<!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>Expert System for Modeling Threats and Protecting Premises from Information Leaks</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Marina Rudenko</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Evgenia Zhivago</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Andrei Rudenko</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>V.I. Vernadsky Crimean Federal University</institution>
          ,
          <addr-line>Prospekt Vernadskogo 4, Simferopol, 295007, Crimea</addr-line>
        </aff>
      </contrib-group>
      <fpage>151</fpage>
      <lpage>156</lpage>
      <abstract>
        <p>This paper considers the evolutionary path of developing physical ways of information leakage and methods of their minimization or possible elimination in the premises allocated for negotiations. The work aims to development of an expert system that interacts with the knowledge base, focused on modeling threats and means of protecting meeting rooms based on user-specified data and parameters. The features of the application, the principles of operation and the architecture of expert systems, a knowledge base developed for the intellectualization of the developed expert system are considered. The result of the work done is an expert system that implements the ability to simulate threats and protect premises from information leaks. The developed expert system can have practical application in organizations, the importance of information secrecy for which is one of the main tasks.</p>
      </abstract>
      <kwd-group>
        <kwd>1 Expert System</kwd>
        <kwd>Information Leakage Channels</kwd>
        <kwd>Knowledge Base</kwd>
        <kwd>Knowledge Representation Model</kwd>
        <kwd>Meeting Room</kwd>
        <kwd>Information Security</kwd>
        <kwd>Technical Leakage Channels</kwd>
      </kwd-group>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>1. Introduction</title>
      <p>In the realities of the modern world, information and its possession play a key role both in the ability
to be competitive and in the process of economic development in general.</p>
      <p>
        Information can be a product, service, raw material, and its correct application leads to various kinds
of material benefits for its owner. As a result, each owner of information strives to protect it from
dissemination and intruders. Often, such attackers are individuals or organizations interested in the
possibility of unauthorized gaining access to confidential information and intending or attempting to
organize such access [
        <xref ref-type="bibr" rid="ref1 ref3">1, 3</xref>
        ].
      </p>
      <p>
        Also, do not forget that some companies work with confidential information, including personal
data, the protection of which required by law [
        <xref ref-type="bibr" rid="ref4">4</xref>
        ][
        <xref ref-type="bibr" rid="ref5">5</xref>
        ]. The interests of the state in the information sphere
are to create conditions for the harmonious development of the Russian information infrastructure, for
the implementation of constitutional rights and freedoms of man and citizen in the field of obtaining
information and using it in order to ensure the inviolability of the constitutional system, sovereignty
and territorial integrity of Russia, political, economic and social stability, in the unconditional provision
of law and order, the development of equal and mutually beneficial international cooperation [
        <xref ref-type="bibr" rid="ref2">2</xref>
        ].
Therefore, behind the successful development and maintenance of the functioning of enterprises, there
is concern about ensuring information security in the field of business, entrepreneurship and production.
      </p>
      <p>
        The topic of protecting premises and meeting rooms in particular is relevant, and in this area,
developments are underway that will help to move from manual calculation of indicators to an
automated process [
        <xref ref-type="bibr" rid="ref6">6</xref>
        ]. A large base of theoretical knowledge accumulated with recommendations for
choosing and arranging a meeting room. However, in reality, it is not always possible to choose a room
that meets the criteria for the selection of premises, due to various reasons - constructive, requiring large
changes that may entail delays in the work of the company or the subjective capabilities of a manager
or a responsible person [
        <xref ref-type="bibr" rid="ref7">7</xref>
        ][
        <xref ref-type="bibr" rid="ref9">9</xref>
        ].
      </p>
    </sec>
    <sec id="sec-2">
      <title>2. Physical paths of information leakage</title>
      <p>Physical paths of information leakage according to state standard specification GOST R 50922-96,
information security threat is a set of conditions and factors that create a threat of information security
breach. All technical channels of information leakage can be divided into 4 groups are shown in the
Table 1.
● financial damage associated with the costs of restoring the company's information system, as
well as due to downtime caused by changes in the information security system;
● material and moral damage caused to information owners, whose information was stolen and,
as a result, damage to business reputation and business relationships was caused.</p>
    </sec>
    <sec id="sec-3">
      <title>3. Expert system design</title>
      <p>
        The most difficult and common are semi-structured or unstructured tasks associated with
decisionmaking, control and management tasks. Expert systems are aimed at solving such problems [
        <xref ref-type="bibr" rid="ref10">10</xref>
        ][
        <xref ref-type="bibr" rid="ref11">11</xref>
        ].
