=Paper=
{{Paper
|id=Vol-2762/short5
|storemode=property
|title=Cognitive Analysis and Modeling of the Investment Situation in Ukraine
|pdfUrl=https://ceur-ws.org/Vol-2762/short5.pdf
|volume=Vol-2762
|authors=Volodymyr Beglytsia,Sergey Tymofeev,Aleksandr Gozhyj
|dblpUrl=https://dblp.org/rec/conf/ictes/BeglytsiaTG20
}}
==Cognitive Analysis and Modeling of the Investment Situation in Ukraine==
https://ceur-ws.org/Vol-2762/short5.pdf
Cognitive Analysis and Modeling of the Investment
Situation in Ukraine
Volodymyr Beglytsiaa , Sergey Tymofeeva and Aleksandr Gozhyja
a
Petro Mohyla Black Sea National University, Mykolaiv, 54003, Ukraine
Abstract
The article deals with the issues of cognitive analysis and modeling of the investment situation in
Ukraine. With the help of fuzzy cognitive analysis, the main factors influencing the investment situ-
ation were identified. Based on these factors, a fuzzy cognitive model of the investment situation was
built. To construct a fuzzy cognitive model, Silov’s cognitive maps were used. The systemic indicators
of the cognitive map were determined, which made it possible to identify the main factors that affect
the investment situation. The ranking of factors influencing the development of the situation has been
carried out. Based on the analysis of these factors, a probable model of the development of the invest-
ment situation was built. The model was built on the basis of a Bayesian network. The performance of
the model was verified on benchmark data. The probabilities of three possible variants of the situation
development have been determined. The analysis of simulation results is carried out.
Keywords
SInvestment situation, Fuzzy cognitive analysis, Fuzzy cognitive model, Cognitive maps, Systemic indi-
cators, Predictive models, Bayesian network
1. Introduction
Complex socio-economic systems, parameters and laws of behavior which are described, mainly
on a qualitative level, are difficult to predict because the change in their parameters can lead
to difficult predictable changes in their structure and behavior. Therefore, it should be noted
that analyzing and modeling such systems and managing them using traditional approaches
based on analytical description or statistical observation of dependencies between inputs or
output parameters is often impossible, and we have to resort to subjective models based on
information obtained from experts and which is cultivated with the involvement of logical
thinking, intuition and heuristics. Traditional methods of system analysis are not suitable for
solving problems of analysis and forecasting of the state of such phenomena and systems.
Most methods used to analyze weakly structured systems often do not work effectively and
do not find practice in a widespread use, if they are not used systematically with other methods
in the context of the task to be solved.
Not only is the use of sign graphs in the simulation of complex systems, which are based on
expert information [1]. When expert construction of real models of complex socio-economic
ICT&ES-2020: Information-Communication Technologies & Embedded Systems, November 12, 2020, Mykolaiv, Ukraine
" science@chmnu.edu.ua (V. Beglytsia); stymofeev@gmail.com (S. Tymofeev); alex.gozhyj@gmail.com (A.
Gozhyj)
0000-0002-8994-4600 (V. Beglytsia); 0000-0002-9223-1468 (S. Tymofeev); 0000-0002-3517-580X (A. Gozhyj)
© 2020 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR
Workshop
Proceedings
http://ceur-ws.org
ISSN 1613-0073
CEUR Workshop Proceedings (CEUR-WS.org)
systems is practically impossible to build an adequate model and make a forecast of the devel-
opment of the system. But from this position one can find a solution using fuzzy graphs to
analyze the information and use the results of the analysis for further research. One kind of
fuzzy graphs is fuzzy cognitive maps. Fuzzy cognitive maps have an advantage over traditional
models in the form of sign graphs: using special algorithms, it is possible to "trace" the path of
influence of positive and negative connections on one or several elements of the system on a
fuzzy cognitive map (given by a fuzzy graph) [1, 2].
Note that cognitive analysis is very specific and this is manifested in the fact that the formal
methods of analysis are applied to models that describe the subjective vision of the situation.
