<!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>Communicative Behavior and Speech Perception of Social Media Actors: A Neural Network Approach</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Maria Pilgun</string-name>
          <email>mpilgun@iling-ran.ru</email>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Moscow</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Russia</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="editor">
          <string-name>Perception, Social Media, Neural Network Technology</string-name>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>Institute of Linguistics, Russian Academy of Sciences</institution>
          ,
          <addr-line>1 bld. 1 Bolshoi Kislovsky lane, app. 47, 49, 125009</addr-line>
        </aff>
      </contrib-group>
      <fpage>0000</fpage>
      <lpage>0002</lpage>
      <abstract>
        <p>The paper concerns the analysis of the residents' attitude towards the construction of the chord ring in Moscow, as well as the identification of digital conflictogenic zones and predictive analytics. The material for the study was network data collected during the period from 1.07.2019 00:00 to 31.12.2019 23:59. To interpret the data, neural network text analysis, analysis of lexical associations, content analysis and sentiment analysis were used. As a result, digital conflictogenic zones were identified; users' perception of each facility was characterized; the risks of protests in the virtual and real space during the project implementation were assessed.</p>
      </abstract>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>1. Introduction</title>
      <p>The new empirical reality with a multifold increase in the amount of information
ensured the dominance of social media, changed the assessment and approaches to data
processing. The importance of semantic analysis for large volumes of verbal material has
increased, since the focus of attention has shifted from the ways of collecting data to data
processing. Advancements in automatic processing of natural language data are increasingly
dependent on the use of artificial neural networks. Thus, Microsoft’s</p>
      <sec id="sec-1-1">
        <title>MT-DNN (Multi-Task</title>
        <p>
          Deep Neural Network) is a natural language processing (NLP) model that outperforms Google
BERT in nine of eleven benchmark NLP tasks. Google’s new Multilingual T5 model trained
on a new generic Crawl-based dataset for 101 languages also marks a new milestone in the
NLP development [
          <xref ref-type="bibr" rid="ref1">1</xref>
          ]. GPT3 (OpenIA), an autoregressive language model with 175 billion
parameters, achieves strong performance on many
        </p>
      </sec>
      <sec id="sec-1-2">
        <title>NLP datasets, including translation,</title>
        <p>
          question-answering, and cloze tasks, as well as several tasks that require on-the-fly reasoning
or domain adaptation, such as unscrambling words, using a novel word in a sentence, or
performing 3-digit arithmetic [
          <xref ref-type="bibr" rid="ref2">2</xref>
          ].
        </p>
        <p>2020 Copyright for this paper by its authors.</p>
      </sec>
    </sec>
    <sec id="sec-2">
      <title>Related work</title>
      <p>
        Speech perception in its multilevel processing system has a rich tradition of research [
        <xref ref-type="bibr" rid="ref3 ref4 ref5 ref6 ref7">3-7</xref>
        ]. During
the pandemic, the problems of transformation of perception in a virtual environment [
        <xref ref-type="bibr" rid="ref8">8</xref>
        ], perception of
stress contrasts in semantic and non-semantic contexts, etc. came to the fore.
      </p>
      <p>
        The study of the conflict has also acquired particular importance in recent years. This led to the
development of a wide range of computational methods used to study the dynamics of conflicts, a
variety of computational approaches from text mining and machine learning to agent-based modeling
and analysis of social networks [
        <xref ref-type="bibr" rid="ref9">9</xref>
        ], and analysis of new personality constructs based on speech data
of social networks [10].
      </p>
      <p>Moreover, the digital environment makes it possible to automatically receive large amounts of data
in real time and to implement interdisciplinary projects [11-13].</p>
      <p>The purpose of this study is to analyze the attitude of Moscow residents towards the construction
of the chord ring using digital data, as well as to identify digital conflictgenic zones and to conduct
predictive analytics.
1.2.</p>
    </sec>
    <sec id="sec-3">
      <title>Data</title>
      <p>The material for the study was the data of social media, microblogs, blogs, instant messengers,
videos, forums and reviews dedicated to the road construction in Moscow (the South-East Chord
(SEC); the Southern Beltroad (reconstruction of Verkhnie Polya str. on the segment (section) from St.
Maryinsky Park to the Moscow Ring Road), North-East Chord (NEC), North-West Chord (NWC)).
Date of collection: 01.07.2019 00:00 - 31.12.2019 23:59 (Figure. 1, Table 1).
∑ % =</p>
      <p>,</p>
      <p>The study involved a transdisciplinary approach; also neural network text analysis and analysis of
lexical associations, content analysis and sentiment analysis were used.</p>
      <p>The data was collected using Brand Analytics algorithms. Content collecting and filtering included
bot identification and filtering, identification of posts, comments, reposts (with and without add-ons).
Also, digital traces (likes, views) were identified and analyzed.</p>
      <p>Content clustering was performed by sentiment, presence and degree of aggression.</p>
      <p>The sentiment analysis was performed using the Eureka Engine sentiment module. The technique
is based on a statistical algorithm for conditional random fields (CRF) using sentiment dictionaries.