      </p>
      <p>Expert system - a computer system capable of partially replacing a specialist-expert in resolving a
problem situation.</p>
      <p>The use of expert systems can help achieve a number of positive aspects:
● exclude possible subjectivity;
● the ability to simplify the process by dividing the problem to be solved into smaller subtasks;
● there is no need to involve expensive experts and specialists;
● development of ES allows to accumulate human knowledge and experience;
● there is no need to study a large number of materials in order to make a decision or check the
correctness of the decision of the involved expert;
● the ability to minimize contacts with strangers, keep confidential data that is undesirable for
publicity;
● the solution can be obtained in a human-readable form.</p>
      <p>Also, the ES has mandatory components that form it:
● user interface;
● ES user - it can be both an end user who wants to get a solution, and an expert who introduces
new knowledge into the database, as well as a knowledge engineer who debugs the work of the
expert system;
● knowledge base editor - implements the ability to edit, delete, change and add data;
● expert - a specialist in the required field, able to find reasonable, correct and effective solutions
to the task;
● knowledge engineer - a person who has knowledge and skills in the field of computer science,
artificial intelligence and the construction of expert systems, who can systematize, properly present
knowledge and help the programmer in writing a software product;
● memory;
● inference mechanism - an inferred solution obtained during the operation of the expert system;
● system of explanations - a system of justifications, based on which this or that decision was
made.</p>
      <p>The basis of ES is the so-called knowledge base in a specific subject area. The data is constantly
entered and accumulated in the process of building and operating an expert system.</p>
    </sec>
    <sec id="sec-4">
      <title>4. Modeling threats and protecting premises from information leaks</title>
      <p>
        In the modern world, market relations are moving away from industrial ones, and are increasingly
becoming informational. Thus, information plays a key role in the development and functioning of
market participants [
        <xref ref-type="bibr" rid="ref12">12</xref>
        ].
      </p>
      <p>
        The reality is that a manager cannot always fully assess the existing threats of information leakage,
of which there are many [
        <xref ref-type="bibr" rid="ref13">13</xref>
        ].
      </p>
      <p>For example, consider a typical office is also used as a meeting room.</p>
      <p>Sometimes, the leader does not even realize that his office is essentially a sieve from which valuable
information flows without any special obstacles.</p>
      <p>Physical methods of protection against information leaks are understood as a complex of technical,
organizational and organizational-technical measures aimed at weakening or excluding the uncontrolled
exit of classified information outside the control zone.</p>
      <p>The following restrictions are imposed on protection methods:
● for protection, only funds certified and approved for circulation in the Russian Federation can
be used;
● methods and means untested, prohibited or related to the spyware category are unacceptable;
● protection against leaks must be based on legal methods of dealing with them;
● for the successful functioning of the protective complex of measures, it is also necessary that it
constitutes a single system.</p>
      <p>The complexes of protective measures are divided according to the principle of grouping of leakage
channels - methods of protection of visual, acoustic and vibration, electromagnetic and
materialmaterial channels of information leakage.</p>
    </sec>
    <sec id="sec-5">
      <title>5. Knowledge base of the expert system</title>
      <p>
        The basis of the expert system is the so-called knowledge base in a specific subject area. Тhe data is
constantly entered and accumulated in the process of building and operating the expert system. The
organization of knowledge and its accumulation are important properties of any expert systems[
        <xref ref-type="bibr" rid="ref13">13</xref>
        ][
        <xref ref-type="bibr" rid="ref15">15</xref>
        ].
      </p>
      <p>
        To solve the tasks assigned to the expert system, three databases needed to store information and
indicators, which will form the knowledge base [
        <xref ref-type="bibr" rid="ref14">14</xref>
        ].
