B. M. Gavrikov, M. B. Gavrikov, N. V. Pestryakova, “On the ability of a statistical classifier to generalize”, Informatsionnye Tekhnologii i Vychslitel'nye Sistemy, 2021, Issue 4,Pages <nobr>38

This article is cited in 4 papers

CONTROL AND DECISION MAKING

On the ability of a statistical classifier to generalize

B. M. Gavrikov^a, M. B. Gavrikov^b, N. V. Pestryakova^a

^a Federal Research Center "Computer Science and Control" of Russian Academy of Sciences, Moscow, Russia
^b Federal Research Center "Keldysh Institute of Applied Mathematics" of Russian Academy of Sciences, Moscow, Russia

Abstract: The ability to generalize the statistical classifier designed to assess the state of human health by the parameters of peripheral blood is being studied. A mathematical model is described and imple-mented for the numerical study of interpolation and extrapolation properties of the classifier developed by the authors, based on the polynomial-regression approach and having probabilistic estimates.

Keywords: human health state, body system, peripheral blood, classification, polynomial regression, learning set.

DOI: 10.14357/20718632210404