Sequences of lexemes are used as input data, after which the algorithm calculates the probabilities of
possible sequences of tags and chooses the most probable one.</p>
      <p>The degree of aggression was identified on the basis of the analysis of lexical tags of aggression
determined expertly. The set   of lexical tags that make up the lexical mask  = {  },  = 1. .  , were
ranked automatically within the analyzed text corpus by determining their semantic weight   in this
corpus. The summation of the ranks of lexical tags weighted by the degree of their significance   in
this domain area assigned by the expert, allows calculating the integral degree of aggression present in
the corpus texts:  = ∑     [14].</p>
      <p>To correctly mark the attitude towards the construction of each facility by Muscovites, indices of
stress and well-being were derived. The algorithm for obtaining these indexes consisted of extracting
tag lists from the database of users’ comments and reposts. One list was formed using the neural
network technology, the other – from the results of the work of 3-5 experts. For the social stress
index, markers were identified from the negative cluster, for the social well-being index – from the
positive and neutral clusters. After these procedures, the resulting lists were combined into a single
dictionary with the average weights of the matching markers. Next, the neural network technology
was adjusted, taking into account the resulting dictionary, and the indices of social stress and
wellbeing were derived according to the following formula:</p>
      <p>∑ 
∑    
where R and B are current, and   and   are the maximum values of ranks and weights (in
%), respectively [14].</p>
      <p>Within this study, digital aggression was considered as an indicator of social conflict [15; 16]. And
social well-being was interpreted according to the WHO Wellness Index-5 (Psychiatric Research
Unit, WHO Collaborating Center for Mental Health, Frederiksborg General Hospital, DK-3400
Hillerød).</p>
      <p>The content analysis was performed in accordance with [17; 18] using the AutoMap text mining
tool.</p>
      <p>For visual analytics, the Tableu platform was used.</p>
      <p>Using a model based on the neural network paradigm of using neural-like elements with temporal
summation of signals (the so-called corticomorphic associative memory), the topic structure of the
database and the results of summarization were analyzed, which made it possible to extract explicit
information. On the other hand, such a network representation of the text made it possible to identify
and interpret the semantic network as a set of interrelated concepts. With the help of the semantic
network, implicatures and semantic accents were analyzed that were most important for actors, which
helped obtain their rating [14].</p>
      <p>An important stage in the analysis of the actors’ attitude was the construction of associative
networks. The Word Association (WA) paradigm analysis has a long and multifaceted history
[1921]. With the help of implicit association tests (IAT), it helped study implicit social cognition,
subconscious motivations, attitudes towards the presented stimulus, as well as automatic associations
for subjects who prefer not to demonstrate at a conscious level (see, for example, Project Implicit
(https://www.projectimplicit.net/). The potential of associations in the analysis of various types of
network data has also already shown its effectiveness [22-24]. In this study, the analysis of lexical
associations was performed on the material of the clusters identified in the course of the sentiment
analysis and analysis of network aggression using neural network technologies [14], which made it
possible to draw conclusions about the preferential perception (attitude) of the users, to identify the
most frequent associations that characterize the actors’ attitude towards the subject of analysis and to
confirm the results of predictive analytics.</p>
      <p>The analysis of network aggression, the sentiment analysis and the analysis of lexical associations
made it possible to identify and analyze the specifics and dynamics of processes that form digital
conflictogenic zones.</p>
    </sec>
    <sec id="sec-4">
      <title>2. Results and discussion</title>
    </sec>
    <sec id="sec-5">
      <title>2.1. Digital platforms, types of sources and messages</title>
      <p>Analysis of the content distribution across digital platforms makes it possible to determine user
preferences and features of communication processes associated with the analyzed facilities. So, to
discuss the SEC issues and to generate the relative content, the users used the following top resources:
VKontakte (98 608 989), Odnoklassniki (38 089 303), YouTube (23 775 381), Facebook (23 098 359)
and Twitter (17 994 117); the Beltroad issues were covered mainly on YouTube (24 135 808), as well
as on the following network platforms: VKontakte (21 995 278), Instagram (8 728 919), Twitter (6
292 593) and Facebook (4 719 839). VKontakte was the leader in posting the NEC and NWC content
(NEC – 10 620 841; NWC – 3 578 719), but the following positions were distributed in different
ways: YouTube (5 706 195), Facebook (3 460 758), Instagram (3 369 284), Odnoklassniki
(2 024 790) for NEC issues; and Odnoklassniki (1 521 277), Facebook (1 493 690), Instagram
(820 960), YouTube (724 602) for NWC issues (Figure 2).</p>
      <p>SEC
NEC</p>
      <sec id="sec-5-1">
        <title>Beltroad NWC</title>
        <p>Analysis of the content distribution by types of sources and types of messages suggests that
the maximum number of messages devoted to the implementation of the SEC and NEC projects were
comments on social networks, which indicates a high degree of interest of the actors and active
discussion with the involvement of a wide range of users. Social media posts dominate in the content
regarding the Beltroad and NWC (Figure 3).</p>
      </sec>
    </sec>
    <sec id="sec-6">
      <title>2.2. Key topics 2.2.1. SEC</title>
      <p>After the selection of the user-generated content from the consolidated database, the analysis of
the semantic network and the topic structure was performed, as well as the content analysis. The
results led to the conclusion that the situation around the construction of the South-East Chord is
extremely tense. The core of the semantic network composed of nominations with a link weight of
98100, clearly shows the negative attitude of the users towards the implementation of the project (Figure
4).</p>
      <p>In a negative cluster, environmental issues come to the fore, which cause serious concerns of the
actors:</p>
      <p>• The danger of an environmental catastrophe caused by the SEC construction, since it will lead
to the destruction of the radioactive repository of the polymer plant:</p>
      <p>(...) There is a real threat of radioactive contamination due to the SEC construction on the
radioactive waste repository of the TVEL plant (...)