      </p>
      <p>The representation of knowledge in expert systems is a fundamental concept that is a decisive aspect
in development. To represent knowledge, a production model and a formal-logical model of knowledge
representation chosen.</p>
      <p>The production refineries are based on the rule (products):</p>
      <p>IF &lt;condition&gt; THEN &lt;action&gt;. This rule consists of two parts: conditions - antecedent and actions
- consequent. In the form of a production model of knowledge representation, the knowledge necessary
for the system to display security recommendations is presented.</p>
      <p>Formally, the logical model of knowledge representation is similar to the production model. This is
partially true, but they make a huge difference. The difference is that the production model of
knowledge representation does not define any connections between the stored objects of the domain.</p>
    </sec>
    <sec id="sec-6">
      <title>6. Expert system for modeling threats and protecting premises from information leaks</title>
      <p>The initial data will be all the necessary information about the room that the user wants to allocate
for the meeting room, check and secure. The user enters information about the floor, the material of the
walls, about the rooms nearby, about the ceiling, decoration, about windows and doors, about the
ventilation, about the presence of radiation elements. The user also edits the scheme of his meeting
room in the program designer.</p>
      <p>As a result, the user will receive a scheme with color identification of the threat, recommendations
in text form on the necessary measures to improve security.</p>
      <p>The subtasks for solving the general problem are as follows:
● Processing of input data;
● Determination of the level of security and highlighting the element, depending on the level of
security;
● Offer security for low-security elements;</p>
      <p>Justification of decisions bases on the knowledge and experience of experts. The more experts
involved in the development of the knowledge base. The more unbiased decision can be issued by the
ES.</p>
      <p>It can be difficult for the user to collect data on the materials of the room, on the presence or absence
of certain elements that affect the security. Therefore, ES can give some error.</p>
      <p>The expert system for modeling threats and protecting premises from information leaks "Avarazh"
consists of a main form for entering data and describing the room considered by the user; forms for
editing the scheme of the room in question; forms for training the expert system, which is necessary for
experts to interact with the system; reference books of objects and materials and information about the
expert system itself.</p>
      <p>The program menu contains reference books for some objects, allowing you to more accurately
classify objects when entering data into the system. The "Training" menu item opens the form that
experts use to train the "Avarazh" expert system. The menu item "About the program" contains the
documentation on the basis of which the expert system works - Federal laws, GOSTs, SNiP and a set
of rules. It also contains an area where the results of processing the entered data are displayed. This area
consists of a visual sub-area, which shows a diagram with color identification of the security levels of
the elements of the tested room, detected leakage channels and recommendations for improving the
level of protection of the room. When you select the "Training" menu item, a window appears in front
of the expert, containing facts and fields for filling in the coefficients.</p>
      <p>To create a room scheme, you need to press the "Edit scheme" button of the main form, by clicking
the user enters the scheme editing mode on the new form.</p>
      <p>Using shape buttons, you can add elements to the room that affect the safety of the room as a whole.
Symbols have been introduced for some elements that may be intuitively incomprehensible. All added
items can be removed using the Clear All button. If the created scheme satisfies the user, then by
clicking the "Add to main window" button the resulting scheme is added to the main form for further
analysis. All elements of the room layout can be resized, moved, deleted or added again. Also, if one
element is varied, a tooltip appears.</p>
    </sec>
    <sec id="sec-7">
      <title>7. Conclusions</title>
      <p>On average, an expert needs about 6 hours to conduct a special survey of a room to measure
indicators when choosing 15 control points, and about 2 hours for calculations. Thanks to the use of the
"Avaraj" expert system, within 30 minutes, you cannot only receive recommendations that are
understandable to a nonprofessional, but also clearly see the hazard levels of the existing leakage
channels.</p>
      <p>In the course of the work, all the tasks were completed and the goal of the final qualification work
was achieved - the development of an expert system interacting with the knowledge base, focused on
modeling threats and means of protecting meeting rooms based on user-specified data and parameters.</p>
      <p>The novelty of this work lies in the development of an expert system in the field of threat modeling
and protection of meeting rooms, which not only generates recommendations for protection, but also
analyzes the existing leakage channels and has a graphical identification of the security level of room
objects and their ability to cause these leakage channels.</p>
      <p>This expert system has directions for further development and improvement, but already at this stage,
it has an extensive and sufficient toolkit to complete the task confirmed by its commissioning at UNIG
LLC.
8. References</p>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          [1]
          <string-name>
            <given-names>SEI</given-names>
            <surname>Cyber</surname>
          </string-name>
          <article-title>Minute: Insider Threats</article-title>
          . Available online: http://resources.sei.cum.edu/library/assetview.cfm?
          <source>assetid=496626 (accessed on 30 September</source>
          <year>2021</year>
          ).
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          [2]
          <string-name>
            <surname>Soomro</surname>
            ,
            <given-names>Z.A.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Shah</surname>
            ,
            <given-names>M.H.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Ahmed</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          <article-title>Information security management needs more holistic approach: A literature review</article-title>
          .
          <source>Int. J. Inf. Manag</source>
          .
          <year>2016</year>
          ,
          <volume>36</volume>
          ,
          <fpage>215</fpage>
          -
          <lpage>225</lpage>
          . https://doi.org/10.1016/j.ijinfomgt.
          <year>2015</year>
          .
          <volume>11</volume>
          .009
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          [3]
          <string-name>
            <surname>Al-Dhahri</surname>
            ,
            <given-names>S.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Al-Sarti</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Ahmed</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          <article-title>Information security management system</article-title>
          .
