<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.0 20120330//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta>
      <journal-title-group>
        <journal-title>А.: The formation method of linguistic standards created for the intrusion de-
tection system. Ukrainian Information Security Research Journal</journal-title>
      </journal-title-group>
    </journal-meta>
    <article-meta>
      <article-id pub-id-type="doi">10.1109/TCSET.2016.7452207</article-id>
      <title-group>
        <article-title>Method of Informational and Psychological Influence Evaluation in Social Networks Based on Fuzzy Logic</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>National Aviation University</string-name>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Ukraine</string-name>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Institute of Special Communication</string-name>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>Central China Normal University</institution>
          ,
          <addr-line>Wuhan</addr-line>
          ,
          <country country="CN">China</country>
        </aff>
        <aff id="aff1">
          <label>1</label>
          <institution>ty of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”</institution>
          ,
          <addr-line>Kyiv</addr-line>
          ,
          <country country="UA">Ukraine</country>
        </aff>
      </contrib-group>
      <pub-date>
        <year>2018</year>
      </pub-date>
      <volume>11</volume>
      <issue>2</issue>
      <fpage>0000</fpage>
      <lpage>0002</lpage>
      <abstract>
        <p>Recent years number of influences on the information environment of the states has been increased by non-military methods, which is due to the high level of development of information technologies and social engineering. Information and psychological influences become extremely widespread, as evidenced by the publications of the advanced countries where information security issues are discussed in the context of information and psychological influences. The main areas that become the goals of such influences are economic, military, political, and the means through which they are implemented, mainly media, social networks. Obviously, for effective counteraction, timely identification and identification of the impact is necessary. However, an equally important task is to clearly and accurately assess possible destructive effects, which leads to informational and psychological interference. There have been described the processes of information confrontation in the aspect of its psychological component. The developed method can be used in information and psychological security, in particular to prevent information and psychological influence on the individual and social group, effective implementation of countermeasures. The method of evaluation of informational and psychological influences and the corresponding structural solution of the system of evaluation of information and psychological influences, which, by processing fuzzy identifying parameters, allows us to assess the destructive effect of informational and psychological influences and create such systems for fuzzy environment.</p>
      </abstract>
      <kwd-group>
        <kwd>Model</kwd>
        <kwd>Informational and Psychological Influence</kwd>
        <kwd>Methods of Informational and Psychological Influence</kwd>
        <kwd>Evaluation</kwd>
        <kwd>Fuzzy Logic</kwd>
      </kwd-group>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>-</title>
      <p>Information has become an integral part of the activity in the modern environment.
Modern external and internal policies are impossible without taking into account and
using effective forms of informational and psychological confrontation. In recent
years, the number of influences on the information environment of the states has been
increased by non-strain methods, which is due to the high level of development of
information technologies and social engineering. Information and psychological
influences become extremely widespread, as evidenced by the publications of advanced
countries where information security issues are discussed in the context of
information and psychological influences. The main areas that become the goals of such
influences are economic, military, political, and the means through which they are
implemented, mainly media, social networks, global computer networks, rumors, etc.
Obviously, for effective counteraction, timely identification and identification of
influence is necessary. However, an equally important task is to clearly and accurately
assess of the possible destructive effects, which leads to information and
psychological intervention. In this regard, the actual task is to develop a system for assessing the
destructive actions of information and psychological influences.
2</p>
    </sec>
    <sec id="sec-2">
      <title>Modern expert evaluation methods</title>
      <p>
        The role of expert methods assumes special importance during the period of unstable
development of informational, social, economic and other processes that directly or
indirectly affect the activity of public administration bodies and local governments.
Thus, the expert systems are actively implemented in priority tasks [
        <xref ref-type="bibr" rid="ref1">1</xref>
        ], risk
identification and management [
        <xref ref-type="bibr" rid="ref2">2</xref>
        ], various tasks of information systems option [
        <xref ref-type="bibr" rid="ref3">3</xref>
        ], for the
detection and evaluation of critical situations [
        <xref ref-type="bibr" rid="ref4">4</xref>
        ] etc.