• An increase in the incidence of desease among citizens:
(...) Cubic meters of polluting particles from the roads will cover Moscow and its inhabitants and
increase the number of cancers and the incidence (...)</p>
      <p>• The environmental problems caused by the SEC construction, according to the actors, are
mentioned in the all-Russian environmental protest:</p>
      <p>(...) Showing civil solidarity with all regions of Russia where there is a peaceful resistance against
environmental genocide of indigenous people and local residents (...)</p>
      <p>• Among environmental issues, garbage clusters are of utmost importance, as well as the
garbage reform:</p>
      <p>(…) Environmental problems have become more acute than ever before in the country’s history.
Landfills are catastrophically overflown, the garbage reform has failed, hazardous incineration
plants will be built in the Moscow region and Tatarstan, despite the protests of local residents! (...)
• Deforestation of green spaces:
(...) Residents are worried about the ecological situation, since after the felling of trees it will
certainly worsen (...)</p>
      <p>Socio-political problems are also identified in the context:
• Readiness for active protest actions:
(...) It is better to be active today than radioactive tomorrow. Protest activity is increasing in
Moscow (...)
• Strong rejection by the actors of the officials’ actions:
(...) it is very difficult to resist this (...) it is against us (...) officials who have gathered from all
over Russia because in Moscow there is always the biggest money, as you know, and everyone wants
to grab a piece of it, we are the only city activists who resist (…)</p>
      <p>• Problems of the SEC construction are combined with the problems of renovation,
construction of New Moscow and road construction, which, according to the actors, are performed
only to enrich officials:</p>
      <p>(...) under the Yuri Mikhalych Luzhkov period, they built over those areas that were free, but when
Sergei Semyonovich Sobyanin came, the first project was this new Moscow, which is being actively
built up now, and now, in general, the construction complex needs a new territory of Moscow, well, it
is quite difficult to find the territory, therefore we have such a program as the renovation (...)
• Users make accusations of lobbying for laws that violate work methods:
(...) Residents of several districts of Moscow are concerned by the plans of city officials by all
means to build a bridge (part of the SEC) across the Moskva River connecting the
MoskvorechyeSaburovo district and Pechatniki! (...)
• There is a certain consolidation of civil society against construction:
(...) it is very nice to see here the defenders of the Losiny Ostrov, Butovo, Pechatniki, only
together, only civil society, and we can resist, we cannot only unite to show them that it is not them
who are the masters of the city, but us (...)</p>
      <p>• Accusing officials of using dishonest methods to achieve their goals, pitting Muscovites of
different social strata against each other:</p>
      <p>(...) I want to officially declare, know that the residents of Moscow will pit us against each other,
we will be simply pitted against each other, this is a great grief, but why do they do it? because they
cannot admit that they are turning a social project into a commercial one and they are looking for the
guilty and we will be those guilty (...)</p>
      <p>• The belief that the implementation of the project will lead to worsen transport accessibility
and a sharp decline in the living standards of Muscovites:</p>
      <p>(…) In the long term we get increased incidence of disease, traffic congestion and deterioration of
the urban environment. Motorists, come to your senses, the Chord will not solve your problems, but
will only aggravate them, reduce the average life expectancy of many districts of Moscow (...).</p>
      <p>Professional issues:
• Expertise becomes an important factor in the confrontation between activists and the state
structures:</p>
      <p>(…) The main argument of the Moscow officials: the burial ground does exist, but there are no
radioactive burials on the section where the Chord will pass, they are located 100-150 meters to the
west. However, this claim is denied by Greenpeace. Experts put a map of the future chord on this
territory and, taking soil samples exactly where the flyover supports will stand, found six sources of
pollution with a maximum level of 200 micro-roentgens per hour (a level of up to 50 micro-roentgen /
hour is considered safe). (...).</p>
      <p>• The actors have trust only in negative information about the project:
(...) The construction of the bridge and the flyover as part of the projects of the South-East Chord
and the Southern Beltroad of eight lanes in the immediate vicinity of JSC TVEL may provoke soil
sliding, while radiation pollution, together with dust particles, can begin to spread along the
surrounding roads in air and throughout Moscow with cars (...)</p>
      <p>The positive and neutral cluster is made up of official materials that should highlight the
advantages of the project, the benefits for residents and the improvement of the transport situation:
(...) The North-West Chord, the Southern Beltroad, the North-East Chord and the South-East
Chord practically create the Fourth Transport Ring in its other edition (…).</p>
      <p>• Improvement of transport accessibility, development of the city transport system:
(...) As part of the SEC construction, multi-level transport interchanges will be built through the
roads of the Kursk, Kazan and Gorkovsky directions of the Moscow Railway (MZhD), as well as on
the section from Graivoronovskaya Street to Polbina Street and on the section from Polbina Street to
Kuryanovsky Boulevard (...)</p>
    </sec>
    <sec id="sec-7">
      <title>2.2.2. Southern Beltroad</title>
      <p>The content analysis indicates a tense situation around the construction of the beltroad. The level
of aggression and social stress is quite high, intergroup and intragroup communications are
characterized by a high degree of conflict potential.</p>
      <p>Ecological issues:
• Environmental problems of the Southern Beltroad construction is closely related to the SEC.
