OPTIMIZATION OF ECONOMIC SECURITY BY MACHINE LEARNING AND EVOLUTIONARY ALGORITHMS

Abstract

This article discusses the problem of ensuring economic security by choosing the optimal set of operations and measures of an organizational and technological nature, taking into account the criterion of optimality in terms of reduced costs or maximum safety under constraints in the form of given allowable costs for them. The solution of this problem by the methods of machine learning and evolutionary optimization algorithms is proposed. The information economy and digital information space built on distributed and accessible information resources represent the necessary knowledge. This knowledge is extracted from large data sets using machine learning methods. Accordingly, to solve the problem, it is necessary to determine what knowledge (properties, patterns) are needed and how to use it. In the generally accepted guidelines for ensuring economic security, this problem is solved in two stages. At the first stage, threats are assessed by calculating security indicators, and at the second stage, appropriate measures are selected to reduce economic risks.