Communications and Networks: Covid-19 on Social Media

Maria Pilgun

Russian State Social University, st. W. Pieck, Moscow, RF

Keywords: social media, neural network approach, user-generated content, actors, digital platforms, perception.

Abstract: The paper deals with the analysis of the reaction of society, of the perception by various groups of social

media actors of the beginning of Сovid 19 in the Russian-language media landscape. Neural network texts

analysis of content dedicated to the beginning of the pandemic and study of digital footprints of actors

showed that the audience was divided into two groups - metropolitan and regional. Moreover, the perception

of the pandemic, the spread of infection, the assessment of methods to combat the disease, the consequences,

etc., differed significantly in both groups of actors.

1 INTRODUCTION

The impact of Covid-19 on various aspects of society

has already been covered in numerous scientific

studies. researchers pay special attention artificial

intelligence technologies which were actively used in

the fight against the spread of infection and in the

search for effective treatment protocols. In particular,

scientific studies have been published that address:

the use of artificial intelligence to solve the Covid-19

problems, including forecasting, decision-making to

support health care, etc. (Santosh et al., 2021); artificial

intelligence to deal with the Covid-19 consequences

(Hassanien et al., 2020); the use of intelligent systems

to stop the spread of the pandemic (Joshi et al., 2020);

neural network technologies for the fight against the

coronavirus (Fong et al., 2021); technological solutions

to stop the Covid-19outbreak and minimize the risk

(Khosla et al., 2021); data mining in the fight against

the Covid-19outbreak (Niranjanamurthy et al., 2020);

epidemic forecasting models, surveillance and

tracking systems (Raza, 2021). In addition, the use of

nanotechnology to prevent the spread of infection is

explored in (Devasena, 2021). Diseases associated

with coronavirus and their consequences caused

significant changes in perception, required new

research on how people in an extraordinary situation

perceive and evaluate their reactions, react to the

environment and interact with each other (WHO, 2020;

Horesh, 2020).

The aim of the study: the analysis of reactions and

concerns of social media actors during the start of of

the pandemic in Russia, the perception of social

media actors of the consequences of the spread of the

coronavirus.

2 METHODS

To analyze the content was applied neural network

texts analysis, as well as sentiment analysis and

analysis of word association. Digital footprints were

explored using content analysis. Neural network

technology Text Analyst 2.3 was used as a tool. A

detailed description of the methodology is presented

in (Kharlamov & Pilgun, 2020).

The study design is presented in the flowchart:

Pilgun, M.

Communications and Networks: Covid-19 on Social Media.

DOI: 10.5220/0011865700003612

In Proceedings of the 3rd International Symposium on Automation, Information and Computing (ISAIC 2022), pages 44-49

ISBN: 978-989-758-622-4; ISSN: 2975-9463

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

The procedures are detailed in (Kharlamov et al.,

2021; Pilgun et al., 2022).

2.1 Data

The dataset of the study was social media data (social

networks, blogs, videos, podcasts, forums, instant

messengers, reviews) dedicated to the discussion of

Covid-19. The material was collected between March

1, 2020 and March 30, 2020.

In Russia, the spread of coronavirus infection

began in the capital; in the regions, the pandemic

began later. In Russia, the spread of coronavirus

infection began in the capital; in the regions, the

pandemic began later. Analysis of the semantic

network made it possible to identify the most

significant assessments and opinions of actors, as

well as allowed the content was divided into two

groups. The first group consisted of actors with

geolocation from Moscow and the Moscow region,

the second actors with geolocation from other

regions of Russia, as well as Russian-speaking actors

from other countries (Table 1; Figure 1, 2). In addition

to various regions of Russia, the geolocation of regional

actors covers the following countries: Ukraine, Belarus,

USA, Israel, Kazakhstan, Germany, Great Britain,

Switzerland, Azerbaijan, Moldova, Czech Republic,

Fradnce, Australia, Spain, Italy, China, Latvia, Netherlands,

Canada, Georgia, Uzbekistan, Estonia, Lithuania, Armenia,

Poland, Greece, Cuba, Sweden, Bulgaria, Serbia, Tajikistan,

Thailand, Kyrgyzstan, Brazil, Kyrgyzstan, Uruguay, Japan,

India, Cyprus, Angola, Vietnam, Fiji, Austria, Albania,

Lithuania, Mongolia, Denmark, Egypt, Ireland, Nigeria,

Argentina, South Korea, Belgium, Iran, Luxembourg,

Mexico, Nepal, etc. (Figure 2).

Table 1. Quantitative data

Regional

actors

Data

Moscow actors

Data

Number of

messages

157 006

Number of

messages

4 535

Max.

number of

messages

per day

10376

Max. number of

messages per

day

274

Number of

authors

47 186

Number of

authors

388

Activity

(posts per

author)

3,33

Activity (posts

per author)

11,69

Figure 1. Geolocation of the Moscow

active actors.

Figure 2. Geolocation of the regional active

actors.

Communications and Networks: Covid-19 on Social Media

3 RESULTS

In the content of the regional actors, the peak of

generated messages falls on March 27, 2020 (the total

number of messages is 10 376), although the

dynamics of views has 2 peaks: on March 24, 2020

(6 448 541) and March 28, 2020 (6 362 129). The

dynamics of the regional actors’ activity

demonstrates a gradual increase in interest in the

spread of the coronavirus infection, the peak of which

also falls on March 27, 2020 (5 471). These peaks

depend on the official information on the distribution

of Covid 19 in Russia. (Figure 3, 4 and 5). As for the

Moscow actors, the largest number of messages is

observed on March 25, 2020 (245) and on March 27,

2020 (274), the dynamics of views had peaks value

on March 16, 2020 (1 090 578) and on March 18,

2020 (1520161); the dynamics of actors’ activity

begins to grow from March 12, 2020 (68) and reaches

a peak already on March 23, 2020 (93). The reason

for the fluctuation of peaks determined by the

messages of the mayor of Moscow S. Sobyanin about

the situation with the spread of Covid 19 in the capital

(Figures 6, 7 and 8).

