ABSTRACT
Perversely, one can state that contemporary data analysis has developed too vigorously, which negatively caused, in particular, an absence of due care and attention regarding formalism, mathematical justification, and ultimately compact subject methodology. For practical applications, specific coordinates of the random variable can describe diverse quantities. A number of these, in particular representing distance or time, for their correct interpretation, must belong to properly bounded subsets, e.g., nonnegative numbers. The task of identifying atypical elements is one of the fundamental problems of contemporary data analysis, above all, in the preliminary phase of processing. Clustering has become the second basic problem within data analysis - compared to other procedures, it is more loosely defined and at a lesser advanced stage in research. Classification constitutes the third of the basic tasks of data analysis. It is often presented within literature on the subject that classic techniques for the smoothing parameter value calculation are mostly inappropriate for the purposes of classification.