The actors put forward accusations of deliberately causing harm to the residents’ health and of
sabotage that will lead to an environmental disaster:</p>
      <p>(...) problems with the radioactive landfill near the Moscow Polymetals Plant and the current
situation with the construction of the bridge within the Southern Beltroad and South-East Chord
projects (...)</p>
      <p>Socio-political problems:
• Protest against the construction of the Southern Beltroad:
(…) We, residents of the city of Moscow, demand to stop the projects of the Southern Beltroad
and the South-Eastern Chord&amp;#33 (…).
•</p>
      <p>Actors’ readiness for active actions, meetings:
(...) On Sunday, a rally will be held in Moscow against the construction of the South-East Chord
and the Southern Beltroad. We start a fundraising campaign for notification of the rally and printing
posters (...)
• Allegations of violation of regulations and laws:
(...) We demand to check the SEC and SB construction project regarding the
PechatnikiKantemirovskaya street – Bekhtereva for compliance with the environmental legislation of the
Russian Federation, urban planning, and sanitary and hygienic standards (...)
• Allegations of falsification during public hearings:
(...) check for violation of the procedure, forgery and falsification of the results during public
hearings on the SEC and SB construction in the Tsaritsyno and Moskvorechye-Saburovo districts, the
legality of the amendment to the Moscow City Planning Code without holding hearings on this issue
(...).</p>
      <p>• Appeals from residents and attempts to solve problems through a legitimate discussion with
government agencies have not been effectively resolved:</p>
      <p>(...) Residents of Moscow apply to law enforcement agencies regarding violation of the law, but
the result is either ping-pong between the RF IC and the Prosecutor’s Office, but it is also possible
that the appeal, where illegal actions of the executive power are described, may end up in the
executive power itself, for example, in Moskomarkhitektura (...).</p>
      <p>• Destruction of the historical architectural heritage of the city and personal property of
citizens:</p>
      <p>(...) well, a part of the kuskovo was cut down there and only the fact that we lived for six months
in tents not letting the park be cut down, only thanks to this they limited themselves to only 12
hectares; this is a historical monument park, kuskovo, the temple built in 1644 is being destroyed due
to the fact that this chord passes close to it; the garages were demolished (…).</p>
      <p>• Deterioration of the residents’ living standards of as a result of the construction:
(...) means that residents of Kantemirovskaya, Bekhtereva, Kaspiyskaya streets and Saburovo
district can say goodbye to peace and comfort under their windows (...).</p>
      <p>Professional and legal issues:
• Lack of professional expertise:
(...) Please show documents confirming the conduct of examinations and surveys confirming the
absence of radioactive waste and excess radiation background in the area and on the territory of the
construction along the main course of the South-East Chord along the territory of the Moscow
Polymetal Plant and the railway station Moskvorechye (...).</p>
      <p>• Accusation of receiving false expert examinations:
(...) SNIPs are violated; for this, one must be put in jail (...) the expert is criminally responsible
(...)</p>
      <p>• Alternative conclusions with negative project characteristics are taken as objective and
truthful. The presence or absence of professional evidence, sources, and the status of the author of the
conclusion casts a small shadow in this case:</p>
      <p>(…) In the area of the South-East Chord construction, the radiation level is 4 times exceeded –
ecologists (…).</p>
      <p>• Requirements to prosecute design and construction organizations that violate standards and
legislation:</p>
      <p>(...) on the construction of (...) the Southern Beltroad and the South-East Chord passing through
the land plot on which the disposal of radioactive waste of JSC “Moscow Polymetal Plant” is located
without obtaining a building permit, without a warrant for construction and digging works, without
any state environmental and radiation expertise (...)</p>
      <p>The positive and neutral cluster is made up of official materials containing positive
assessments of the project, information on the progress of construction, the benefits that residents will
obtain, and the improvement in the transport situation:</p>
      <p>(...) Recently, another section of the Southern Beltroad has opened – from Balaklavsky to
Proletarsky Avenue (...).