Figure 3. Dynamics of the total number of messages

and unique messages of the regional actors.

Figure 4. Dynamics of views of the regional actors.

Figure 5. Dynamics of the regional actors’ activity.

Figure 6. Dynamics of the total number of messages

and unique messages of the Moscow actors.

Figure 7. Dyamics of views of the Moscow actors.

Figure 8. Dynamics of the Moscow actors’ activity

ISAIC 2022 - International Symposium on Automation, Information and Computing

In the database of both groups, a neutral cluster

prevails; meanwhile, in the Moscow content, the

negative cluster accounts for 18,24 %, and in the

regional content - 5,04 %. The positive cluster is

represented by insignificant data, practically absent.

(Figures 9, 10). The sentiment analysis of digital

footprints confirms a higher degree of negative

reactions in the Moscow group compared with the

regional group (Figures 11, 12).

Figure 9. Tonality of

the regional actors’

messages.

Figure 10. Tonality

of the Moscow

actors’ messages.

Figure 11. Tonality

of the regional

actors’ footprints.

Figure 12. Tonality

of the Moscow

actors’ footprints.

The topic structure of the zonal content dedicated

to Covid 19 is exclusively related to the problems of

the infection attack in Moscow and then in the

Moscow region and further in Russia, as well as the

discussion of ways to prevent the spread of a new

disease (Figure 13).

In the content of Moscow' actors, along with

health problems, a significant portion is taken by

political issues, criticism of the authorities’ actions,

both in relation to the organization of the fight against

the c Covid 19, and in relation to other issues, for

example, amendments to the Constitution (Figure

14).

Figure 13. Topic structure of the regional

actors’ content.

Figure 14. Topic structure of the Moscow actors’

content.

The most interesting are the results of the analysis of

the core of the semantic network, which make it

possible to reveal the hidden assessments and

opinions of the actors (Figures 15 and 16).

Figure 15. The semantic network of the content of

province actors.

Communications and Networks: Covid-19 on Social Media

Figure 16. The semantic network of the content of

Moscow actors.

3 CONCLUSIONS

Analysis of the data on the beginning of the pandemic

in Russia showed that the perception of Covid-

19differs significantly in the Moscow and regional

group of actors. The impact of the pandemic has

mostly affected the inhabitants of the metropolis. The

discussion of the coronavirus topic in the Russian-

speaking media space was begun by the Moscow

actors; their intensity in the generation and the

negative connotation of the content prevail over the

regional data. The regional content is devoted to the

problems of the infection spread in Moscow and then

in the Moscow region and further in Russia, as well

as the organization of life in the new conditions, the

consequences of the pandemic: unemployment in the

first instance. The Moscow actors focused primarily

on the problems of the pandemic in the capital, and

also paid great attention to the rise of the coronavirus

in China, Europe and the USA. The coronavirus

topics in the Moscow content are closely related to

political and economic issues.

REFERENCES

[1] Santosh K C, Joshi A (Eds.) 2021 COVID-19:

Prediction, Decision-Making, and its Impacts

(Springer Singapore, Singapore)

[2] Hassanien, A, Dey, N, Elghamrawy, S M: 2020 (Eds.)

Big Data Analytics and Artificial Intelligence Against

COVID-19: Innovation Vision and Approach

(Springer International Publishing, Cham)

[3] Joshi A, Dey N, Santosh K C (Eds.) 2020 Intelligent

Systems and Methods to Combat Covid-19

(Springer Singapore, Singapore)

[4] Fong S J, Dey N, Chaki, J 2021 Artificial

Intelligence for Coronavirus Outbreak (Springer

Singapore, Singapore)

[5] Khosla P K, Mittal M, Sharma D, Goyal L M (Eds.)

2021 Predictive and Preventive Measures for

Covid-19 Pandemic (Springer Singapore,

Singapore)

[6] Niranjanamurthy M B, Siddhartha K N (Eds.) 2021

Intelligent Data Analysis for COVID-19 Pandemic

(Springer Singapore, Singapore)

[7] Raza Kh (Ed.) 2021 Computational Intelligence

Methods in COVID-19: Surveillance, Prevention,

Prediction and Diagnosis (Springer Singapore,

Singapore

[8] Devasena T 2021 Nanotechnology-COVID-

19Interface (Springer Singapore, Singapore)

[9] WHO 2020 COVID-19 situation reports– www who

int

[10] Horesh D, Brown A D 2020 Traumatic stress in the

age of COVOID-19: A call to close critical gaps and

adapt to new reality Psychological Trauma: Theory,

Research, Practice, and Policy 12(4) 23-36

[11] Kharlamov A; Pilgun M (Eds) 2020

Neuroinformatics and Semantic Representations

Theory and Applications (Cambridge Scholars

Publishing, Newcastle upon Tyne)

[12] Kharlamov A A, Raskhodchikov A N, Pilgun M 2021

Smart City Data Sensing during COVID-19: Public

Reaction to Accelerating Digital Transformation

Sensors 21 12 3965

[13] Pilgun M, Raskhodchikov A N, Koreneva Antonova

O 2022 Effects of COVID-19 on Multilingual

Communication Frontiers in Psychology 12.

ISAIC 2022 - International Symposium on Automation, Information and Computing

Communications and Networks: Covid-19 on Social Media