Combinatorial Algorithm for Constructing a Parametric Feature Space for the Classification of Multidimensional Models

This article considers a combinatorial algorithm for finding the best set of generalized variables to construct a unified structure of models of objects being classified from given data sets on the basis of the principles of the group method of data handling (GMDH). A classifier giving the best representation of characteristics of the objects being classified is proposed to be constructed in the parameter space of the found structure of models.