ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

INVESTIGATION OF A NONCLASSICAL APPROACH TO THE BEST CLASSIFICATION OF ECOLOGICAL HAZARD OF OBJECTS UNDER THE UNKNOWN FUNCTIONS OF LOSSES

Journal: Technogenic and ecological safety (Vol.3, No. 1)

Publication Date:

Authors : ;

Page : 70-74

Keywords : ecological hazard classification; observed data; loss function; decision rule;

Source : Downloadexternal Find it from : Google Scholarexternal

Abstract

Purpose. The no classical approach to the solution of the problem of the best classification of the ecological hazard of objects of the technical and natural sphere from the observable data is investigated with the impossibility of an exact definition of the loss functions for the decisive rules. Methodology. The methodology for solving the problem is based on the representation of unknown loss functions by functions of an arbitrary form, depending on the observed data and some composite parameter vector. Results. The best decisive rules for the classification of three characteristic states of ecological danger for the objects of the technical and natural spheres have been obtained. The resulting decision rules depend on observations and the composite vector of parameters, optimized in accordance with the minimum of the average classification risk. Practical value. This allows us to use the rules for the best classification of the environmental hazard of objects with a minimum of a priori information about the loss functions

Last modified: 2018-01-31 17:00:17