      </p>
      <p>
        The basis of such systems, besides the actual knowledge created by experts, is the
socalled heuristic or production rules [
        <xref ref-type="bibr" rid="ref3 ref5">3, 5</xref>
        ]. Under such circumstances, a high degree of
uncertainty in the environmental factors influence is a defining attribute, and therefore
an acceptable accuracy of results cannot be provided by any statistical or other
formalized methods, no matter how perfect they are [
        <xref ref-type="bibr" rid="ref3">3</xref>
        ].
      </p>
      <p>Expert methods are intended to predict qualitative and quantitative characteristics, the
development of which is not completely or partially subject to mathematical
formalization due to the lack of sufficient and reliable statistics. The core of expert
forecasting method is based on qualified specialists assessments on a specific problem
(experts or groups of experts), formed according to the certain rules for problems solving
or forecasts development, the conclusions are drawn about the development paths of
the forecasting object.</p>
      <p>Among the expert techniques there are two main groups distinguished: quantitative
and qualitative. Quantitative methods of expert technologies are based on the
logicalmathematical and statistical methodologies application for generalizing expert
opinions, testing statistical significance of the assessment results, confirming the
refutation of the assessment quality in general.</p>
      <p>
        Similar expert approaches were set up to create decisive rules relating to the attacker
detection in the information-communication networks and systems [
        <xref ref-type="bibr" rid="ref6 ref7">6-10</xref>
        ] and in the
method of criticality level assessment as the emergence result of the information
security incidents [11].
      </p>
      <p>Method of detection and identification of informational and
psychological impact
During the modern warfare, the following classification can be made:
1) methods aimed at people who perceive the information critically:
─ change of opinion by persuasion;
─ psychological isolation of the object;
─ coercion;
─ propaganda.
2) methods aimed at people who perceive the information uncritically:
─ misinformation;
─ propaganda;
─ change of sights by suggestion;
─ infection;
─ manipulation;
─ reframing [12].</p>
      <p>As we can see, there is a certain imbalance among the methods, which are directed at
people who perceive information uncritically in most cases. This situation is due to
the fact that it is much easier to achieve a result, to carry out an attack, if the attacker's
actions are aimed at non-critical thinking, since they will bypass a certain
"psychological shield" of a person.</p>
      <p>Propaganda is attributed to both groups, because of the variety of means, it is evident
that its use is equally effective for all people [13].</p>
      <p>Concepts and classifications regarding information psychological impact analysis has
shown that today there is no single classification that would cover all aspects and
characteristics of its implementation during the information warfare [13].
In the course of the research, the following evaluation parameters of the informational
and psychological influences were identified: CSA – «Completeness and strength of
argument», CGN – «Consistency with the norms of general public opinion», PR –
«Public reaction», GAF – «Growth of the anxiety factor», VD – «Velocity of
distribution», NAT – «Number of affected targets».</p>
      <p>The standard values were constructed in accordance with [10-12].</p>
      <p>
        For the CSA parameter, the following linguistic estimates are: {low (L), medium (M),
high (H)}. Intervals for defining reference values = {[
        <xref ref-type="bibr" rid="ref1 ref2 ref3 ref4 ref5 ref6 ref7">0-20</xref>
        ], [21-40], [41-60]}
intervals of time.
      </p>
      <p>After the operations we will form the following terms of the linguistic variables for
this parameter:
L = {0/0,33; 1/0,33, 0,63/0,75; 0,43/1; 0/1},
M = {0/0,33; 0,22/0,33; 1/0,75; 0,57/1; 0/1},
H= {0/0,33; 0,33/0,33; 0,75/0,75; 1/1; 0/1}.</p>
      <p>
        For the CGN parameter, the following linguistic estimates are typical: {unmatched
(U), medium agreed (M), agreed (A)}. Intervals for determining reference values =
{[
        <xref ref-type="bibr" rid="ref1 ref2 ref3 ref4 ref5 ref6 ref7">0-33</xref>
        ], [34-66], [67-100]} percent.