And that at each stage of the formation of the model, the analyst or expert must make decisions
on which the adequacy of the model depends, so the expert is a very important task, and his
decisions form the further course of the analysis of the model. Such decisions include the choice
of the model itself, the formation of a set of factors and relationships between them, the choice
of scales and weight of ties, the choice of methods for calculating impacts. Different methods
will yield different results. The set of models and methods proposed in the literature does not in
itself guarantee the adequacy of the model. Adequacy is finally cleared up only in the process
of real work.
Problem statement. The article explores the use of fuzzy cognitive modeling to study the
investment situation in Ukraine and the use of Bayesian networks for making forecasts based
on fuzzy cognitive analysis.
2. Materials and Methods
This section discusses the issues of modeling the investment situation in Ukraine. The analysis
of the investment situation and the construction of a fuzzy cognitive model of investment
processes are presented. The analysis of system indicators of a fuzzy cognitive map is carried
out and the main factors influencing the investment situation are determined.
2.1. Analysis of the investment situation in Ukraine
For many years, the European Union has been one example of a fairly successful political and
economic system. This applies to both domestic and economic relations. Thus, the European
Union has created favorable conditions for doing business. Large, medium and small businesses
here have enough opportunities for their operation and development. Such positions, together
with political stability and the stability of legislation, have become the basis for ensuring
competitive products, attracting high technology and increasing investment potential.
In view of this, it is significant that in recent years Ukraine has taken confident steps towards
integration with the European Union. In fact, we can say that internal and external conflicts have
led to an increase in the intensity of European integration processes. Pro-European positions
among the political elite and society in general have significantly strengthened. At the same
time, the politically active share of citizens has already clearly decided on their choice to build
a modern democratic state.
One of the main indicators of European integration relations is the investment climate in
Ukraine. The development of the investment market depends on many factors, one of the main
being the development of social and legal institutions in the country and the level of favorable
conditions for the development of the investment situation.
Investment policy is one of the most important components of Ukraine’s economic policy. In
today’s financial and economic crisis, it gets a special priority. At the disposal of the state are
the main means of regulating the production process. Investment policy should help to boost
the economy, increase production efficiency, ensure social and economic stability and solve a
number of other problems.
The analysis of the investment situation in Ukraine in recent years suggests a reduction
of direct investment from foreign investors [3]. The economic situation in Ukraine can be
characterized as unstable. The economy of Ukraine is significantly behind the developed
countries of the world in terms of the combined productivity of all factors of production
and, accordingly, the level of welfare of the population. Most enterprises are technologically
backward, energy intensive, with a weak diversification of products and markets. This is due to
the fact that Ukraine is in a situation of military conflict and instability of the state investment
policy, investment law issues, lack of proper development of the investment market and its
instruments, and as a consequence - insufficient level of domestic and foreign investments [4, 5].
The dynamics of investment in Ukraine is presented in Fig. 1 [4].
Figure 1: The dynamics of investment in Ukraine
Investments are diverted into already developed areas of economic activity. The main volumes
of direct investment revenues were sent to institutions and organizations engaged in financial
and insurance activities - 26.1% and industrial enterprises - 27.3% [4].
The key investor countries include the Netherlands - 33.2%, the Russian Federation - 17.3%,
Cyprus - 16.6%, Austria – 7.1%, France - 3.9%, the United Kingdom - 3.4%, Poland – 3.2%, Germany
– 2.4%, the Virgin Islands (Brit.) - 2.1%, and Switzerland - 1.9% [4].
The following branches of the Ukrainian economy, which are the most attractive for investors,
are agriculture, aerospace technologies, transit potential of Ukraine, metallurgical and chemical
industries. It also attracts investors with an inexpensive and highly skilled workforce. But
a potential investor faces an obstacle already during his first business visit to Ukraine. This
obstacle is corruption. This is a major obstacle to investment, according to an annual poll of
foreign investors conducted by leading sociological agencies.
