FIELD: physics.
SUBSTANCE: method includes access to a trained prediction model in the form of a decision tree, created, at least, in part, on the basis of a set of training objects; creation of a subgroup of random parameters of interest; linking the subgroup of random parameters of interest to the decision tree; determination of the sheet accuracy parameter based on the parameters of interest associated with the given sheet and the subgroup of random parameters of interest in this sheet; determination of the accuracy parameter of the trained prediction model in the form of a decision tree based on a certain sheet accuracy parameter.
EFFECT: higher prediction model accuracy.
41 cl, 11 dwg
