production degradation indicators using statistical data for 2005 - 2022. Results. The regions were divided into three clusters, with Cluster 2 (42 regions) being the most degraded in terms of agricultural production. These regions show significant declines in per capita production and yield indicators. Conclusions. Cluster analysis made it possible to identify regions with the most pronounced degradation of agricultural production, enabling the development of targeted support measures, such as targeted subsidies, tax incentives, and infrastructure restoration programs, to improve the situation in these regions.