On the territory of Ukraine, the investor, besides corruption, is waiting for "surprises"
connected with the judicial system, financial instability. But also significantly affect his work
various sectoral monopolies and large corporations that belong to businessmen close to the
authorities (the so-called oligarchs).
On the basis of the analysis of information sources [3, 4] the rating of "negative" factors,
which essentially affect the investment situation in Ukraine, is as follows:
1. Large-scale corruption.
2. Corrupt court system.
3. Non-stable currency and financial system.
4. Monopolization of markets and the influence of oligarchs on power.
5. A conflict with Russia.
6. Problems in the legislative sphere.
7. Corrupt law enforcement structures.
8. Restrictions on the movement of capital and foreign exchange operations.
9. Solid tax administration.
10. Level of labor migration (Given the COVID-19 epidemic).
According to this rating, the actual war with Russia in the East of Ukraine - only in the
fifth place in the rating of obstacles to investment. Also a substantial barriers to investment
in Ukraine is complex and changing legislation, the actions of law enforcement, restrictions
on financial markets, difficult tax administration and labor migration of Ukrainian specialists
abroad.
The analysis of the investment situation in Ukraine shows that the main obstacles to invest-
ments are corruption, a corrupt judicial system and problems in the legislative sphere. In order
to overcome them, it is necessary to systematically consider all factors that affect the investment
climate in the state and determine the priority ways to overcome corruption.
2.2. Fuzzy cognitive model of the investment situation
The development of cognitive modeling methods is largely due to the need to study semi-
structured systems and situations that include many elements of a different nature, and the
relationships between the elements of which are both quantitative and qualitative. Roberts F.
[6] proposed a cognitive approach to the study of semi-structured and complex problems due
to the limited applicability of exact analytical models of complex systems and the study of their
behavior and situations arising from the functioning and development of such systems. With
this approach, the construction of models of semi-structured systems or situations is based on
the subjective understanding and presentation of the control object about the parameters of the
controlled system and the relationship between them.
To understand and analyze the behavior of a complex system, it is necessary to construct a
structural diagram of cause-and-effect relationships in the form of a cognitive map.
To analyze the investment situation, the method of fuzzy cognitive analysis is used. The
method of fuzzy cognitive analysis is based on building a fuzzy map of cognitive problems,
Table 1
List of concepts
№ Identification concept Concept Name
1 C1 Corruption
2 C2 Corrupt court system
3 C3 Non-stable currency and financial system
4 C4 Monopolization
5 C5 A conflict with Russia
6 C6 Problem in the legislative sphere
7 C7 Corrupt law enforcement structures
8 C8 Restrictions on the movement of capital and foreign exchange operations
9 C9 Solid tax administration
10 C10 High level of labor migration from Ukraine from Ukraine
calculating system indicators and analyzing modeling results. The modeling method based on
fuzzy cognitive maps is described in [7, 8].
A fuzzy map of a cognitive problem displays the relationship between the elements of the
system and shows the influence of the elements on each other. The values of the mutual influence
of elements (concepts) on a fuzzy cognitive map are given by fuzzy numbers. Modeling of the
investment situation in Ukraine was carried out using fuzzy cognitive maps of Silov [8].
To build a cognitive map for modeling an investment situation, a list of concepts was developed
based on information about the investment situation (Table 1). Based on these concepts, a
fuzzy cognitive map was built, shown in Figure 2. This fuzzy cognitive map shows the causal
relationships between the main concepts affecting the investment situation in Ukraine.
Build of a fuzzy cognitive map and calculation of system parameters and modeling was
carried out in FCMS environment [9].
Figure 2: Fuzzy cognitive map for modeling investment situation
Based on the fuzzy cognitive map (Fig. 2), a matrix of concepts (Fig. 3) was created, which
describes the causal relationships of the fuzzy cognitive map. Based on the fuzzy cognitive map
(Fig. 2), a matrix of concepts (Fig. 3) was created, which describes the causal relationships of
the fuzzy cognitive map. The values of bonds are determined based on expert assessments.