2.2.3. NEC
2.2.4. NWC</p>
      <p>The key topics of the content represent the official versions of the construction and are related to
similar projects; they represent the North-West Chord as part of a new transport project (Figure 6):
(...) The new transport framework that will have been built in Moscow by 2023, will be based on
the North-East, South-East and North-West Chords, as well as the Southern Beltroad (…).</p>
      <p>The negative connotation in the users’ perception is reflected in a small number of messages and is
associated with the following claims:
• Deterioration of the ecological situation:
(...) How will people live in hotspot construction areas? how will they breathe? The Department of
Transport and Development of Road and Transport Infrastructure in Moscow does not seem to care
(...)
• Decrease of the living standards of local residents:
(...) In Moscow, the North-West Chord is now being built through a residential area – right under
the windows of buildings, with trees cut down (...)
• Deterioration of transport accessibility within the area:
(...) Oddly enough: it was possible to drive 80 km per hour along Kalininsky Prospekt in 1973, but
along the NORTH-WEST CHORD in 2017 there are no underground passages and ONLY 60 KM
PER HOUR is allowed (...)
• Integration with the problems of construction of similar facilities, the SEC first of all:
(...) will lead to the possibility of radioactive waste contamination of a large territory of Moscow,
where several hundred thousand residents live (...)
2.3.</p>
    </sec>
    <sec id="sec-8">
      <title>Associative network</title>
      <p>The users’ perception is reflected most expressly in the data obtained using associative search
and analysis of lexical associations. The stimuli were nominations of facilities and nominations with
the maximum link weight in the sematic core of the corresponding data set (Figures 7 - 14).
2.3.1. SEC
(...) and here is the problem with garbage, this is a new stalingrad, here is a new Chernobyl;
until now Chernobyl has not happened, but if the chord is built through this unowned waste
disposal site – and this is not just an exercise in semantics, – this is indeed a radiation
hazard, all of us here can be exposed to this; what to do with these 60,000 tons, that’s a good
question (...)</p>
    </sec>
    <sec id="sec-9">
      <title>2.3.2. Beltroad</title>
      <p>Contexts with responses:</p>
      <p>• (...) represent the district of Moskvorechye-Saburovo and Tsaritsyno (...) the areas densely
populated, with many children, they were green before the urban development plans for our districts
started a year ago, and we all gathered to talk about these problems that are hanging over us; this is
radiation; these are 2 highways – the southern beltroad, the south-east chord and renovation that will
massively affect the Tsaritsyno area and directly affect the Moskvorechye-Saburovo area (...)
• (...) I will speak as a resident of this area, I have lived here for a very long time and I still
remember it green, we had a lot of trees, our ecology is bad, but they protected us from the road and
from cars and from any negativity. Our building 41 is the first on the street and is surrounded by the
railway, Kashirskoye Highway, a sump plus a couple of other burial grounds; it is very scary, since a
lot of people in our building have died from cancer recently (...)
2.3.3. NEC
Contexts with responses:
• (...) They want to cut down thousands of trees in Losiny Ostrov for the sake of new roads,
shopping centers and residential complexes. Part of the national park in the north-east of Moscow is
planned to be cut down for the construction of the eight-lane back-up of Shchelkovskoye highway, the
six-lane North-East Chord, the Fairytale Forest residential complex and a shopping center (...)
• (...) There were a number of useful speeches from our deputies: in addition to my speech on
the chord, Yevgeny Stupin spoke out on the garbage cluster and attempts to ban the rally, Helen
Yanchuk – on fraudulent electronic voting, Viktor Maksximov – on deprivation of paternal rights for
participation in rallies, Oleg Sheremetev – on the problems of the North-East Chord and related
issues, Sergey Savostyanov – (...)
2.3.4. NWC</p>
    </sec>
    <sec id="sec-10">
      <title>Digital conflictogenic zones</title>
      <p>Digital conflictogenic zones were identified taking into account the dynamics of the actors’
activity, messages, analysis of the negative cluster and the level of network aggression. The largest
conflictogenic zone was identified in the data set dedicated to the SEC construction, with a high level
of aggression and social stress approaching a critical point, the achievement of which could have
irreversible consequences. It should be noted that the content with the absence of social tension
consists mainly of official messages, and not only they do not arouse much interest or any positive
reaction from the users, but, on the contrary, they receive abhorrence. At the same time, the
usergenerated content is distinguished by negative evaluativeness and aggression (Figures 15 - 16).
