Modeling of risk level of the socio-economic systems functioning

Abstract

In the process of human evolution, the nature of human activity changes and it is necessary ”tools” are available for solving new problems. Recently, problems related to decision-making have become increasingly important. Particularly relevant are the problems of supporting decisions in the process of managing socio-economic systems. Decision- makers are usually faced with issues of information retrieval, uncertainty, and in some cases, conflict in the decision-making process. At the same time, it is assumed that the implementation of any of the variants of decisions implies the occurrence of certain consequences, the analysis and evaluation of which fully characterizes the chosen variant. Traditionally, complex analytical calculations, expert knowledge, modern information technology tools are used to evaluate the possible consequences. The analysis of the existing practice of managing social and economic systems makes it possible to propose new directions of its optimization, which, in turn, provides an orientation to the programmed indicators of development of both internal system characteristics and parameters of the external environment, taking into account the forecast values of key parameters of the management object. It is the orientation to the projected development indicators that allows you to develop and implement effective strategies for managing pro- cesses in social and economic systems. The importance of owning the tools and techniques of forecasting for the economist and manager in today’s context is undeniable. The purpose of this work, based on the analysis of literature sources, is to draw conclusions about the features, perspectives of use and opportunities for the development of data mining in the current environment of computer technology. The basic methods of machine learning are considered in the paper and the peculiarities and results of their application to solving problems of prediction problems are analyzed. In order to solve the problem, there is a need to identify what areas of technology development scientists need to improve and research. Machine learning is a unit of a fairly broad field of science that studies artificial intelligence. Related algorithms are used to solve problems that often make it difficult or impossible to come up with an explicit algorithm for solving them.

Keywords

• передбачення
• iнтелектуальний аналiз даних
• алгоритми машинного навчання.