Developing a model and algorithm for decision support in self-government bodies using machine learning

The article deals with the problem of developing a model and algorithm for assessing the state and supporting decision-making in self-government bodies using a neural network of machine learning. A multi-layer direct propagation neural network is constructed as a neural network. The neural network has three inputs, the hidden layer has two neurons and one output. The training sample consists of parametric assessments of educational activities, the social environment and the state of crime in local governments. The dependent parameter in the sample consists of generalized expert assessments of self-government bodies, in numerical form. A model and algorithm of the decision support process using a multilayer neural network of direct propagation perceptron are constructed. Based on the constructed model and the proposed algorithm, the weight coefficients of neurons are calculated and a linear function is constructed to support decision-making. Using this model, a generalized expert assessment is determined for the new self-government body in numerical form, which is interpreted as a proposed solution for improving the condition of the object.