=Paper=
{{Paper
|id=Vol-2649/paper1
|storemode=property
|title=Machine Learning Methods and Models, Predictive Analytics and Applications
|pdfUrl=https://ceur-ws.org/Vol-2649/paper1.pdf
|volume=Vol-2649
|authors=Lidiya Guryanova,Roman Yatsenko,Nadija Dubrovina,Vitalina Babenko
}}
==Machine Learning Methods and Models, Predictive Analytics and Applications==
https://ceur-ws.org/Vol-2649/paper1.pdf
1
Machine learning methods and models, predictive analytics
and applications
Lidiya Guryanova 1[0000-0002-2009-1451], Roman Yatsenko2 [0000-0001-7968-6890], Nadija
Dubrovina3[ 0000-0003-1346-9708], Vitalina Babenko4 [0000-0002-4816-4579]
1.2,.
Simon Kuznets Kharkiv National University of Economics, Ukraine,
guryanovalidiya@gmail.com, roman.yatsenko@hneu.net
3
School of Economics and Management in Public Administration in Bratislava, University of
Economics in Bratislava, Slovakia, nadija@mail.ru
4
V. N. Karazin Kharkiv National University, Ukraine, vitalinababenko@karazin.ua
Abstract. This is an introductory text to a collection of articles selected from the
MPSESM-XII conference: Modern Problems of Modelling Socio-Economic
Systems, which took place in Kharkiv, Ukraine, on April 9-10th, 2020. Due to
quarantine measures in connection with COVID-19, several sections were working
on June 25th, 2020 on the basis of S. Kuznets Kharkiv National University of
Economics https://ek.hneu.edu.ua/. The following is a brief overview of the main
scientific schools on modelling systems, of the results of their work.
Key words: predictive analytics, machine learning methods, Data Science
applications for economics, business, healthcare
1 Introduction
The current stage of development of systems is characterized by the strengthening of
globalization and digitalization trends, which is confirmed by the dynamics of the index
of the level of globalization of macroregions (KOF Globalization Index, 2019) and the
index of the level of development of information and communication technologies (ICT
Development Index, 2019). The current trends lead to a qualitative change in the
conditions for the functioning of socio-economic systems and business structures
(hereinafter SES) at various levels of the hierarchy. In particular, the main features of the
external environment for the functioning of SES are:
increased competition, which turns out to be more and more aggressive in the global
market. This forces the SES management to look for new strategies for innovative
development, which are aimed at creating value innovation, reconstructing market
boundaries, creating a new space free from competition;
Copyright © 2020 for this paper by its authors. Use permitted under Creative Commons
License Attribution 4.0 International (CC BY 4.0).
2
wide opportunities for changing business models and optimizing the company's
business processes. Networked forms of business organization are becoming more and
more widespread, virtual corporations that have almost no assets, unite geographically
distributed groups of employees who create a digital asset, a virtual product that is
implemented in the Internet space;
increased volatility of the external environment. Globalization and digitalization, on
one hand, provide ample opportunities for outstripping growth in sectors with a high level
of international competitiveness, and, on the other hand, they cause the emergence of new
risks and threats in the field of financial, energetical, environmental, etc. security. It is no
coincidence that the last decades have been characterized by permanently recurring
financial, economic, social, etc. crises of a strong severity due to the resonant interaction
of cycling factors, an increase in the likelihood of infection by a crisis from partner
countries.
The changed conditions for the functioning of the SES require the improvement of the
model basis of information and analytical control systems, which make it possible to
analyze poorly predictable trends, to identify patterns of development and cause-and-
effect relationships in a multifactorial space, when the dynamics of the development of
research objects is determined by thousands of rapidly changing factors, to diagnose
unfavorable trends in their early stages and to develop preventive management solutions
to ensure safe, sustainable and competitive development of SES.
2 MPSESM: main scientific schools
MPSESM is an international scientific and practical conference, which, since 2009, has
been bringing together scientists from different countries in the field of problems of
developing a model basis for information and analytical control systems. The following
scientific schools function within the framework of the conference:
Predictive analytics and econometric modelling. The main focus of research by
scientists of this scientific school is modelling the dynamics of financial markets,
optimizing portfolio and trading strategies, developing adequate models for predicting the
characteristics, state and behavior of systems under conditions of uncertainty and risk, of
incomplete information and increasing turbulence of the external environment. So, within
the framework of this direction, a methodological approach has been developed, which,
using simple random number generators and a number of hypotheses about the properties
of a random process, the socio-demographic characteristics of the sample, allows you to
form a random sample with indicators: gender, age, health status and the likelihood of
contracting a certain disease, which depends on gender, age and health status, for which it
is possible to obtain expert assessments at the given points. The developed approach
enables healthcare professionals to model and predict various morbidity processes for
3
a particular population, assess possible social and economic risks, determine the need for
the necessary resources for treatment and implementation of preventive programs aimed
at reducing morbidity. The proposed approach is recommended for use in the design of an
information and analytical system for the formation of a comprehensive anti-crisis
program for leveling the consequences of the "shock" of COVID-19. According to
experts, the following developments recognized as interesting which study the possibility
of using models with multifactor volatility to predict the price of derivatives, using the
Hobson-Rogers models to study the dynamics of the PFTS index and find the volatility of
the value of financial instruments. Models of forecasting systemic transformations in the
resort and recreational economy, which allow decision makers to carry out scenario
analysis and parameterization of strategies that ensure their sustainable development, also
aroused quite high interest.
