<!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 />
    <article-meta>
      <title-group>
        <article-title>Remote Selection of Staff Based on Socionic Analysis of Social Network Content</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Oleksandr Morushko</string-name>
          <email>morushkoo@gmail.com</email>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Natalіia Khymytsia</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Vasyl Teslyuk</string-name>
          <email>vasyl.m.teslyuk@lpnu.ua</email>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>Lviv Polytechnic National University</institution>
          ,
          <addr-line>Lviv, 79013</addr-line>
          ,
          <country country="UA">Ukraine</country>
        </aff>
      </contrib-group>
      <abstract>
        <p>In modern conditions, the ability to work remotely is becoming increasingly important. On the other hand, the limited possibility of direct contact with candidates for a job highlights the ability and skills that someone has when finding skilled workers using remote sensing techniques. In particular, this applies to the assessment of soft skills of applicants, which include high communication skills, language skills, cognitive or emotional empathy, leadership traits. Thus, recruitment managers must have clear criteria for remote assessment of the business qualities of potential employees. As such criteria, the authors recommend the use of the Jung basis. With the help of four pairs of dichotomous features with a high probability, the mentality of a person is determined and on this basis, their ability to be involved in performing a certain type of work is assessed. To do this, the authors propose to use the following methods of analysis: visual, verbal, and if possible - and testing. The study also describes the algorithm for using these methods to form teams, assess their cohesion, and determine the optimal leader. It is suggested to use photos and text messages from social networks as a source of information for such analysis. This type of information is publicly available and quite informative. In addition, such information can be monitored and analyzed without the knowledge of social network users applying for the job. The advantage of the proposed methods is their relative simplicity and accessibility in use. The novelty of the proposed method is the possibility of remote selection of staff, which does not require direct contact with candidates and saves time. This provides significant benefits both during a pandemic Web communities, social networks, socionic analysis, sociometry, commands, visual COLINS-2022: 6th International Conference on Computational Linguistics and Intelligent Systems, May 12-13, 2022, Gliwice, Poland ORCID: 0000-0001-8872-2830 (O. Morushko); 0000-0003-4076-3830 (N. Khymytsia); 0000-0002-5974-9310 (V. Teslyuk)</p>
      </abstract>
      <kwd-group>
        <kwd>1</kwd>
        <kwd>assessment</kwd>
        <kwd>verbal assessment</kwd>
        <kwd>Jung basis</kwd>
      </kwd-group>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>-</title>
      <p>and
when
organizing
remote
staff
work,
without specific
office
space.</p>
      <p>Psychological compatibility of staff will play a key role. However, it should be noted that
they also require sufficient qualified knowledge in the field of practical psychology and
significant practical experience of the expert.</p>
    </sec>
    <sec id="sec-2">
      <title>1. Introduction</title>
      <sec id="sec-2-1">
        <title>Over the past year, we have faced a problem</title>
        <p>where interpersonal and organizational
communications, including oral live communication, have been limited due to the coronavirus
pandemic. Therefore, the task of organizing and supporting business communications, selecting
candidates for vacant positions and creating capable work teams, remotely, has become very
important.</p>
        <p>One of the ways to find the best candidates for vacant positions, create effective teams remotely, in
such conditions of limited access to information about the participants themselves, can be socionic
diagnostics of content from social networks (including photo content and text messages). Analysis of</p>
        <p>2022 Copyright for this paper by its authors.
photos and text messages makes it possible with a high probability to determine the sociotype of each
applicant. Having information about the sociotype of the applicant and the vacancy and the type of
work that needs to be done, it is possible with high probability to predict which candidates to invite to
vacant positions. Also, based on the analysis of the collected information and socionic diagnostics, it
is possible to select a capable working team in a short time to perform tasks remotely.</p>
        <p>To do this, you must meet the following criteria:
• to determine as accurately as possible the sociotype of applicants for vacant positions;
• form work teams based on the principle of comfort for all its members.</p>
        <p>First of all, the task is to determine the sociotype of each applicant using the Jung basis. Of the
three main methods of determination - visual, verbal and testing, without direct contact with the
subject, we can use only the first two methods. If we can confirm the visual information with verbal
messages, it allows us to move to the active phase - the interview.</p>
        <p>If the candidates themselves are interested in working in a team where they will be most
comfortable, they have the opportunity to fill out a questionnaire to determine their sociotype. If we
get identical results based on all these methods, then determining the sociotypes of candidates for
vacant positions and for teamwork will be correct.</p>
        <p>The next stage is the formation of a capable team. In order for a team to function effectively
remotely, it is important to consider the psychological compatibility of the members of that team.
Since it is very important that each team member clearly understands the tasks he has to perform and
what his personal contribution is to these tasks, it is necessary to analyze the intertype relationships
that develop between team members within. Such an indicator of the comfort of being in this team is
the personal coefficient of conflict between the candidates and the team as a whole.</p>
        <p>Determining the coefficient of conflict is proposed by analyzing the psychological distances of
intertype relationships between team members. After the information about the applicants is collected
and processed mathematically, based on the following data analysis, the selection of the optimal
leader or team leader is carried out.</p>
      </sec>
    </sec>
    <sec id="sec-3">
      <title>2. Related Works</title>
      <p>
        Problems of sociometric analysis of groups and communities, various aspects of personnel
selection, are widely covered in scientific papers. The article by Ion Georgiou, Ronald Concer, Andrej
Mrvar offers a method of analyzing the compatibility and diversity of different, interconnected,
structural configurations of groups that are focused on achieving a specific goal. In particular,
scientists consider an approach that meets sociometric principles and methodological and
measurement standards [
        <xref ref-type="bibr" rid="ref1">1</xref>
        ]. Current issues of team formation, based on sociometric analysis of social
networks, are considered in [
        <xref ref-type="bibr" rid="ref2">2</xref>
        ]. Researchers offer their method of solving the problem of finding a set
of paired commands, as harmonious as possible [
        <xref ref-type="bibr" rid="ref3">3,4</xref>
        ]. In [5], and evaluation of the system for the
analysis of human relations based on the use of indistinct theory is proposed.
      </p>
      <p>Analysis of social networks and research on how to appoint experts among several projects to
maximize the efficiency of distribution is presented [6]. To calculate the social relations between
experts in each project, the authors consider an approach based on genetic algorithm and sociometry</p>
      <p>In work [7] the analysis of advantages of the web tool Agent SocialMetric based on the alternation
of the analysis of social networks with intelligent dialog agents is given. Researchers Jiamou Liu,
Ziheng Wei in their work proposed a game model of cohesion, which is not only based on the social
network but also reflects the social needs of people [8]. This model is presented as a type of joint
game in which all participants can gain popularity through the strategic formation of groups.</p>
      <p>The team of authors in [9], using sociometry, investigates the problem of forming a multiple
teams. It is proposed to solve the team optimization model using the following algorithms: constraint
programming provided by a commercial solver; local search heuristics and variable meta-heuristics
for a variable environment.</p>
      <p>Based on an analysis of the social network Twitter, Martin Grandjean identified the structure of
relationships and identified users with a special position. His work also shows that language groups
are key factors to justify clustering in the network [10].</p>
      <p>A new approach to the analysis of social networks is used to measure the social status of students
in the study [11]. It is based on a genetic algorithm (GA) and analysis of social networks for grouping
partners, for the purpose of joint learning in the classroom.</p>
      <p>Also, in the context of our study, interesting is the work [12], where based on the analysis,
identified two structures of communication networks for continuous communication and
communication twice a day: as a shared network and a coherent network. The main nodes or groups
for network architecture in this study are identified through graphical mapping of the network,
quantitative analysis of subjective impressions, and quantitative statistical analysis using sociometric
techniques.</p>
      <p>In the study of I. Khomytska, V. Teslyuk, A. Holovatyy, O. Morushko [13] developed methods
and models based on the theory of mathematical statistics and allow to increase the accuracy of
differentiation of phonostatistical structures of styles. The program system of text differentiation
proposed by the authors and author attribution of the text at the phonological level, where the criterion
of text differentiation is the average frequencies of groups of consonant phonemes, in our opinion, can
also be effectively used for remote staff selection.</p>
    </sec>
    <sec id="sec-4">
      <title>3. Methods of remote selection of personnel based on socionic analysis of the content of social networks</title>
    </sec>
    <sec id="sec-5">
      <title>3.1. Method of selection of socionic types</title>
      <p>Everyone perceives, processes and transmits information differently. These processes depend on
the location of mental functions, their agility and signs of functions (Fig. 1.).</p>
      <p>(
iihnkngT ILCOG
)
1
3</p>
      <sec id="sec-5-1">
        <title>SENSORY (Feeling) EGO</title>
      </sec>
      <sec id="sec-5-2">
        <title>INTUITION (Intuition) 2 4</title>
        <p>( E
S T
e H
sen IC
) S</p>
        <p>Based on Jung's "coordinate system" [14] and the setting of consciousness
(extroverted/introverted), Aushra Augustinavičiūtė singled out 16 socionic types [15], four of which
are formed at the intersection of rational (logic/ethics) and irrational (sensory / intuition) coordinates.)
functions:
1) LSI - logical-sensory introvert;</p>
        <p>SLE - sensory-logical extrovert;
LSE - logical-sensory extrovert;</p>
      </sec>
    </sec>
    <sec id="sec-6">
      <title>3.2. Visual signs of determining sociotypes</title>
      <p>Since we are dealing with 16 basic sociotypes, we must have clear criteria according to which we
can identify the subjects belonging to one of them.</p>
      <p>In general, there are three main methods of identification - visual, verbal, and testing. Obviously,
the best results can be given by the simultaneous application of all three methods. However, in
practice, this is not always possible. Therefore, sometimes it is necessary to limit oneself to visual
assessment, or visual/verbal definition of sociotype.</p>
      <p>In the process of visually / verbally determining the sociotype, we must consistently determine
which of the four dichotomous features of Jung is leading, and accordingly determine the sociotype of
the subject. Tables 1-4 show the visual signs for identification:</p>
      <sec id="sec-6-1">
        <title>Energy and information</title>
      </sec>
      <sec id="sec-6-2">
        <title>Behavior and work style</title>
      </sec>
      <sec id="sec-6-3">
        <title>Psychological orientation</title>
      </sec>
      <sec id="sec-6-4">
        <title>Relationship behavior</title>
      </sec>
      <sec id="sec-6-5">
        <title>Social behavior</title>
      </sec>
      <sec id="sec-6-6">
        <title>Cognitive interests</title>
      </sec>
      <sec id="sec-6-7">
        <title>Gives more</title>
      </sec>
      <sec id="sec-6-8">
        <title>Active, expensive</title>
      </sec>
      <sec id="sec-6-9">
        <title>Understands others better</title>
      </sec>
      <sec id="sec-6-10">
        <title>Attempts to dominate</title>
      </sec>
      <sec id="sec-6-11">
        <title>Expansion, expansion</title>
      </sec>
      <sec id="sec-6-12">
        <title>Large-scale</title>
      </sec>
      <sec id="sec-6-13">
        <title>Accumulates more</title>
      </sec>
      <sec id="sec-6-14">
        <title>Passive, frugal</title>
      </sec>
      <sec id="sec-6-15">
        <title>He understands himself better</title>
      </sec>
      <sec id="sec-6-16">
        <title>Ability to adapt</title>
      </sec>
      <sec id="sec-6-17">
        <title>Deepening, improvement</title>
      </sec>
      <sec id="sec-6-18">
        <title>Local</title>
        <p>To identify verbal signs, we can use the technique already described in our work [16, 17].</p>
      </sec>
    </sec>
    <sec id="sec-7">
      <title>3.3. Methods of determining psychological compatibility</title>
      <p>The comfort/discomfort of the relationship between team members will depend on the type of
intertype relationship and the numerical equivalent into which we can translate it - the psychological
distance. Therefore, having the results of the definition of socionic types, we can use tables 5 and 6 to
obtain the initial data to assess the potential coherence of the team, which includes:</p>
      <p>X  x1, x2 , x3 ,...,xn ,  – some team consisting of n members, and xi ,
i  1, n – some individual of this team. According to the table 5, there is a complete binary relation
 , which determines the intertype relationship between a pair of individuals in the team X. Weight
matrix of the ratio  will be a matrix, defined in table 6 with elements mi, j – absolute values of
intertype distances between individuals of the team X.</p>
      <p>Let's mark ki – the coefficient of conflict of the i-th individual of the team X relative to all its
other members, which will be calculated by the following formula: ki 
, for all values
i  1, n .
coefficients of conflict of all team members and calculated by the formula F ( X ) 
Let's mark F(X) – team conflict function X, the value of which will be the sum of the values of the
n
 k j .</p>
      <p>Through the use of this technique, we can determine both the integral coefficient of conflict and
the personal coefficient of conflict of each member of the team. This allows us to choose the optimal
team leader. This is how we will consider a team member with the minimum value of the coefficient
of conflict. In addition, we have the opportunity to assess the comfort of each member of the team in
this team. The higher the coefficient of conflict, the less comfortable a person will feel in this team
[18].</p>
      <p>Regarding the purposeful selection of staff for vacant positions, in our opinion, it would be
appropriate to use the method of creating a socionic portrait of the ideal candidate. To do this, based
on the application of Jung's basis, we consistently determine which of the four pairs of dichotomous
features will be more optimal. For example, we are looking for a person to be a salesperson. This
profession requires a high degree of contact with clients, the ability to empathize, the ability to
persuade here and now, constant efficiency, and stay in shape. So, it must be ethical, extrovert,
sensory, rational. An ethical-sensory extrovert meets these criteria.</p>
      <p>For the formed team, we can also define such a value as an integral sociotype. Knowledge of the
integrated sociotype of the team allows us to predict what kind of tasks the team can better cope with
[19].</p>
      <p>To search for potential candidates to fill a vacancy, we can use information from open sources
Internet sites for job search / job offers, or information that is posted on social networks by job
seekers. The main criterion for us will be the presence of several fairly clear photos (profile / full
face), which can be analyzed according to Jung's basis, as well as the presence of text messages
(summaries) - for verbal analysis by the above method [20].</p>
    </sec>
    <sec id="sec-8">
      <title>3.4. Examples of application of socionic analysis</title>
      <p>We will consider how this technique can work on concrete examples. Photos and text data for
analysis are taken from the social networks Linkedin and Facebook.</p>
      <p>Example 1.</p>
      <sec id="sec-8-1">
        <title>User Daria Norikova. Figure 1: Photos by Daria Norikova on Linkedin and Facebook.</title>
        <p>From the above photos, we can conclude that the user is a logician because the upper part of the
face is not round, but rectangular. Also with a high probability can be said about extraversion
dynamic poses and the desire to be the center of attention. The shape of the face is pentagonal, which
indicates intuition. And finally - the shoulders are straight, the figure is tightened, that is, it is rational.
Thus, based on the diagnosis, we can conclude that Daria corresponds to the sociotype of a
logicalintuitive extrovert (LIE) - "Jack London".</p>
        <p>Now we will conduct a verbal analysis of Daria Norikova's text messages.</p>
        <p>A detailed analysis of the text message shows us first of all extraversion (very dynamic
presentation of information) and intuition - sensors most likely will not risk going on vacation in the
mountains without their own equipment. The ability not to get confused and use the situation to your
advantage indicates the presence of business logic. It follows that the program function is business
logic, and creative - the intuition of possibilities. So, before us is a logical-intuitive extrovert.</p>
        <p>In this candidate, the upper part of the face is rectangular, which indicates logic. The shape of the
face itself is elongated, teardrop-shaped. This is typical of intuitions. The photo shows that the body
shape is streamlined, the shoulders are lowered. So, before us is irrational. Also in the photo we see
static poses, so most likely we have an introvert. Therefore, our previous hypothesis of Pavel Khud’s
sociotype is an intuitive-logical introvert (OR) - "Balzac".</p>
        <p>Now let's do a verbal analysis.</p>
        <p>From the above in Fig. 4. A fragment of a text message can draw the following conclusions. At the
beginning, the word "assume" is used twice. The use of such words is inherent in intuitions. The
emphasis on the timeliness of certain actions gives us clarification - it is the intuition of time. The text
is quite pragmatic.</p>
        <p>In addition, we did not ignore the text message from the second photo (Fig. 5). Such statements
are inherent in business logic. Actually, we can say that there is a fact of manipulation of business
logic (value in deeds, not words). Therefore, business logic in this case is a manipulative (creative)
function, and software is the intuition of possibilities. There is a high probability that we have an
intuitive-logical introvert.</p>
        <p>Example 2.</p>
      </sec>
      <sec id="sec-8-2">
        <title>User Pavlo Khud. Figure. 3: Photos by Pavel Khud on Linkedin and Facebook.</title>
        <p>If Daria and Pavlo had to work in the same team, their intertype relationship is a mirror image.
Both belong to the representatives of the third quadrant. The psychological distance is minimal - the
first. This is a very favorable, comfortable relationship, provided that there are no elements of direct
competition. Therefore, it would be better for both of them to have the working group they belong to
headed by a third person, preferably an ethicist from the third quadrant.</p>
      </sec>
    </sec>
    <sec id="sec-9">
      <title>4. The analysis of the obtained results</title>
      <p>The outcome obtained as a result of socionic analysis can be presented in the form of a table 7:</p>
      <p>Substitute the results obtained in tables 5 and 6. According to them, these intertype relationships
are mirror images. Both belong to the representatives of the third quadrant. The psychological
distance is minimal - the first. Both belong to the socionic club of researchers. This is a very
favorable, comfortable relationship, provided that there are no elements of direct competition.
Communication is pleasant, easy and almost conflict-free. Quite good interaction and understanding.
Both sociotypes have strong (working) and weak (problem functions), for which they are not able to
provide each other with effective assistance. Therefore, it would be better for both of them to have the
working group they belong to headed by a third person, preferably an ethicist from the third quadrant.
This harmonizes these relationships, makes them constructive [16, P. 157].</p>
    </sec>
    <sec id="sec-10">
      <title>5. Conclusions</title>
      <p>Modern trends in the development of the information society suggest that the relationship between
real and virtual communication will steadily increase in favor of the latter. In turn, this will necessitate
the search for technologies and techniques that will compensate for the loss of some of the
information that was obtained during direct communication, through accessible content from social
networks (photos and text messages).</p>
      <p>Applying the method of socionic analysis in gathering the necessary information can provide a
number of competitive advantages. By knowing the information language in which our
communication partners communicate, it is possible to build a communicative model of
communication in which it will be easier to reach a consensus [21].</p>
      <p>Knowledge of practical psychology in our time is becoming an important communication tool.
They allow you to effectively use human resources, build successful teams, much more effectively
resolve industrial conflicts.</p>
      <p>An important step in this direction is proper personal self-identification. It allows you to develop
individual tools for determining the sociotypes of other people, optimized for your own sociotype.
The method of determination cannot be absolutely the same, as each of the basic 16 sociotypes has its
own characteristics.</p>
      <p>Knowledge of one's own sociotype and the sociotype of one's interlocutor makes it possible to
determine the type of intertype relations and the optimal psychological distance. This information also
allows you to choose the most appropriate information channel (channels) for communication, and
also allows you to consciously avoid communication on controversial topics.</p>
      <p>The successful implementation of projects directly depends on skillfully selected staff, clearly
defined tasks that will face them, minimizing misunderstandings and conflicts that arise in the process
of working together. The biggest responsibility for this lies with line managers, although much will
also depend on how well-coordinated the team was. Proper and appropriate use of socionic analysis
methods can minimize the risks of remote recruitment. The scientific novelty of the method of
socionic analysis proposed by the authors lies in the relative simplicity and accessibility of the
method. It does not require significant financial costs. A qualified specialist can teach any manager in
a relatively short period of time the basic elements of applying this method.</p>
      <p>It is enough to correctly analyze the visual and verbal information available from social networks.
6. References
[4] O. Mulesa, F. Geche, A. Batyuk, Information technology for determining structure of social
group based on fuzzy c-means, in Proceedings of the IEEE Xth International Scientific and
Technical Conference on Computer Sciences and Information Technologies, CSIT 2015, Lviv,
2015, pp. 60-62. doi: 10.1109/STC-CSIT.2015.7325431
[5] T. Yoshizumi, T. Sumida, Y. Shiono, M. Namekawa, K. Tsuchida, (2020) Evaluation of
Advanced Analysis Method for Human Relationship Using Fuzzy Theory. In: Abraham A.,
Cherukuri A., Melin P., Gandhi N. (eds) Intelligent Systems Design and Applications. ISDA
2018. Advances in Intelligent Systems and Computing, vol 940. Springer, Cham.
https://doi.org/10.1007/978-3-030-16657-1_72.
[6] V. S. Baghel and S. D. Bhavani, Multiple Team Formation Using an Evolutionary Approach,
2018 Eleventh International Conference on Contemporary Computing (IC3), 2018, pp. 1-6, doi:
10.1109/IC3.2018.8530662.
[7] А. Kuz, Agent SocialMetric: a web-based platform for teaching environments and conflict
assessment support in the classroom. Complutense Journal of Education, 28 (3), (2017) 941-958.
doi.org/10.5209/rev_RCED.2017.v28.n3.51366.
[8] J. Liu, Z. Wei, Network, popularity and social cohesion: A game-theoretic approach, CoRR,
2016. URL:https://arxiv.org/abs/1612.08351.
[9] Jimmy H. Gutiérrez, César A. Astudillo, Pablo Ballesteros-Pérez, Daniel Mora-Melià,
CandiaVejar, Alfredo. The multiple team formation problem using sociometry, Computer &amp; Operations
Research, Vol. 75, 2016, pp.150-162. doi.org/10.1016/j.cor.2016.05.012.
[10] Martin Grandjean, Aaron Mauro (Reviewing Editor), A social network analysis of Twitter:
Mapping the digital humanities community, Cogent Arts &amp; Humanities, Volume 3, Issue 13,
2016, doi.org/10.1080/23311983.2016.1171458.
[11] Rong-Chang Chen, Shih-Ying Chen, Jyun-You Fan, Yen-Ting Chen, Grouping partners for
cooperative learning using genetic algorithm and social network analysis, Procedia Engineering,
Volume 29, 2012, pp. 3888-3893. doi.org/10.1016/j.proeng.2012.01.589.
[12] Harry L. Litaker, Robert L. Howard, Social network analysis and dual rover communications,
Acta Astronautica, Volume 90, Issue 2, 2013, pp. 367-377.
doi.org/10.1016/j.actaastro.2012.05.013.
[13] I. Khomytska, V. Teslyuk, A. Holovatyy, О. Morushko, Methods, models and means of the
system for differentiation of phonostatistical structures of engglish functional styles.
Development of methods, models and means of authorship attribution of a text. In:
EasternEuropean Journal of Enterprise Technologies. № 3/12 (93), 2018, pp. 41–46.
[14] C.G. Jung, Psychological Types. In: Princeton University Press, 1976.
[15] A. Augustinavichyute, Socionics. Introduction. In: Terra Fantastica, 1998.
[16] O. O. Morushko, N.O. Khymytsia, Humanization of information exchange, Triada Plus, Lviv,
2017.
[17] A. Morushko, N. Khymytsia, N. Shakhovska, Determining the Psychological Portrait of
Members of Web Communities through Socionic Analysis, Proceedings of the 2nd International
Workshop on Control, Optimisation and Analytical Processing of Social Networks (COAPSN
2020), Lviv, Ukraine, 2020, pp. 112-124.
[18] K. Pietrak, The foundations of socionics – A review, Cognitive Systems Research 47, (2018),
pp. 1-11.
[19] M. Vaida, Collaborative Education Teams Development Using Alternative Methodologies,
ICETC 2019: Proceedings of the 2019 11th International Conference on Education Technology
and Computers, Amsterdam Netherlands, October 28 - 31, (2019), pp. 223–227.
[20] D. Zakharov, Recruitment Using Social Networks, Control, 6(1), (2018), pp. 25-30.
[21] V. Kraev, A. Tikhonov, Modern recruitment approaches in personnel selection, Revista,
Espacios, Vol. 41 (Issue 12), (2020), p. 12.</p>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          [1]
          <string-name>
            <given-names>Ion</given-names>
            <surname>Georgiou</surname>
          </string-name>
          , Ronald Concer,
          <string-name>
            <given-names>Andrej</given-names>
            <surname>Mrvar</surname>
          </string-name>
          ,
          <article-title>A systemic approach to sociometric group research: Advancing the work of Leslie Day Zeleny,</article-title>
          <year>1939</year>
          -
          <fpage>1947</fpage>
          ,
          <string-name>
            <given-names>Social</given-names>
            <surname>Networks</surname>
          </string-name>
          , Volume
          <volume>63</volume>
          (
          <year>2020</year>
          )
          <fpage>174</fpage>
          -
          <lpage>200</lpage>
          . doi.org/10.1016/j.socnet.
          <year>2020</year>
          .
          <volume>01</volume>
          .002
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          [2]
          <string-name>
            <given-names>M.</given-names>
            <surname>Campêlo</surname>
          </string-name>
          ,
          <string-name>
            <given-names>T.</given-names>
            <surname>Figueiredo</surname>
          </string-name>
          , &amp; A.
          <string-name>
            <surname>Silva</surname>
          </string-name>
          ,
          <article-title>The sociotechnical teams formation problem: a mathematical optimization approach</article-title>
          .
          <source>Ann Oper Res</source>
          <volume>286</volume>
          , (
          <year>2020</year>
          )
          <fpage>201</fpage>
          -
          <lpage>216</lpage>
          . https://doi.org/10.1007/s10479-018-2759-5.
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          [3]
          <string-name>
            <given-names>A.</given-names>
            <surname>Batyuk</surname>
          </string-name>
          ,
          <string-name>
            <given-names>V.</given-names>
            <surname>Voityshyn</surname>
          </string-name>
          ,
          <article-title>Apache storm based on topology for real-time processing of streaming data from social networks in</article-title>
          <source>Proceedings of the 2016 IEEE 1st International Conference on Data Stream Mining and Processing</source>
          ,
          <string-name>
            <surname>DSMP</surname>
          </string-name>
          <year>2016</year>
          ,
          <year>2016</year>
          , стр.
          <fpage>345</fpage>
          -
          <lpage>349</lpage>
          .
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>