      </p>
      <p>After the operations we will form the following terms of the linguistic variables for
this parameter:
U = {0/0,33; 1/0,33; 0,92/0,66; 0,4/1; 0/1},
M= {0/0,33; 0,35/0,33; 1/0,66; 0,07/1; 0/1},
A = {0/0,33; 0,1/0,33; 0,44/0,66; 1/1; 0/1}.</p>
      <p>
        For the PR parameter, the following linguistic estimates are: {small (S), medium (M),
high (H)}. Intervals for determining reference values = {[
        <xref ref-type="bibr" rid="ref1 ref2 ref3 ref4 ref5 ref6 ref7">0-33</xref>
        ], [34-66], [67-100]}.
After the operations we will form the following terms of the linguistic variables for
this parameter:
S = {0/0,33; 1/0,33; 0,89/0,66; 0,36/1; 0/1},
M = {0/0,33; 0,7/0,33; 1/0,66; 0,27/1; 0/1},
H = {0/0,33; 0,3/0,33, 0,56/0,66; 1/1; 0/1}.
      </p>
      <p>
        For the GAF parameter, the following linguistic estimates are: {slow (S), medium(M),
high (H)}. Intervals for determining reference values = {[
        <xref ref-type="bibr" rid="ref1 ref2 ref3 ref4 ref5 ref6 ref7">0-33</xref>
        ], [34-66], [67-100]}
percent.
      </p>
      <p>After the operations we will form the following terms of the linguistic variables for
this parameter:
S = {0/0,33; 1/0,33, 0,78/0,66; 0,6/1; 0/1},
M = {0/0,33; 0,27/0,33; 1/0,66; 0,5/1; 0/1},
H = {0/0,33; 0,18/0,33; 0,44/0,67; 1/1, 0/1}.</p>
      <p>
        For the VD parameter, the following linguistic estimates are: {slow (S), medium (M),
high (H)}. Intervals for determining reference values = {[
        <xref ref-type="bibr" rid="ref1 ref2 ref3 ref4 ref5 ref6 ref7">0-33</xref>
        ], [34-66], [67-100]}
percent.
      </p>
      <p>After the operations we will form the following terms of the linguistic variables for
this parameter:
S = {0/0,33; 1/0,33, 0,91/0,66; 0,75/1; 0/1},
M = {0/0,33; 0,54/0,33; 1/0,66; 0,33/1; 0/1},
H = {0/0,33; 0,23/0,33; 0,73/0,67; 1/1, 0/1}.</p>
      <p>
        For NAT characteristic parameter such linguistic assessment {low (L), medium (M),
high (H)}. Intervals for determining reference values = {[
        <xref ref-type="bibr" rid="ref1 ref2 ref3 ref4 ref5 ref6 ref7">0-33</xref>
        ], [34-66], [67-100]}.
After the operations we will form the following terms of the linguistic variables for
this parameter:
L = {0/0,33; 1/0,33, 0,57/0,66; 0,33/1; 0/1},
M = {0/0,33; 0,5/0,33; 1/0,66; 0,56/1; 0/1},
H = {0/0,33; 0,17/0,33; 0,36/0,66; 1/1, 0/1}.
      </p>
      <p>
        For the DR parameter, the following linguistic estimates are: {short-term (S),
medium-term (M), long-term (L)}. Intervals for determining reference values = {[
        <xref ref-type="bibr" rid="ref1 ref2 ref3 ref4 ref5 ref6 ref7">0-33</xref>
        ],
[34-66], [67-100]}.
      </p>
      <p>After the operations we will form the following terms of the linguistic variables for
this parameter:
S = {0/0,33; 1/0,33, 0,95/0,66; 0,78/1; 0/1},
M = {0/0,33; 0,42/0,33; 1/0,66; 0,39/1; 0/1},
L = {0/0,33; 0,13/0,33; 0,4/0,66; 1/1, 0/1}.</p>
      <p>Let’s represent the calculated reference values in the form of graphs (see Помилка!
Джерело посилання не знайдено.) [13].</p>
      <p>In the further, we calculate the for each estimation parameter. At this stage, an
assessment is made of the criticality of the impact of each valuation parameter and their
respective ranking. We apply for this method a quantitative pair comparison with the
definition of the square root, which is a kind of method of quantitative pair comparison.
