Publication date: 2021

DOI: 10.3103/S000510552104004X

An intellectual system has been created that implements the JSM method of automated research support for the task of predicting the risk of pelvic organ prolapse in women with a history of birth through the birth canal. Methods for expanding fact bases are applied. For the first time, a lattice of strategies formed by predicates of simple similarity, difference, and similarity-difference (a, ad(0), ad(2)) has been implemented and an algorithm for optimizing the replenishment of the fact base that uses the value of explainability has been proposed. Three experiments that were carried out according to the same scheme are described; in the second experiment, noise was removed when analyzing a subset of the initial data. New tasks have been formulated that allow further improvement of research results.

Издатель: PLEIADES PUBLISHING INCPLEIADES HOUSE, 7 W 54 ST, NEW YORK, NY, UNITED STATES

Тип: Article