2.4.1. Social stress and well-being indices</p>
      <p>The quantitative expression of the residents’ attitude towards the implementation of urban
development projects was evaluated using the social stress and well-being indices. The social stress
and well-being indices were calculated by the ratio of the weights of markers and ranks of notions in
the corresponding content. The measurements were taken using the database of comments and shares
(reposts) dedicated to the corresponding facility. It should be noted that active actors generate a large
amount of negative content dedicated to the SEC construction and participate in online and offline
protest activities. A fairly high Social Approval Index arises from official information and messages
from actors affiliated with stakeholders (city authorities, construction organizations). However, the
content with positive sentiment evaluation not only does not convince opponents of the projects, but,
on the contrary, causes more bitterness and deepens the conflict (Figure 17).</p>
      <p>Social well-being index</p>
      <p>Social stress index
North-West Chord</p>
      <p>North-East Chord
Southern beltroad</p>
      <p>South-East Chord
0,4
14,1
15,6
19,5
18,9
19,7
0
5
10
15
20
25
30
35
40
45</p>
    </sec>
    <sec id="sec-11">
      <title>3. Conclusion</title>
      <p>The analysis of the residents’ attitude towards the implementation of the Moscow chord ring
project based on network data, as well as the study of digital conflictogenic zones and the predictive
analytics results led to the following conclusions:</p>
      <p>• The actors’ attitude towards the implementation of the South-East Chord project is extremely
negative, characterized by the presence of vast digital conflictogenic zones, a high level of social
stress and network aggression. Further implementation of the project will be accompanied by active
protests in virtual and real spaces. Only force majeure circumstances, such as a pandemic, can make
significant adjustments.</p>
      <p>• The facilities of the Southern beltroad (reconstruction of Verkhniye Polya street on the
section from Maryinsky Park street to the Moscow Ring Road) and the North-East Chord are
evaluated with an average level of negative reaction, are distinguished by the presence of digital
conflictogenic zones and characterized by an average level of social stress and network aggression.
Meanwhile, it can be predicted that protest activity will develop mainly within the virtual
communications.</p>
      <p>• The content dedicated to the North-West Chord reflects a high degree of the residents’
positive perception, does not contain digital conflictogeniс zones or signs of social stress and network
aggression.</p>
      <p>It is important to note the synergy of problems associated with the construction of similar
facilities: in the users’ perception, construction facilities causing conflicts with residents are
combined with the problems of other urban planning projects in Moscow and the Moscow region, as
well as with the renovation issue. The integrated nature of the users’ claims, the transfer of conflict
situations from the construction of similar facilities in Moscow and the Moscow region to the
“allRussian” context (for example, the Shiyes problems, garbage clusters) emphasizes the growing
importance of environmental issues for Russian society.</p>
    </sec>
    <sec id="sec-12">
      <title>4. References</title>
      <p>[10] V. Kulkarni, M. L. Kern, D. Stillwell, M. Kosinski, S. Matz, L. Ungar, S. Skiena, H. A.</p>
      <p>Schwartz. Latent human traits in the language of social media: An open-vocabulary approach,
PLoS ONE. November 28, 13, 11 (2018). doi:10.1371/journal.pone.0201703.
[11] A. Gonçalves, A. Social Media Analytics Strategy, Apress, New York, NY, 2017.
[12] G. Meiselwitz (Ed.), Social Computing and Social Media. Communication and Social</p>
      <p>Communities. Cham, Springer International Publishing, 2019.
[13] Yu. Shibuya. Social Media Communication Data for Recovery, Singapore, Springer Singapore,
2020.
[14] A. Kharlamov, M. Pilgun (Eds), Neuroinformatics and Semantic Representations. Theory and</p>
      <p>Applications, Cambridge Scholars Publishing, Newcastle upon Tyne, 2020.
[15] A. M. Keynesianism, Social Conflict and Political Economy, Palgrave Macmillan UK,</p>
      <p>London, 2020.
[16] C. M. Lemos, Agent-Based Modeling of Social Conflict. From Mechanisms to Complex</p>
      <p>Behavior. Springer International Publishing, Cham (2018).
[17] M. D. White, E. Marsh. Content analysis: a flexible methodology, Library trends 1(55) 2006:
22-45.
[18] K. Krippendorff. Content Analysis. An Introduction to Its Methodology, 3nd. ed., Sage</p>
      <p>Publications, Inc., Los Angeles, 2012.
[19] D. L. Nelson, C. Nelson, C. Mcevoy, and T. A. Schreiber. The University of South Florida free
association, rhyme, and word fragment norms, Behavior Research Methods, Instruments, &amp;
Computers 36 2004: 402–407.
[20] P. J. Brooks, L. Siger-Gardner, K. Sailor. Contrasting effects of associates and coordinates in
children with and without language impairment: A picture–word interference study, Applied
Psycholinguistics 35(3) 2014: 515-545. doi:10.1017/S0142716412000495.