Pair comparison is the procedure for setting preferences options by comparing all
possible pairs and further streamline object on the basis of comparison [18]. The
paired comparison method is one of the most widely used expert procedures for
determining the relative weights of objects.</p>
      <p>The basis is the comparison of each of the table parameters and the formation of the
matrix of the pair comparison A  aij , where aij selected according to experts on a
scale of relative importance: 1 - alternatives are equally important, 3 - moderate
advantage of one parameter over another, 5 - significant advantage of one parameter
over another, 7 - significant advantage (convincing evidence available), 9 - obvious
advantage of one of the parameters; 2, 4, 6, 8 - intermediate solutions.
The expert fills the locations of the comparison table of the factor with itself gives a
unit. In the first location of the first line, the expert writes a unit, in the second - the
result of a comparison of the first factor with the second, in the third - the result of a
comparison of the first factor with the third, etc. Moving to the second line, the expert
writes in the first location the result of the comparison of the second factor with the
first, in the second - the unit, in the third - the result of comparison of the second
factor with the third, etc. [19].</p>
      <p>Next, we calculate the weight coefficients according to the expression
i  n  iI1 aij , where i  1, I , I is number of evaluation parameters, in this case 7.
After that the valuation of the obtained coefficients is carried out according to the
formula: i  i / ( iI1i , so that</p>
      <p> iI1i  1 .</p>
      <p>Consider an example of determining the importance coefficients for the
predetermined valuation parameters (Table 1). The expert evaluates the importance of each of
them in comparison with the other and puts the information in the table. The
coefficients of importance are calculated and their rationing is carried out. The result of a
pair wise comparison of the estimated parameters of informational and psychological
influences Pi
The next step is to conduct a ranking of the estimated parameters on the calculated
and normalized factors of importance. As a result of calculations, the parameter
«Growth the factor of anxiety» gets the highest score, and therefore, according to the
expert, is the most priority among the other parameters (Table 2).
The problem of assessing the level of criticality of informational and psychological
influence as one of the processes of providing informational and psychological
influence is determined by the fact that its occurrence and development are difficult to
predict (and often not even predictable), that is, we are dealing with an event in an
unclearly formalized space. In addition, there are no generally accepted criteria for
assessing the level of criticality, most of them have different nature (including clear
and unclear) and mathematical properties, which makes it impossible to use most of
the currently known estimation methods to the general set of these criteria. Therefore,
the formation of parameters and the development of methods for assessing the level of
criticality of informational and psychological influence and the methods for its
identification is an relevant task. The method uses the following methods of fuzzy logic as
a method of linguistic terms using statistical data (MLTS) - for constructing reference
values of parameters and evaluation standards, linear approximation by local maxima
(LALM), generalized Heming’s distance (DH) - for processing fuzzy data and
conducting operations of fuzzy logic. In addition, expert methods of evaluation and
ranking are used: the method of average grades (AG).</p>
      <p>The next step is to calculate the overall assessment of the criticality of the situation.
Initially, taking into account the specific methods of information and psychological
impact, the low frequency is formed:</p>
      <p>E
LSCi   (e Le ) (1)</p>
      <p>e1
The current low frequency is compared with the reference standard by one of the
known methods of comparison of the low frequency. For these purposes, we use the
method of forming the -- level nominalization of the low frequency [16] and the
method for identifying the terms [14]. The procedure is to calculate nominalized
(transformed) standards and levels. The definition of the generalized Heming’s
distance is carried out then. The criterion for LCS compliance with one of the terms of
the benchmark is the smallest Heming distance.</p>
      <p>At the final stage, visualization of the results takes place. In addition, in order to
better reflect the criticality of the IPCC, it is proposed to display the criticality
parameters using the critical indicator. To do this, the appropriate Le parameters should be
pre-defused. The most expedient in this case is the application of the method of the
center of gravity, by which the LF is converted into a clear by the formula:
q q
L  100 ( X Lq (xLq ) / (xLq )) (2)</p>
      <p>i1 i1
where q is number of LOAs. A case where the values of individual parameters are
calculated directly without the use of expert methods is possible. In this case, they are
displayed on the indicator by histogram. The development of the method reflects a
new approach to solving the problem of assessing information and psychological
influence. The method is based on fuzzy logic. During its implementation there are
several stages, which are aimed at determining the reference and current values of a
certain information space for the detection of information and psychological
influence. A special feature is the provision of the information and psychological influence
evaluation process, which will be useful in improving the effectiveness of the
development and implementation of countermeasures.
