APPLICATION OF BAYESIAN NETWORKS TO DETERMINE THE IMPACT OF HIGHER EDUCATION ON ECONOMIC DEVELOPMENT

Abstract

In modern society, higher education institutions not only serve as educational and research centers but also exert a certain influence on the economy, politics, and social sphere of their presence region. Therefore, the question of evaluating the functioning of universities in a specific region in the interest of its sustainable development becomes relevant.

In this article, the authors propose a model for analyzing the influence of factors characterizing the state of the education system on the gross domestic product (GDP) of the country, based on the mathematical apparatus of Bayesian networks. The advantage of using Bayesian networks (BNs) lies in the robustness of these networks to incomplete, inaccurate, and noisy information.  BNs are used for forecasting, direct and inverse modeling of complex relationships, and decision-making. The proposed model predicts the GDP based on factors such as funding for higher education, research activity, the number of students, and staff. The constructed Bayesian network allows not only direct forecasting of the GDP level based on factors characterizing the state of higher education but also inverse modeling, that is, determining which parameters need to be changed to achieve a certain level of GDP.