Research of varying levels of greenhouse gas emissions in European countries using the k-means method

Greenhouse gas emissions are a global problem. Although the EU countries from 1990 to 2012 reduced their total emissions by 19.2% (CO₂ eq.), it is still necessary to limit their emissions. In the article the possibility of using the taxonomic methods that allow grouping (classifying) objects described by many attributes (variables) is presented. In particular, cluster analysis was used, in which some methods for the isolation of homogeneous subsets of surveyed objects can be distinguished. One of such method is k-means algorithm. As a measure of similarity of objects in clusters the Euclidean distance was applied. In the analysis 28 European countries were taken as objects of research and they were described by four attributes (variables), i.e. the emission levels of greenhouse gases such as carbon dioxide, methane, nitrogen oxides and nitrous oxide. The aim of the analysis is to grouping objects – the European countries – into clusters that are most similar to each other in the same cluster and most unlike in other clusters. The research was carried out according to total greenhouse gas emissions, and according to emissions of these gases per capita of the countries surveyed. The analyses are based on Eurostat reports.