Answers are grounded in this paper and cite their sources.
Introduction: Introduction: Currently, in various spheres of state and regional management, problems related to the need for constant monitoring of key parameters of the functioning of complex facilities are becoming more acute. The need for continuous updating of knowledge about the functioning of objects in the form of formalized models, in the conditions of continuously changing situational conditions of the processes under study, determines the need to use new data processing technologies when conducting information and analytical work. The purpose of the work is to develop a conceptual approach to improving the processes of information and analytical work in solving problems of recognizing the state of partially observed control objects using machine learning methods. Methods used: the algorithms implemented in machine learning methods allow us to identify various patterns in the available source data and reflect them in the generated model of the object of control in the form of parameters, and the use of a finite automaton determines the content of the processes of operational calculations implemented in various situational conditions set by the current completeness and reliability of the source data. The scientific novelty lies in the fact that the proposed method determines a qualitatively new content of information processing processes, and automation of the processes of developing and updating models of control objects creates conditions for a significant expansion of the feature space taken into account when conducting information and analytical work, and also provides the opportunity to obtain more reliable results in the interests of rational management of complex objects and processes. Result: the presented method makes it possible to evaluate the quality of the information solution when combining the results obtained using machine learning methods. The basis of the proposed approach is formed by the method of aggregating the results of non-precision recognition of the states of objects using several identification methods. Practical significance: the presented method is universal and can be applied in information support systems of public administration bodies in solving problems related to ensuring public safety.
No takes yet. Share an insight, caveat, or question.
PANKOV et al. (2022) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: