Software implementation of the main cluster analysis tools

This article discusses an approach to creating a complex of programs for the implementation cluster analysis methods. A number tools processing initial data set and their software are analyzed, as well complexity application analysis. An is generalized from point view factual material that supplies information problem under study basis discussion, decision-making. Cluster procedure combines objects or variables into groups based on given rule. The work provides grouping multivariate using proximity measures such sample correlation coefficient its module, cosine angle between vectors Euclidean distance. authors proposed method by centers, nearest neighbor selected standards. results can be used analysts in process structure will improve efficiency clustering algorithms. practical significance developed algorithms expressed package created means C ++ language VS environment.