Topical evolution patterns and temporal trends of microblogs on public health emergencies. An exploratory study of Ebola on Twitter and Weibo
| DOI | https://doi.org/10.1108/OIR-04-2016-0100 |
| Date | 08 October 2018 |
| Published date | 08 October 2018 |
| Pages | 821-846 |
| Author | Lu An,Chuanming Yu,Xia Lin,Tingyao Du,Liqin Zhou,Gang Li |
| Subject Matter | Library & information science,Information behaviour & retrieval,Collection building & management,Bibliometrics,Databases,Information & knowledge management,Information & communications technology,Internet,Records management & preservation,Document management |
Topical evolution patterns and
temporal trends of microblogs on
public health emergencies
An exploratory study of Ebola on Twitter
and Weibo
Lu An
Center for Studies of Information Resources, Wuhan University, Wuhan, China and
School of Information Management, Wuhan University, Wuhan, China
Chuanming Yu
School of Information and Safety Engineering,
Zhongnan University of Economics and Law, Wuhan, China
Xia Lin
College of Computing and Informatics, Drexel University,
Philadelphia, Pennsylvania, USA, and
Tingyao Du, Liqin Zhou and Gang Li
School of Information Management, Wuhan University, Wuhan, China
Abstract
Purpose –The purpose of this paper is to identify salient topic categories and outline their evolution
patterns and temporal trends in microblogs on a public health emergency across different stages.
Comparisons were also examined to reveal the similarities and differences between those patterns and trends
on microblog platforms of different languages and from different nations.
Design/methodology/approach –A total of 459,266 microblog entries about the Ebola outbreak in West
Africa in 2014on Twitter and Weibo were collectedfor nine months after the inceptionof the outbreak. Topics
were detected by the latent Dirichlet allocation modeland classified into several categories. The daily tweets
were analyzedwith the self-organizingmap technique and labeledwith the most salient topics.The investigated
time span was dividedinto three stages, and the most salient topic categories were identified for each stage.
Findings –In total, 14 salient topic categories were identified in microblogs about the Ebola outbreak and
were summarized as increasing, decreasing, fluctuating or ephemeral types. The topical evolution patterns of
microblogs and temporal trends for topic categories vary on different microblog platforms. Twitter users
were keen on the dynamics of the Ebola outbreak, such as status description, secondary events and so forth,
while Weibo users focused on background knowledge of Ebola and precautions.
Originality/value –This study revealed evolution patterns and temporal trends of microblog topics on a
public health emergency. The findings can help administrators of public health emergencies and microblog
communities work together to better satisfy information needs and physical demands by the public when
public health emergencies are in progress.
Keywords Twitter,Microblog, Weibo, Public health emergency, Temporal trends, Topicalevolution pattern
Paper type Research paper
Introduction
Public health emergencies are important types of events that attracthigh attention. As public
health emergencies usually occur unexpectedly and deteriorate rapidly, emergency
management departments need to quickly and fully grasp the development status of
emergencies, predict potential problems and take effective measures. Currently, microblog Online Information Review
Vol. 42 No. 6, 2018
pp. 821-846
© Emerald PublishingLimited
1468-4527
DOI 10.1108/OIR-04-2016-0100
Received 5 April 2016
Revised 9 January 2017
10 October 2017
6 November 2017
Accepted 14 November 2017
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/1468-4527.htm
This work was supported by the Major Project of the Ministry of Education of China (Grant
No. 17JZD034) and the National Natural Science Foundation of China (Grant No. 71603189, 71790612).
821
Microblogs on
public health
emergencies
platforms, such as Twitter and Weibo, are popular channels for obtaining and generating a
great deal of information and comments on various events, such as public health events.
According to Reuter and Spielhofer (2017), 75 percent of citizens used or plan to use social
media to shareinformation duringan emergency, and 77 percent of citizensbelieve that social
media provides emergencyinformation faster than traditional media channels. Microblogs are
an important type of social media and provide a wealth of clues to public emergency
management. In September 2013, Twitter Alert, an emergency alert system was released, and
the American Red Cross, Federal Emergency Management Agency, and global non-profits
such as the World Health Organization have participated in this system (Mashable, 2013). The
topical analysis of microblogs on public health emergencies can reveal the development track
of emergencies and, thus,facilitate understanding thepublic’s concerns about the emergencies.
As a microblog platform often has a fixed range of users, e.g., from the same or similar
countries or of the same language, the analysis of microblogs on one platform may produce
limited findings. A comparison between two different microblog platforms may reveal different
topical evolution patterns or temporal trends for the same emergency and help emergency
management departments obtain insights and identify problems that cannot be found upon the
analysis of one microblog platform.
Since February 2014, a severe Ebola outbreak has raged in West Africa and has led to
the death of more than 11,000 people as of March 16, 2016 (World Health Organization,
2016). Hundreds of thousands of tweets on the Ebola outbreak have been generated on
microblog platforms since then. Tens of thousands of tweets and internet searches have
been generated upon each news video of Ebola (Towers et al., 2015). To explore temporal
characteristics of topic categories in related microblogs, we took the Ebola outbreak in West
Africa since February 2014 as the investigation case.
The purpose of this study was, first, to detect salient topic categories in microblogs on a
major public health event at different stages, second, to outline the evolutionary pattern and
temporal trends of topic categories in the related microblogs, and third, to reveal and
compare usage patterns of Chinese and English microblog platforms when a major public
health event unfolds. The findings can be used to understand temporal characteristics of
topic categories in microblogs of public health emergencies and usage differences between
Chinese and English microblog platforms. Accordingly, public health emergency
administrators can provide informational and physical support to mitigate the
consequences of infectious diseases in similar cases, such as the current Zika virus.
Related research
Topical analysis of microblogs
Common methods oftopical analysis of microblogs include statistical techniques and manual
classification. Studies on microblogs involving general topics, e.g., Yun (2012), were usually
based on fairly rough topic classifications. More fine-grained topic categories were identified
in the topical analysis of microblogs focusing on a specific topic, such as classifying main
topics of public concern toward Ebola into several categories (Lazard et al., 2015). However,
manual classification of topics limited the number of investigated microblogs.
As major events usually trigger a large number of tweets, researchers generally sample a
small percentage of related tweets (Qu et al., 2011). To automatically analyze large-scale
microblog entries, topic modeling is an indispensable step. Latent Dirichlet allocation (LDA)
and term frequencyvectors are common approaches fortopic modeling. Topics of microblogs
were oftenstudied in a static manner, andthe evolution of topics was notsufficiently explored.
Temporal analysis of microblogs
To reveal the temporal characteristics of microblogs, some novel topic models were
developed, such as trend sensitive-latent Dirichlet allocation (TS-LDA) (Yang and Rim, 2014)
822
OIR
42,6
To continue reading
Request your trial