Figure 3: Matrix concepts of fuzzy cognitive map for modeling
On the basis of the concept matrix, system indicators were calculated. System indicators this
outdegree, indegree and impact on system concepts [10, 8]. According to the values of the system
indicators, the most important concepts in the system and the level of their influence on the
system were identified. The results of calculating system indicators are shown in Figure 4.
Figure 4: Fuzzy cognitive modeling results
Based on the analysis of the results of fuzzy cognitive modeling (calculation and analysis of
system indicators of a fuzzy cognitive map), we can conclude that the most important concepts
in the system that affect the investment situation in Ukraine are: C1 (corruption), C2 (corrupt
judicial system), C3 (corrupt law enforcement structures). The concepts C5 (conflict with Russia),
C9 (tax administration), C10 (labor migration) are more likely a consequence of C1, C2, C3.
The concepts C4 (monopolization), C6 (problem in the legislative sphere), C7 (corrupt law
enforcement structures), C8 (restrictions on the movement of capital and foreign exchange
operations) are causal concepts.
2.3. Investigation of the investment situation of Bayesian networks
For the research and investment situation modeling was used Bayesian network (BN) [2].
Bayesian networks, [11] are widely used for probabilistic modeling and forecasting in infor-
mation processing systems, statistics represented by time series and time sections, as well
as qualitative data represented by expert estimates, linguistic variables and interval values.
The main feature of Bayesian networks is that they allow you to establish cause-and-effect
relationships between events and determine the probability of a given situation occurring when
new information is received about a change in the state of any network node.
Bayesian network can be considered as a model for representing probabilistic dependencies
(relations) between the vertices of an acyclic graph. In this case, with respect to the set of
variables, the Markov condition is satisfied, each variable of the network does not depend on all
other variables, except for the parent predecessors of this variable [12, 13, 14, 15, 16].
To model the investment situation in Ukraine, a model was built in the form of a Bayesian
network. The simulation was performed using a Bayesian network.
The main criteria for determining the current level of investment in Ukraine are tax legislation,
an unstable monetary situation, restrictions on the movement of capital and foreign exchange
transactions, monopolization of markets and influence on the power of oligarchs, labor migration.
, the presence of a military conflict and a significant level of corruption. The Bayesian network
model is shown in Figure 5.
Figure 5: Bayesian network is presented to model the investment situation
Table 2
Bayesian Network Probabilities
№ Outcomes Probability
1 Corruption 0.82
2 Military conflict with Russia 0.59
3 Unstable monetary situation 0.32
4 Monopolization of market 0.68
5 Labor migration 0.45
6 Tax legislation 0.32
7 Restrictions on the movement of capital 0.18
8 Situation 1 0.36
9 Situation 2 0.57
10 Situation 3 0.06
The built Bayesian network models the investment situation in Ukraine. Bayesian network is
built in the environment Genie. The model is based on expert data characterizing the current
situation. As a result of the work of this Bayesian network, the probabilities of investment
situations in Ukraine are calculated. This Bayesian network considers the probabilities of three
situations. Situation 1 - investments are insignificant, Situation 2 - investments grow, the
situation is positive, Situation 3 - there are no investments. Table 2 presents the probabilities
for estimating the current investment situation. By varying the probabilities in the network,
you can model various investment situations. To check the adequacy of the model, modeling
was carried out for the indicators of 2017 and 2018.The model is effective. For more accurate
modeling, it is necessary to build a large cognitive map. But there is a high probability of
error when defining fuzzy links in the model. This may affect the accuracy of the model. The
probabilities of possible outcomes are presented in Table 2. The most probable situation 2.
3. Conclusions
The article discusses an approach to the analysis of the investment situation based on fuzzy
cognitive analysis and probabilistic modeling based on Bayesian networks. A fuzzy cognitive
map was developed to analyze and model the investment situation. With the help of system
indicators, the main factors influencing investments in Ukraine were identified. The main
factors affecting investment are high levels of corruption, hostilities, monopolization of markets,
restrictions on the financial market and high levels of labor migration. Based on the analysis of a
fuzzy cognitive map, a Bayesian network was built for probabilistic modeling of the investment
situation in Ukraine. Using the Bayesian network, the probabilities of the possible development
of the investment situation were calculated.
