Application of cluster analysis for oilseeds identification

Abstract

A search for mathematical methods to assess the level of balance of the main quantitative traits in oilseed varieties is described in the article. The main quantitative traits and crop yield are analyzed on the basis of experimental data of cluster analysis application for assessment varieties of oilseeds. The identification of varieties of oilseeds, grown in different regions of the country was carried out with the help of cluster analysis.   The main cultivation areas of oilseeds in 2017-19 have been also studied. As a result of the research, it was established that the cultivation area of flax, mustard and camelina crops has reduced, which is connected with the crop yield and low exports. The state of growth and the development of oilseeds, the duration of their cultivation, which depends on the biological characteristics of the breed, as well as the cultivation technology, have been analyzed. The specific features of the oilseeds have been studied. Unified measures which include a number of investigated features are identified with the help of the method. These features are unified using the metric into one resemblance cluster of the grouping of the studied objects. The breeds are grouped into three clusters according to years. The composition of each cluster changes depending on the natural and climatic conditions. Exploratory analysis using a point diagram is performed. A dendogram of the comparative characteristics of oilseed varieties by the elements of the yield structure is given. Using the K-means clustering method, the Euclidean distances of the studied parameters of the average values of the corresponding 5 clusters that have similar varieties of oilseeds are determined. The research shows that it is possible to determine the competitiveness of various oilseeds and to develop new food products.