4</p>
    </sec>
    <sec id="sec-3">
      <title>Experimental study of the proposed method</title>
      <p>We use the example of the political anti-advertising campaign of the election
company P. Poroshenko - the video "50 shades of Poroshenko", which was shown on the 1 +
1 channel before the first round of presidential elections in Ukraine [21]. The video
tells about corruption schemes headed by Petro Poroshenko, but not a single concrete
fact was given, which indicates this. The video and individual cuts from it, the
information was actively disseminated to the official page of the TV channel 1 + 1,
affiliated with it pages, hostile to the presidential candidate by public groups, as well as
networks of bots on the social network Facebook. Previously, the following method
of information and psychological influence was identified and identified as
"Manipulation". Identifying parameters of this method are CSA "Completeness and Strength
of Argumentation", CGN "Consistency with Common Public Opinion Standards",
GAF "Anxiety Rise".</p>
      <p>By expert comparison, coefficients of importance for the estimation parameters
"Manipulation" were established. Thus, the following coefficients were obtained for the
importance of fuzzy estimation parameters: Ω1 = 0,115; Ω2 = 0.333; Ω3 = 0.552.
The measured and analytical data are entered in the table and after the phasation
process the value of the estimating parameters is determined and the index of the level of
criticality in the fuzzy form is calculated, which then translates into a clear form after
the dephasing, and the results are displayed on the indicator of the criticality of the
situation. In Table 3 the results of critical assessment of "Manipulation" in the social
network Facebook about the video "50 shades of Poroshenko" are given.
To reflect the level of criticality of the situation in a linguistic form, the procedure for
determining the Heming distance between the terms of the estimated benchmark and
the level of LCS is carried out. Having carried out necessary calculations of Heming
distance, with the help of SOKS received that the current level of criticality situation
"High". In addition, the assessment was carried out at various stages of information
and psychological impact, which enabled to show the dynamics of the situation and
assess the adequacy of the countermeasures involved. Figure 1 shows an indicator of
the criticality of the situation in the development of the IPV for "Manipulation. The
overall criticality level is 88.</p>
      <p>0
20
40
60
80
88
The system of evaluating destructive actions of informational and psychological
influence was developed in the work to solve the problem of destructive effects of
informational and psychological influence, which can be used in a weakly formalized
environment close to real conditions. To achieve the goal were performed the
following tasks:
There was made the analysis of the concept of informational and psychological
influence, its place in the modern information environment was determined, the main
methods and methods of implementation of informational and psychological influence
were investigated and the existing methods of estimating informational and
psychological influence were analyzed. This has made it possible to find that there is no
generalized and sufficiently universal of informational and psychological influence of
evaluation system, and, on the basis of existing highly specialized systems, identify
the shortcomings that most need to be finalized.</p>
      <p>The method of carrying out an evaluation of informational and psychological
influence destructive actions is developed. All the deficiencies of existing evaluation
systems identified because of the analysis are taken into account. The most universal
estimation parameters are determined. This method is based on quantitative methods
of expert evaluation, which gives advantages in the absence of the need to collect
large amounts of statistical data and clear formalization of the current situation.
The structure of evaluation of destructive of informational and psychological
influence actions is developed, which allows estimating the level of destructive
informational and psychological influence actions in a fuzzy environment close to real
conditions. The scheme of the architecture of the system is presented, which allows more
detailed presentation of the modules and processes with which they interact, input and
output data to each of the blocks and modules.</p>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          1.
          <string-name>
            <surname>Gambo</surname>
            ,
            <given-names>Ishaya P.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Abimbola</surname>
            <given-names>H.</given-names>
          </string-name>
          <string-name>
            <surname>Soriyan</surname>
          </string-name>
          , Rhoda N.
