UDC 004.8
COMBINING MACHINE LEARNING METHODS AND SIMULATION MODELING OF SOCIO-ECONOMIC PROCESSES IN DECISION SUPPORT SYSTEMS
O. D. Kazakov, Ph.D. (Econ.), assistant professor, Head of the Department of Information Technology, BSETU, Bryansk, Russia;
orcid.org/ 0000-0001-9665-8138, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
N. Yu. Azarenko, Ph.D. (Econ.), assistant professor, Associate Professor of the Department of Public Administration, Economic and Information Security, BSETU, Bryansk, Russia;
orcid.org/ 0000-0001-6644-418X, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
Three modes of joint application of machine learning and simulation methods in decision support systems are studied. The purpose of this work is to apply these methods jointly to improve the efficiency of generating proposals and conducting appropriate analysis in decision support systems for managing socioeconomic processes and systems on the example of a human potential research in the Bryansk region. In the context of approbation of design solutions, this work aims to develop a predicative model of the human potential of the Bryansk region. To generate synthetic data necessary for training the primary neural network of long-term short-term memory, the authors have developed a simulation model of the region's human potential. The main approaches to modeling are defined as system dynamics and agent-oriented approach. On the basis of transfer training, the pre-trained model is transferred to the solution of forecasting tasks on real data.
Key words: machine learning; simulation modeling; decision support systems; human potential model of the region.