A bibliometric analysis of event detection in social media
| Pages | 29-52 |
| DOI | https://doi.org/10.1108/OIR-03-2018-0068 |
| Published date | 11 February 2019 |
| Date | 11 February 2019 |
| Author | Xieling Chen,Shan Wang,Yong Tang,Tianyong Hao |
| 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 |
A bibliometric analysis of event
detection in social media
Xieling Chen
Jinan University, Guangzhou, China
Shan Wang
University of Macau, Macao, China, and
Yong Tang and Tianyong Hao
South China Normal University, Guangzhou, China
Abstract
Purpose –The purpose of this paper is to explore the research status and development trend of the field of
event detection in social media (ED in SM) through a bibliometric analysis of academic publications.
Design/methodology/approach –First, publication distributions are analyzed including the trends of
publications and citations, subject distribution, predominant journals, affiliations, authors, etc. Second, an
indicator of collaboration degree is used to measure scientific connective relations from different perspectives.
A network analysis method is then applied to reveal scientific collaboration relations. Furthermore, based on
keyword co-occurrence analysis, major research themes and their evolutions throughout time span are
discovered. Finally, a network analysis method is applied to visualize the analysis results.
Findings –The area of ED in SM has received increasing attention and interest in academiawith Computer
Science and Engineering as two major research subjects. The USA andChina contribute the most to the area
development.Affiliations and authorstend to collaboratemore with those within the samecountry. Among the
14 identifiedresearch themes,newly emerged themes suchas Pharmacovigilanceevent detection are discovered.
Originality/value –This study is the first to comprehensively illustrate the research status of ED in SM by
conducting a bibliometric analysis. Up-to-date findings are reported, which can help relevant researchers
understand the research trend, seek scientific collaborators and optimize research topic choices.
Keywords Network analysis, Social media, Bibliometric analysis, Event detection
Paper type Research paper
Introduction
Social media are forms of electronic communication (such as websites for social networking
and microblogging) through which users create online communities to share information,
ideas, personal messages and other contents, such as videos (www.merriam-webster.com/
dictionary/social%20media). The ubiquity of social media is increasing rapidly due to the
spread of internet and the development of mobile devices. With the continuous growth of
social networks and the active use of social media services, overwhelming amount of
user-generated social media data is available as participants continuously interact with each
other (Weiler et al., 2016). The explosive increasing of the data has promoted the
development of social media data mining for supporting better decision making (Wang et al.,
2017), thus it draws much attention from both academia and industry ( Jiang et al., 2017).
Meanwhile, social med ia data processing bri ngs a lot of research opp ortunities
(Nurwidyantoro and Winarko, 2013), among which, event detection has been one of the
most important research topics (Dong et al., 2015) since social media data carry abundant
hidden occurrences of real-time events (Kaleel and Abhari, 2015).
Event detection is defined as identifying the first story on the topics of interest through
constantly monitoring news streams (Dou et al., 2012). The streaming nature of social media
with no limitation about space and time makes it a valuable source for event detection
(Nurwidyantoro and Winarko, 2013). Specially, an event in social media has been defined by
Online Information Review
Vol. 43 No. 1, 2019
pp. 29-52
© Emerald PublishingLimited
1468-4527
DOI 10.1108/OIR-03-2018-0068
Received 5 March 2018
Revised 26 June 2018
Accepted 11 September 2018
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/1468-4527.htm
This paper forms part of a special section “Social media mining for journalism”.
29
Event
detection in
social media
Dou et al. (2012) as “an occurrence causing change in the volume of text data that discusses
the associated topic at a specific time. This occurrence is characterized by topic and time,
and often associated with entities such as people and location.”Thus, event detection in
social media (ED in SM) can be defined as the detection of an occurrence causing change
in the volume of social media data that discusses the associated topic at a specific time.
ED in SM is challenging because of short and noisy contents, diverse and fast changing
topics, as well as large data volumes (Petrovićet al., 2010; Li et al., 2012).
The research of ED in SM has attracted more and more interests from academia.
Researchers have published a wealth of relevant publications to report their research
findings. The growing number of publications witnesses its rapid development (e.g. Wang
et al., 2017; Hua et al., 2016; Kaleel and Abhari, 2015; Cheng and Wicks, 2014). These
research publications, conveying newly developed technologies, reflect the status and trend
of the research field to a certain extent. Therefore, a systematic and comprehensive analysis
of publications in this research field to help people understand its current status and
development trend is of great demand.
However, to the best of our knowledge, currently there is no scientific an
d comprehensive analysis of ED in SM research field based on quantitative and statistical
perspective. Therefore, we propose a bibliometric analysis of ED in SM to comprehensively
map the landscape of this research field. Specifically, this paper addresses the following three
research questions:
RQ1. What are the publication distributions (publication and citation quantities,
research subject distribution, predominant journals, countries/regions, affiliations
and authors) of the ED in SM research field?
RQ2. What are the scientific collaborations among countries/regions, affiliations
and authors?
RQ3. What are the major research themes and their evolutions upon time in this
research field?
To answer these questions, this study conducts a bibliometric analysis on academic
publications in this field during the period 2009–2017 to explore the general publication
distributions, reveal the collaboration relations, discover hot keywords and analyze major
research themes and their evolutions.
Literature review
Bibliometric analysis is the evaluation of scholarly publications from a quantitative
perspective within a certain field using statistical methods (Chiu and Ho, 2007). It enables
researchers to organize information in a specific field (Merigo et al., 2015), evaluate scientific
developments in the knowledge of a specific subject (Bouyssou and Marchant, 2011),
compare research performance across different countries (Pu et al., 2016), identify emerging
research focuses and predict future research success (Mazloumian, 2012), etc. Bibliometric
analysis has been widely applied to various fields for the measurement of quality and
productivity of academic outputs, e.g., public service management ( Juliani and de Oliveira,
2016), multimorbidity (Xu et al., 2017) and fuzzy theory (Yu et al., 2018). Especially, it has
also been applied to interdisciplinary research fields, e.g., natural language processing in
mobile computing (Chen, Ding, Xu, Wang, Hao and Zhou, 2018), and natural language
processing in medical research (Chen, Xie, Wang, Liu, Xu and Hao, 2018). Specially, some
bibliometric studies centered on social media-related topics. Leung et al. (2017) provided a
systematic review on 406 social media-related literatures during the period 2007–2016 with
co-citation and co-word analysis. Three topics including big data using text mining
methods, diverse research designs and word-of-mouth based on social theories were
30
OIR
43,1
To continue reading
Request your trial