          <source>Int. J. Comp. Appl</source>
          .
          <year>2017</year>
          ,
          <volume>158</volume>
          ,
          <fpage>29</fpage>
          -
          <lpage>33</lpage>
          . DOI:
          <volume>10</volume>
          .5120/ijca2017912851
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          [4]
          <string-name>
            <surname>Dupuis</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Khadeer</surname>
            ,
            <given-names>S.</given-names>
          </string-name>
          <article-title>Curiosity killed the organization: A psychological comparison between malicious and non-malicious insiders and the insider threat</article-title>
          .
          <source>In Proceedings of the 5th Annual Conference on Research in Information Technology</source>
          , Boston, MA, USA, 28 September-1
          <source>October</source>
          <year>2016</year>
          . DOI:
          <volume>10</volume>
          .1145/2978178.2978185
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          [5]
          <string-name>
            <given-names>Insider</given-names>
            <surname>Threat Report</surname>
          </string-name>
          .
          <article-title>Insider threat report. Insider threat related data breach detection time</article-title>
          .
          <source>In Insider Threat Report: Executive Summary; Verizon business ready:</source>
          New York, USA,
          <year>2019</year>
          . https://www.verizon.com/business/resources/reports/insider-threat
          <source>-report/</source>
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          [6]
          <string-name>
            <surname>Alzhrani</surname>
            ,
            <given-names>K.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Rudd</surname>
            ,
            <given-names>E.M.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Boult</surname>
            ,
            <given-names>T.E.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Chow</surname>
            ,
            <given-names>C.E.</given-names>
          </string-name>
          <article-title>Automated big text security classification</article-title>
          .
          <source>In Proceedings of the 2016 IEEE Conference on Intelligence and Security Informatics (ISI)</source>
          , Tucson,
          <string-name>
            <surname>AZ</surname>
          </string-name>
          , USA,
          <fpage>28</fpage>
          -
          <issue>30</issue>
          <year>September 2016</year>
          .
          <source>Sustainability</source>
          <year>2020</year>
          ,
          <volume>12</volume>
          , 6217
          <volume>14</volume>
          of 14 https://www.researchgate.net/publication/309388110_Automated_Big_Text_Security_Classificat ion
        </mixed-citation>
      </ref>
      <ref id="ref7">
        <mixed-citation>
          [7]
          <string-name>
            <surname>Kim</surname>
            ,
            <given-names>K.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Kim</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          <article-title>A Study on analyzing risk scenarios about vulnerabilities of security monitoring system: Focused on information leakage by insider</article-title>
          .
          <source>In Proceedings of the International Workshop on Information Security Applications</source>
          , Jeju Island, Korea,
          <fpage>23</fpage>
          -
          <lpage>25</lpage>
          August 2018; Springer: Cham, Switzerland,
          <year>2018</year>
          . DOI:
          <volume>10</volume>
          .1007/978-3-
          <fpage>030</fpage>
          -17982-3_
          <fpage>13</fpage>
        </mixed-citation>
      </ref>
      <ref id="ref8">
        <mixed-citation>
          [8]
          <string-name>
            <surname>Shin</surname>
            ,
            <given-names>H.J.</given-names>
          </string-name>
          ; Kim,
          <string-name>
            <surname>M.H.</surname>
          </string-name>
          <article-title>A detection method of data leakage by cooperation of insiders</article-title>
          .
          <source>Int. J. Appl. Eng. Res</source>
          .
          <year>2017</year>
          ,
          <volume>12</volume>
          ,
          <fpage>13321</fpage>
          -
          <lpage>13327</lpage>
          .
          <string-name>
            <surname>Corpus</surname>
            <given-names>ID</given-names>
          </string-name>
          :
          <fpage>51808393</fpage>
        </mixed-citation>
      </ref>
      <ref id="ref9">
        <mixed-citation>
          [9]
          <string-name>
            <surname>Bromiley</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          <article-title>Defend Your Business against Insider Threats;</article-title>
          SANS Institute Information Security Reading Room, Sans Institute: Boston, MA, USA,
          <year>2019</year>
          . Corpus ID:
          <fpage>173171997</fpage>
        </mixed-citation>
      </ref>
      <ref id="ref10">
        <mixed-citation>
          [10]
          <string-name>
            <surname>Mandelli</surname>
            ,
            <given-names>D.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Yilmaz</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Aldemir</surname>
            ,
            <given-names>T.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Metzroth</surname>
            ,
            <given-names>K.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Denning</surname>
            ,
            <given-names>R.</given-names>
          </string-name>
          <article-title>Scenario clustering and dynamic probabilistic risk assessment</article-title>
          .