[21] L. Vivas, L. Manoiloff, A. M. García, F. Lizarralde. J. Vivas. Core Semantic Links or Lexical
Associations: Assessing the Nature of Responses in Word, J Psycholinguist Res February 48(1)
2019: 243–256. doi:10.1007/s10936-018-9601-8.
[22] K. Houben, R. W. Wiers. Measuring implicit alcohol associations via the Internet: Validation of
Web-based implicit association tests, Behavior Research Methods 40 2008: 1134–1143. doi:
10.3758/BRM.40.4.1134.
[23] P. Bedi, C. Sharma. Community detection in social networks, Wiley Interdisciplinary Reviews:</p>
      <p>Data Mining and Knowledge Discovery. 6 2016:115–135. doi:10.1002/widm.1178.
[24] B. File, Z. Keczer, A. Vancsó, B. Bőthe, I. Tóth-Király, M. Hunyadi, A. Ujhelyi, I. Ulbert, J.</p>
      <p>Góth, G. Orosz. Emergence of polarized opinions from free association networks, Behavior
Research Methods 51 2019: 280–294. doi:10.3758/s13428-018-1090-z.</p>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          [1]
          <string-name>
            <given-names>L.</given-names>
            <surname>Xue</surname>
          </string-name>
          ,
          <string-name>
            <given-names>N.</given-names>
            <surname>Constant</surname>
          </string-name>
          ,
          <string-name>
            <given-names>A.</given-names>
            <surname>Roberts</surname>
          </string-name>
          ,
          <string-name>
            <given-names>M.</given-names>
            <surname>Kale</surname>
          </string-name>
          ,
          <string-name>
            <given-names>R.</given-names>
            <surname>Al-Rfou</surname>
          </string-name>
          ,
          <string-name>
            <given-names>A.</given-names>
            <surname>Siddhant</surname>
          </string-name>
          ,
          <string-name>
            <given-names>A.</given-names>
            <surname>Barua</surname>
          </string-name>
          ,
          <string-name>
            <given-names>C.</given-names>
            <surname>Raffel</surname>
          </string-name>
          . mT5:
          <article-title>A massively multilingual pre-trained text-to-text transformer</article-title>
          ,
          <source>arXiv:2010.11934v1 [cs.CL] 22.10</source>
          (
          <year>2020</year>
          ):
          <fpage>1</fpage>
          -
          <lpage>13</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          [2]
          <string-name>
            <surname>T. B. Brown</surname>
            ,
            <given-names>B.</given-names>
          </string-name>
          <string-name>
            <surname>Mann</surname>
            ,
            <given-names>N.</given-names>
          </string-name>
          <string-name>
            <surname>Ryder</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          <string-name>
            <surname>Subbiah</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          <string-name>
            <surname>Kaplan</surname>
            ,
            <given-names>P.</given-names>
          </string-name>
          <string-name>
            <surname>Dhariwal</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          <string-name>
            <surname>Neelakantan</surname>
            ,
            <given-names>P.</given-names>
          </string-name>
          <string-name>
            <surname>Shyam</surname>
            ,
            <given-names>G.</given-names>
          </string-name>
          <string-name>
            <surname>Sastry</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          <string-name>
            <surname>Askell</surname>
            ,
            <given-names>S.</given-names>
          </string-name>
          <string-name>
            <surname>Agarwal</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          <string-name>
            <surname>Herbert-Voss</surname>
            , G. Krueger,
            <given-names>T.</given-names>
          </string-name>
          <string-name>
            <surname>Henighan</surname>
            ,
            <given-names>R.</given-names>
          </string-name>
          <string-name>
            <surname>Child</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          <string-name>
            <surname>Ramesh</surname>
            ,
            <given-names>D.M.</given-names>
          </string-name>
          <string-name>
            <surname>Ziegler</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          <string-name>
            <surname>Wu</surname>
            ,
            <given-names>C.</given-names>
          </string-name>
          <string-name>
            <surname>Winter</surname>
            , Ch. Hesse,
            <given-names>M.</given-names>
          </string-name>
          <string-name>
            <surname>Chen</surname>
            , E. Sigler,
            <given-names>M.</given-names>
          </string-name>
          <string-name>
            <surname>Litwin</surname>
            ,
            <given-names>S.</given-names>
          </string-name>
          <string-name>
            <surname>Gray</surname>
            ,
            <given-names>B.</given-names>
          </string-name>
          <string-name>
            <surname>Chess</surname>
            ,
            <given-names>J.</given-names>
          </string-name>
          <string-name>
            <surname>Clark</surname>
            , Ch. Berner,
            <given-names>S.</given-names>
          </string-name>
          <string-name>
            <surname>McCandlish</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          <string-name>
            <surname>Radford</surname>
            ,
            <given-names>I.</given-names>
          </string-name>
          <string-name>
            <surname>Sutskever</surname>
            ,
            <given-names>D.</given-names>
          </string-name>
          <string-name>
            <surname>Amodei</surname>
          </string-name>
          ,
          <article-title>Language Models are Few-Shot Learners</article-title>
          .
