Shlukovací metody v data miningu

Abstract

One of the oppotunities in data mining is a use of clustering analysis. Clustering analysis belongs to unsupervised methods of data mining. We put here a focus on this method. Some basic principles are described in the second part of this paper. This method is examined on two examples from the marketing field. In the first example is used software Statgraphics 5.0Plus to solve clustering problem (nearest neighbour algorithm and Eucleidian distance); and in the second example is used Statistica 6.0Cz software (from Statoft; Inc.).

One of the oppotunities in data mining is a use of clustering analysis. Clustering analysis belongs to unsupervised methods of data mining. We put here a focus on this method. Some basic principles are described in the second part of this paper. This method is examined on two examples from the marketing field. In the first example is used software Statgraphics 5.0Plus to solve clustering problem (nearest neighbour algorithm and Eucleidian distance); and in the second example is used Statistica 6.0Cz software (from Statoft; Inc.).