          <article-title>Ikono: A Proposed Process Model for Requirements Engineering using Delphi Techniques for Prioritisation</article-title>
          .
          <source>IJ Information Technology, and Computer Science</source>
          <volume>1</volume>
          :
          <fpage>73</fpage>
          -
          <lpage>80</lpage>
          (
          <year>2015</year>
          ).
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          2.
          <string-name>
            <given-names>Danik</given-names>
            <surname>Yu</surname>
          </string-name>
          .,
          <string-name>
            <surname>Hryschuk</surname>
            <given-names>R.</given-names>
          </string-name>
          , Gnatyuk S.:
          <article-title>Synergistic effects of information and cybernetic interaction in civil aviation</article-title>
          ,
          <source>Aviation</source>
          , vol.
          <volume>20</volume>
          , №3, рр.
          <fpage>137</fpage>
          -
          <lpage>144</lpage>
          (
          <year>2016</year>
          ).
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          3.
          <string-name>
            <surname>Zaied</surname>
          </string-name>
          , Abdel Nasser H.,
          <article-title>Samah Ibrahim Abdel Aal, and Mohamed Monir Hassan</article-title>
          .
          <article-title>Rulebased expert systems for selecting information systems development methodologies."</article-title>
          <source>International Journal of Intelligent Systems and Applications 5</source>
          .9:
          <issue>19</issue>
          (
          <year>2013</year>
          ).
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          4.
          <string-name>
            <surname>Gizun</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Gnatyuk</surname>
            ,
            <given-names>V.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Balyk</surname>
            ,
            <given-names>N.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Falat</surname>
            ,
            <given-names>P..</given-names>
          </string-name>
          <article-title>Approaches to improve the activity of computer incident response teams</article-title>
          .
          <source>In Proceedings of the 2015 IEEE 8th International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications</source>
          , Warsaw, Poland,
          <source>September 24-26</source>
          ,
          <year>2015</year>
          : vol.
          <volume>1</volume>
          , pp.
          <fpage>478</fpage>
          -
          <lpage>483</lpage>
          (
          <year>2015</year>
          ).
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          5.
          <string-name>
            <surname>Trach</surname>
            ,
            <given-names>O.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Peleshchyshyn</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          :
          <article-title>Development of directions tasks indicators of virtual community life cycle organization</article-title>
          .
          <source>In Proceedings of 2017 12th International Scientific and Technical Conference on Computer Sciences and Information Technologies</source>
          (
          <year>2017</year>
          ).
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          6.
          <string-name>
            <surname>Gizun</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Gnatyuk</surname>
            ,
            <given-names>V.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Suprun</surname>
            ,
            <given-names>O.</given-names>
          </string-name>
          :
          <article-title>Formalized model for building heuristic rules for detecting incidents</article-title>
          , Bulletin of Engineering Academy of Ukraine, №
          <volume>1</volume>
          , pp.
          <fpage>110</fpage>
          -
          <lpage>115</lpage>
          (
          <year>2015</year>
          ).
        </mixed-citation>
      </ref>
      <ref id="ref7">
        <mixed-citation>
          7.
          <string-name>
            <surname>Korobiichuk</surname>
            ,
            <given-names>I.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Fedushko</surname>
            ,
            <given-names>S.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Juś</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Syerov</surname>
            ,
            <given-names>Y.</given-names>
          </string-name>
          :
          <article-title>Methods of Determining Information Support of Web Community User Personal Data Verification System</article-title>
          . In: Szewczyk R.,
          <string-name>
            <surname>Zieliński</surname>
            <given-names>C.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Kaliczyńska</surname>
            <given-names>M</given-names>
          </string-name>
          . (eds).
          <source>Automation</source>
          <year>2017</year>
          .
          <article-title>ICA 2017</article-title>
          .
          <article-title>Advances in Intelligent Systems and Computing</article-title>
          . Volume
          <volume>550</volume>
          . pp
          <fpage>144</fpage>
          -
          <lpage>150</lpage>
          . Springer (
          <year>2017</year>
          ). DOI:
          <volume>10</volume>
          .1007/978-3-
          <fpage>319</fpage>
          -54042-9_
          <fpage>13</fpage>
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>