          <source>arXiv:2005.14165v2 [cs.CL] 1.06</source>
          (
          <year>2020</year>
          ):
          <fpage>1</fpage>
          -
          <lpage>74</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          [3]
          <string-name>
            <given-names>K.</given-names>
            <surname>Demuth</surname>
          </string-name>
          . Perception, production, and individual differences,
          <source>Applied Psycholinguistics</source>
          <volume>39</volume>
          (
          <issue>4</issue>
          ) (
          <year>2018</year>
          ):
          <fpage>735</fpage>
          -
          <lpage>741</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          [4]
          <string-name>
            <given-names>T. G.</given-names>
            <surname>Ransom</surname>
          </string-name>
          ,
          <string-name>
            <given-names>R.</given-names>
            <surname>Dale</surname>
          </string-name>
          ,
          <string-name>
            <given-names>R. J.</given-names>
            <surname>Kreuz</surname>
          </string-name>
          ,
          <string-name>
            <given-names>D.</given-names>
            <surname>Tollefsen</surname>
          </string-name>
          .
          <article-title>How Do Different Types of Alignment Affect Perceived Entity Status?</article-title>
          ,
          <source>J Psycholinguist Res</source>
          <volume>48</volume>
          (
          <year>2019</year>
          ):
          <fpage>961</fpage>
          -
          <lpage>985</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          [5]
          <string-name>
            <given-names>M.</given-names>
            <surname>Judith</surname>
          </string-name>
          ,
          <string-name>
            <given-names>J.</given-names>
            <surname>Bryant</surname>
          </string-name>
          .
          <article-title>Blue is in the Eye of the Beholder: A Cross-Linguistic Study on Color Perception and Memory</article-title>
          ,
          <source>J Psycholinguist Res</source>
          <volume>48</volume>
          (
          <year>2019</year>
          ):
          <fpage>163</fpage>
          -
          <lpage>179</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          [6]
          <string-name>
            <given-names>N.</given-names>
            <surname>Kreyßig</surname>
          </string-name>
          ,
          <string-name>
            <given-names>A.E.</given-names>
            <surname>Krautz</surname>
          </string-name>
          .
          <article-title>Lying and perception of lies by bilingual speakers</article-title>
          ,
          <source>Applied Psycholinguistics</source>
          <volume>40</volume>
          (
          <issue>5</issue>
          ) (
          <year>2019</year>
          ):
          <fpage>1313</fpage>
          -
          <lpage>1329</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref7">
        <mixed-citation>
          [7]
          <string-name>
            <given-names>P. P.</given-names>
            <surname>Krishna</surname>
          </string-name>
          ,
          <string-name>
            <given-names>S.</given-names>
            <surname>Arulmozi</surname>
          </string-name>
          ,
          <string-name>
            <given-names>M.</given-names>
            <surname>Sh</surname>
          </string-name>
          .
          <string-name>
            <surname>Ram</surname>
            ,
            <given-names>R. K.</given-names>
          </string-name>
          <string-name>
            <surname>Mishra</surname>
          </string-name>
          .
          <article-title>Sensory Perception in Blind Bilinguals and Monolinguals</article-title>
          ,
          <source>J Psycholinguist Res</source>
          <volume>49</volume>
          (
          <year>2020</year>
          ):
          <fpage>631</fpage>
          -
          <lpage>639</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref8">
        <mixed-citation>
          [8]
          <string-name>
            <given-names>B. A</given-names>
            ,
            <surname>Essam</surname>
          </string-name>
          ,
          <string-name>
            <given-names>M. S.</given-names>
            <surname>Abdo</surname>
          </string-name>
          .
          <article-title>How Do Arab Tweeters Perceive the COVID-19 Pandemic?</article-title>
          ,
          <string-name>
            <surname>J Psycholinguist Res</surname>
          </string-name>
          (
          <year>2020</year>
          ):
          <fpage>1</fpage>
          -
          <lpage>15</lpage>
          . doi:
          <volume>10</volume>
          .1007/s10936-020-09715-6.
        </mixed-citation>
      </ref>
      <ref id="ref9">
        <mixed-citation>
          [9]
          <string-name>
            <given-names>E.</given-names>
            <surname>Deutschmann</surname>
          </string-name>
          ,
          <string-name>
            <given-names>J.</given-names>
            <surname>Lorenz</surname>
          </string-name>
          ,
          <string-name>
            <given-names>L. G.</given-names>
            <surname>Nardin</surname>
          </string-name>
          ,
          <string-name>
            <given-names>D.</given-names>
            <surname>Natalini</surname>
          </string-name>
          ,
          <article-title>and</article-title>
          <string-name>
            <given-names>A.F.X.</given-names>
            <surname>Wilhelm</surname>
          </string-name>
          (Eds.),
          <source>Computational Conflict Research</source>
          , Springer, Heidelberg,
          <year>2020</year>
          .
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>