Models for assessing and analyzing the development of territories. Within the
framework of this scientific school, the focus of research is concentrated on the problem
of uneven development. This direction is in the trend of studies by Stiglitz (2013), Piketty
(2014), which indicate an increase in uneven development, increased economic
concentration, widening inequality in income and wealth, which threatens the long-term
prosperity and sustainable development of SES. Numerous international institutions such
as the World Bank, IMF, OECD and WEF also attach high importance to the identified
problems of economic growth, inequality and sustainability, the need to form an effective
strategy for inclusive economic growth. These problems are considered in the works of
Ukrainian scientists (Klebanova at al., 2011), which propose a model basis for assessing
the unevenness and cyclical development of territories to form a balanced development
strategy that ensures sustainable development of both individual regions and the state as a
whole. According to experts, within the framework of the designated research area, the
proposed selective adaptive model for predicting the index of uneven territorial
development was of particular interest, which allows for both structural and parametric
adaptation of the forecasting system and, on this basis, to increase the validity of medium-
term forecasts.
Methods and models of Data Science, Machine learning: analytical research in
economics. This scientific school is aimed at the development of knowledge management
technologies, cognitive management as a systematic management of the processes through
which knowledge is identified, accumulated, distributed and applied in an organization to
improve its performance. Within the framework of this direction, priority tasks are the
ones of pattern recognition, identification of classes of situations, classification of
situations for which differentiated control strategies can be developed. These tasks are
effectively solved using machine learning and Data Science methods. Within the
framework of this direction, the most interesting, according to experts, were the
developments related to the use of cluster analysis methods to determine the optimal
locations of business structures, to identify homogeneous groups of customers for which
differentiated digital marketing strategies can be developed; to the use of machine
4
learning methods to assess the sustainability of the competitive positions of global audit
companies, to assess the risk of lending to companies in the agro-industrial sector.
It should be noted that the scientific schools represented at the MPSESM conference
have a wide geography (Fig. 1). We thank scientists from Slovakia, Poland, Bulgaria,
Slovenia, Greece, Austria, Lithuania, Great Britain, Ukraine, Russia, Mexico, Canada,
who took part in the organization and work of the conference.
Fig. 1. Geography of MPSESM conference participants
59 papers were submitted to the conference (Fig. 2).
Fig. 2. Number of publications (VI-XII international scientific research conference MPSESM)
The submitted papers were double-blind peer-reviewed by the program committee
members using the conference management system http://mpsesm.org/. 7 selected
publications formed the basis of this collection of articles.
3 Conclusions
The institutional environment of the conference was represented by 47 research
centers, universities, IT companies. In particular, such as "KODA" (Ukraine-Germany),
"Ukrainian Management-Intellect" (Ukraine). The conference was attended by
5
125 scientists, including 47 Doctors of Sciences, 4 corresponding members of the
academies of sciences. We thank the participants for interesting reports and look forward
to further cooperation in the field of predictive analytics and econometric modelling;
modelling the development of territories; modelling of security systems; system analysis
and design of decision support systems; modelling financial processes; information
technology in business and education; development of reflexive control models;
application of machine learning and data science methods for analytical research in
economics and business.
References
1. KOF Globalisation Index. (2020). Retrieved from: https://kof.ethz.ch/en/forecasts-and-
indicators/indicators/kof-globalisation-index.html
2. ICT Development Index. (2019). Retrieved from: https://www.itu.int/en/ITU-D/
3. Piketty, T. (2014). Capital in the Twenty-First Century. Cambridge, MA: Harvard
University Press
4. Stiglitz, J. (2013). The Price of Inequality: How Today’s Divided Society Endangers Our
Future. Retrieved from: http://www.pas.va/content/dam/accademia/pdf/es41/es41-stiglitz.pdf
5. Klebanova, T., Kizim, N., Guryanova, L., Chagovets, L., Chernova, N., Trunova, T.
(2011). Models for assessing the unevenness and cyclical dynamics of the development of
territories. Kharkiv, Kh.: PH "INZHEK"