Reviewing topics of COVID-19 news articles: case study of CNN and China daily
| DOI | https://doi.org/10.1108/AJIM-05-2022-0264 |
| Published date | 29 August 2022 |
| Date | 29 August 2022 |
| Pages | 407-429 |
| Subject Matter | Library & information science,Information behaviour & retrieval,Information & knowledge management,Information management & governance,Information management |
| Author | Yue Yuan,Kan Liu,Yanli Wang |
Reviewing topics of COVID-19
news articles: case study of CNN
and China daily
Yue Yuan
Department of Information Management, Peking University, Beijing, China, and
Kan Liu and Yanli Wang
Zhongnan University of Economics and Law, Wuhan, China
Abstract
Purpose –The purpose of thisstudy is to analyze the topics of COVID-19 news articles for better obtaining the
relationship among and the evolution of news topics, helping to manage the infodemic from a quantified
perspective.
Design/methodology/approach –To analyze COVID-19 news articles explicitly, this paper proposes a
prism architecture. Based on epidemic-related news on China Daily and CNN, this paper identifies the topics of
the two news agencies, elucidates the relationship between and amongst these topics, tracks topic changes as
the epidemic progresses and presents the results visually and compellingly.
Findings –The analysis results show that CNN has a more concentrated distribution of topics than China
Daily, with the former focusing on government-related information, and the latter on medical. Besides, the
pandemic has had a big impact on CNN and China Daily’s reporting preference. The evolution analysis of news
topics indicates that the dynamic changes of topics have a strong relationship with the pandemic process.
Originality/value –This paper offers novel perspectives to review the topics of COVID-19 news articles and
provide new understandings of news articles during the initial outbreak. The analysis results expand the scope
of infodemic-related studies.
Keywords COVID-19, News articles, Prism architecture, Topic analysis, Data visualization
Paper type Research paper
1. Introduction
The outbreak of COVID-19 constitutes an unprecedented public health crisis and has been
declared a global pandemic. Nearly 365 million confirmed cases –and more than five million
deaths –have been reported as of January 28, 2022 [1]. In addition, the pandemic has had a
devastating effect across the societal spectr um: health services; education; tourism;
agriculture; and many other sectors have been adversely affected [2], which in turn has
elicited profound governmental and public concern.
With the ever-increasing coronavirus infectors, a parallel tsunami of information strikes
the world –individuals and news agencies generate a vast volume of accurate and inaccurate
information –forming a so-called “infodemic”(Kim et al., 2020;Gruzd et al., 2021). Similar to
the epidemic, it spreads between humans by news media or social media, causing
psychological problems and risk-taking behaviors that harm mental and physical health.
Therefore, the infodemic has demonstrated equal destruction as the pandemic itself, which
raised significant concern from the World Health Organization (WHO). As Dr. Tedros, the
Director-General of WHO, said, “We are not just fighting an epidemic; we are fighting an
infodemic”(Zarocostas, 2020).
Unfortunately, an infodemic cannot be eliminated but can only be managed
(Tangcharoensathien et al., 2020). Therefore, many researchers dedicate their strength to
manage the infodemic. Most studies on infodemic focus on the topic of misinformation.
Compared with the true information, the damage caused by misinformation is indeed more
significant (Tran et al., 2019). In addition, misinformation is often spread between individuals
Reviewing
topics of
COVID-19
news articles
407
The current issue and full text archive of this journal is available on Emerald Insight at:
https://www.emerald.com/insight/2050-3806.htm
Received 20 May 2022
Revised 10 July 2022
Accepted 28 July 2022
Aslib Journal of Information
Management
Vol. 75 No. 2, 2023
pp. 407-429
© Emerald Publishing Limited
2050-3806
DOI 10.1108/AJIM-05-2022-0264
throughsocial media. Therefore,many researchers focuson misinformation,especially rumors
spread on social media platforms (Chong et al., 2021;Vragaet al., 2020;Janmohamed et al., 2021).
However,the definition of infodemicis “the proliferation of accurateinformationand inaccurate
information during the pandemic”(Gruzd et al., 2021), which means we should be concerned
with both misinformation and true information to manage infodemic. Ideas submitted to the
WHO have suggested that the quantified analysis of digital data (e.g. online newsarticles) is
considered one of the effective ways to manage infodemic (Tangcharoensathien et al., 2020;
Dhawan et al., 2021). Becausethe flooding ofnews articles will alsolead to severe issues,such as
confusing people’s perception of the pandemic (Dhakal, 2018;Yan and Bissell, 2018;Agarwal
and Toshniwal, 2020;Roby et al., 2018;Ophir, 2018,2019), which will lead to the in ability of the
official to implement effectiverisk communication and disaster response (McBride et al., 2020;
Jamieson and Van Belle, 2019;Jeong and Lee, 2019), thus providinga fertile breeding groundfor
thespread of misinformation.The analysisof news articles willreveal the internaltrends hidden
behind the massive text content, such as topics or emotions, which helps to understand and
managethe infodemic from differentoutlooks. Hence,analysis of the news articlesfrom various
perspectives is necessary tomanage infodemic.
Here, “news articles”refer to products of editorial agenda that editors of news companies
decide what the mass could know or could not know. This layer of concern distinguishes
news articles from academic articles. Hence, one should approach and analyze news articles,
or the discourse formed by multiple articles from the news agencies, with caution. However,
as a typical type of unstructured data, news articles have high analytical value. The topic
extraction on COVID-19-related news articles can be used to obtain critical clues from the
large volume of news articles during the initial outbreak of the pandemic. Furthermore, the
relationship among and the evolution of news topics can be analyzed quantitively, helping to
manage the infodemic from a quantified perspective.
The topic relationships and topic evolution can be elucidated if news articles are analyzed
comprehensively and continuously (del Gobbo et al., 2021). To review news articles in this
manner, the words used in different news parts, including the title, keywords and content, can
be utilized to extract topics (L
azaro-Rodr
ıguez, 2020;Abd-Alrazaq et al., 2020). By mining the
relationship and evolution of news topics, this paper attempts to address the following
questions:
(1) What are the topics of COVID-19-related news articles?
(2) What is the relation among the topics of COVID-19-related news articles?
(3) What is evolution track of topics of COVID-19-related news articles?
To deal with the three questions, this paper builds a visual topic analysis architecture that
extracts and displays the topic relationship and evolution route of COVID-19-related news
articles. More specifically, we extract news topics and visualize their relationships, which
provide objective, reproducible and verifiable interpretations.
The rest of the manuscript goes as follows. The next section, Section 2, presents the
literature review of analysis of news articles and other contents related to COVID-19. Section
3elaborates the prism architecture proposed by this study. Section 4 details the experiment
results and analysis. Finally, Section 5 makes discussion and conclusion of the major findings
and implications.
2. Literature review
2.1 Analysis of various contents related to COVID-19
Due to the rapid spread of the pandemic, analysis on various types of content (e.g. academic
literature, news reports, social media content, etc.) that are related to COVID-19 have drawn
AJIM
75,2
408
To continue reading
Request your trial