Computational supporting decision-making in self-government bodies using machine learning

Abstract

The article discusses the problem of building a model and algorithm for supporting decision-making in self-government bodies by machine learning. As a method of machine learning, a multiple linear regression method is selected for processing the training sample. In the training sample, independent data consists of parametric assessments in the numerical form of self-government organs in three areas of activity, as, education, social environment and crime. A dependent parameter consists of generalized expert assessments of self-government bodies, also in numerical form. A model and an algorithm for the decision support process for making decisions using the multiple linear regression method is built. Based on the built model and the proposed algorithm, the ratios of the function coefficients to support the decision-making are identified. With this model, a generalized expert assessment in a numerical form is determined for the new self-government body, which is interpreted as a proposed solution to improve the state of the object.