References
[1] B. Kosko, Fuzzy engineering, Prentice-Hall, Inc., 1996.
[2] A. N. Terentyev, P. I. Bidyuk, L. A. Korshevnyuk, Bayesian network as instru-
ment of intelligent data analysis, Journal of Automation and Information Sciences
39 (2007) 28–38. URL: https://doi.org/10.1615/jautomatinfscien.v39.i8.40. doi:10.1615/
jautomatinfscien.v39.i8.40.
[3] Direct foreign investments in ukraine since 2002 to 2020, 2020. URL: https://www.
prostobank.ua/spravochniki/indikatory_rynka/foreign_direct_investment.
[4] Investment in ukraine: statistics, branches, information for investors, 2020. URL: https:
//dlf.ua/ru/investirovanie-v-ukraine-statistika-otrasli-informatsiya-dlya-investorov/.
[5] Nbu counted foreign investments in ukraine in 2019, 2020. URL: https://finance.liga.net/
ekonomika/novosti/v-nbu-podschitali-inostrannye-investitsii-v-ekonomiku-ukrainy.
[6] F. S. Roberts, Discrete mathematical models, with applications to social, biological, and
environmental problems (1976).
[7] R. Sadiq, Y. Kleiner, B. Rajani, Interpreting fuzzy cognitive maps (fcms) using fuzzy
measures to evaluate water quality failures in distribution networks, in: Joint international
conference on computation in civil and building engineering (ICCCBE XI), 2006, pp. 1–10.
[8] V. Silov, Strategic decision-making in a fuzzy environment, Moscow: INPRO-RES (1995)
228.
[9] A. Gozhyi, I. Kalinina, Information system for a fuzzy cognitive analysis and modeling,
Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska (2014).
[10] E. Papageorgiou, A. Kontogianni, Using fuzzy cognitive mapping in environmental decision
making and management: A methodological primer and an application, in: International
Perspectives on Global Environmental Change, InTech, 2012. URL: https://doi.org/10.5772/
29375. doi:10.5772/29375.
[11] P. Bidyuk, A. Gozhyj, Z. Szymanski, I. Kalinina, V. Beglytsia, The methods bayesian
analysis of the threshold stochastic volatility model, in: 2018 IEEE Second International
Conference on Data Stream Mining Processing (DSMP), 2018, pp. 70–74. doi:10.1109/
DSMP.2018.8478474.
[12] J. Cheng, R. Greiner, J. Kelly, D. Bell, W. Liu, Learning bayesian networks from data: An
information-theory based approach, Artificial Intelligence 137 (2002) 43–90. URL: https:
//doi.org/10.1016/s0004-3702(02)00191-1. doi:10.1016/s0004-3702(02)00191-1.
[13] L. Jouffe, P. Munteanu, New search strategies for learning bayesian networks, in: Pro-
ceedings of the 10th International Symposium on Applied Stochastic Models and Data
Analysis, Compiègne, France, 2001, pp. 591–596.
[14] N. Friedman, M. Linial, I. Nachman, D. Pe'er, Using bayesian networks to analyze expression
data, Journal of Computational Biology 7 (2000) 601–620. URL: https://doi.org/10.1089/
106652700750050961. doi:10.1089/106652700750050961.
[15] L. Falzon, Using bayesian network analysis to support centre of gravity analysis in
military planning, European Journal of Operational Research 170 (2006) 629–643. URL:
https://doi.org/10.1016/j.ejor.2004.06.028. doi:10.1016/j.ejor.2004.06.028.
[16] O. Pourret, P. Naïm, B. Marcot, Bayesian networks: a practical guide to applications, John
Wiley & Sons, 2008.