SOFTWARE IMPLEMENTATION OF THE NONPARAMETRIC ALGORITHM OF BIG STATISTICAL DATA AUTOMATIC CLASSIFICATION
Examined is the software that implements a nonparametric algorithm for automatic classification of big statistical data. The class is characterized by a unimodal fragment of probability density. The proposed algorithm of automatic classification is based on the decomposition of the initial data. The results of the decomposition form a set of multi-dimensional intervals centers and corresponding incidences of random variables values. Classes are discovered on the basis of the received information. The obtained nonpara-metric algorithms are important in the processing of remote sensing data.
Keywords: software, automatic classification, multidimensional histogram, pattern recognition, large-volume sampling, discretization of multidimensional random variables value area, remote sensing data.