STOCHASTIC MODEL OF OBJECT CLASSIFICATION

The article proposes a stochastic model of classification procedures and ways of its possible application. The general formulation of the object classification problem and the presentation of the results of solving this problem in the form of a probabilistic matrix are considered. The use of such a stochastic classification model makes it possible to carry out a comparative analysis of the effectiveness of various classification procedures and to select the most suitable of them according to the criterion of proximity to the reference procedure. A set of basic metrics is presented that can be applied to solve the problem of assessing the degree of proximity of the probabilistic matrix of the analyzed classification procedure to the matrix of the reference procedure. On the example of a double classification procedure, the aggregation of stochastic models of individual procedures and obtaining a model of a complex classification procedure using the Hadamard product are considered. In conclusion, the possible directions of development of the stochastic model of classification procedures are presented.

Authors: E. A. Burkov, Е. A. Tolkacheva, P. I. Paderno, F. E. Sattorov

Direction: Informatics, Computer Technologies And Control

Keywords: Probability matrix, double classification procedure, object classification, authentication coefficient, identification coefficient, stochastic model

---

View full article