          <source>Reliab. Eng. Syst. Saf</source>
          .
          <year>2013</year>
          ,
          <volume>115</volume>
          ,
          <fpage>146</fpage>
          -
          <lpage>160</lpage>
          . https://dblp.org/rec/journals/ress/MandelliYAMD13.html
        </mixed-citation>
      </ref>
      <ref id="ref11">
        <mixed-citation>
          [11]
          <string-name>
            <surname>Ha</surname>
            ,
            <given-names>D.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Kang</surname>
            ,
            <given-names>K.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Ryu</surname>
            ,
            <given-names>Y.</given-names>
          </string-name>
          <article-title>Detecting Insider Threat based on Machine Learning: Anomaly Detection Using RNN Autoencoder</article-title>
          .
          <source>J. Korea Inst. Inf. Secur. Cryptogr</source>
          .
          <year>2017</year>
          ,
          <volume>27</volume>
          ,
          <fpage>763</fpage>
          -
          <lpage>773</lpage>
          . DOI:
          <volume>10</volume>
          .13089/JKIISC.
          <year>2017</year>
          .
          <volume>27</volume>
          .4.763,
          <string-name>
            <surname>Corpus</surname>
            <given-names>ID</given-names>
          </string-name>
          :
          <fpage>208108705</fpage>
        </mixed-citation>
      </ref>
      <ref id="ref12">
        <mixed-citation>
          [12]
          <string-name>
            <surname>Lee</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Kim</surname>
            ,
            <given-names>I.</given-names>
          </string-name>
          <article-title>Detecting Abnormalities in Fraud Detection System through the Analysis of Insider Security Threats</article-title>
          .
          <source>J. Soc. E Bus. Stud</source>
          .
          <year>2019</year>
          ,
          <volume>23</volume>
          ,
          <fpage>153</fpage>
          -
          <lpage>169</lpage>
          . DOI:
          <volume>10</volume>
          .7838/JSEBS.
          <year>2018</year>
          .
          <volume>23</volume>
          .4.153,
          <string-name>
            <surname>Corpus</surname>
            <given-names>ID</given-names>
          </string-name>
          :
          <fpage>199710981</fpage>
        </mixed-citation>
      </ref>
      <ref id="ref13">
        <mixed-citation>
          [13]
          <string-name>
            <surname>Bokova</surname>
            <given-names>O.I.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Drovnikova</surname>
            <given-names>I.G.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Etepnev</surname>
            <given-names>A.S.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Rogozin</surname>
            <given-names>E.A.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Khvostov</surname>
            <given-names>V.A.</given-names>
          </string-name>
          <article-title>Methods of Estimating Reliability of Information Security Systems which Protect from Unauthorized Access in Automated Systems</article-title>
          . DOI:
          <volume>10</volume>
          .15622/sp.
          <year>2019</year>
          .
          <volume>18</volume>
          .6.
          <fpage>1301</fpage>
          -
          <lpage>1332</lpage>
        </mixed-citation>
      </ref>
      <ref id="ref14">
        <mixed-citation>
          [14]
          <string-name>
            <surname>Kim</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Oh</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Ryu</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Lee</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Kwon</surname>
            ,
            <given-names>K.</given-names>
          </string-name>
          ;
          <string-name>
            <surname>Lee</surname>
            ,
            <given-names>K.</given-names>
          </string-name>
          <article-title>SoK: Research on Behavior-Based Data Leakage Incidents for the Sustainable Growth of an Organization</article-title>
          .
          <source>Sustainability</source>
          <year>2020</year>
          ,
          <volume>12</volume>
          (
          <issue>15</issue>
          ),
          <volume>6217</volume>
          ; https://doi.org/10.3390/su12156217
        </mixed-citation>
      </ref>
      <ref id="ref15">
        <mixed-citation>
          [15]
          <string-name>
            <surname>Oh</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          ; Kim,
          <string-name>
            <given-names>T.</given-names>
            ;
            <surname>Lee</surname>
          </string-name>
          ,
          <string-name>
            <surname>K.</surname>
          </string-name>
          <article-title>Advanced insider threat detection model to apply periodic work atmosphere</article-title>
          .
          <source>TIIS</source>
          <year>2019</year>
          ,
          <volume>13</volume>
          ,
          <fpage>1722</fpage>
          -
          <lpage>1737</lpage>
          . DOI:
          <volume>10</volume>
          .3837/tiis.
          <year>2019</year>
          .
          <volume>03</volume>
          .035
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>