What do people say about K-POP?
A Text Mining Approach to the Analysis of
BTS Fever
Research-in-Progress
Soobin Choi Gayeon Park
Hee-Woong Kim
Abstract
K-POP is steadily growing with global competitiveness. The rise of K-POP's
popularity has continued to create Korean idol groups. However, many idol
groups were dismantled and there is lack of measures for overseas advance
and success. Therefore, this study aims to analyze the success factors of
BTS by focusing on the text mining techniques. After collecting Twitter's online
postings using crawling technique, we will analyze in three text mining
techniques: topic modeling, keyword extraction, and term frequency analysis.
By analyzing data with three text mining methods, we will derive how BTS
could success globally and form a huge fandom. And with the derived key
factors, we will suggest a success strategy based on the analysis results. In
contrast to previous studies that were centered on case studies or interview,
this study has implications in that the actual data was collected and
analyzed through three text mining techniques.
Keywords: BTS, K-POP, Text Mining, Topic Modeling, Keyword Extraction, Term
Frequency Analysis
Introduction
The term 'Korean Wave', a term referring to Korean popular culture preference phenomenon, is
constantly expanding around the world. The popularity of K-POP (Korean pop) has been steadily
evolving around the world for the last 20 years and is rapidly emerging as a new content of the Korean
Wave. K-POP has a huge fan base not only in China, Japan, and Southeast Asia, but also in Eastern
Europe (Jang and Song 2017; Fu and Shon 2018). Through the social network service (SNS), K-POP
has formed a global fandom, increasing its awareness and proving its popularity (Choung 2010). BTS,
Korea's representative idol group, have a lot of global fans and the number is continuously increasing.
After one year of the debut, BTS started touring in Seoul with performances in Japan, Philippines,
Singapore, Thailand, etc., and quickly met with global fans. In 2015, BTS was selected as the Korean
representative for the “World Wide Act”, which selected as the most active artist in MTV EMA 2015
(European Music Awards). MTV EMA said, "The next generation of large-scale K-POP artist group
Twenty-Third Pacific Asia Conference on Information Systems, China 2019
What do people say about K-POP?
that has shown remarkable growth rapidly since its debut in 2013. BTS is trying to write and work
directly with the music, and to convey the thoughts and stories of the people through BTS music”.
On the iTunes K-POP chart in the US in 2015, the BTS’s album ‘Hwayang-Yeonhwa PT. 2’ and the
title song ‘RUN’ won the top album chart and top song chart, and the best K-POP album 4th place and
singing section '2015 best K-POP album and 2015 best K-POP song 20' at the same time. And “RUN,”
the album title track, were ranked fourth and third, respectively. When they released their second album
‘WINGS’ in 2016, it was ranked 26th in the US Billboard 200 chart. In addition, Billboard Album Chart
and Billboard Digital Song Chart were ranked # 1. Also, for the first time as a Korean singer, they
entered # 16 on the Indie album chart, including # 62 on the UK official album chart. In addition, it has
ranked # 1 on the latest weekly charts for Japanese music store tower record comprehensive album, # 1
on the Chinese video site, and # 1 on the iTunes music charts in the iTunes 97 countries all over the
world.
BTS entered the Billboard 200 with Billboard # 7 on the first week after releasing the ‘LOVE
YOURSELF’ album in 2017, and the five albums released until that time were on the Billboard 200
charts in succession. “DNA” was also ranked on Billboard's 100 Best Songs of 2017 for # 49, and BTS
was the only K-POP group. The BTS has once again proved its popularity as a Korean idol group for
the second time in the history of the Korean singer in 2012 after the release of ABC's new year's show
"New Year’s Rocking Eve 2018" live on December 31, 2017 throughout the United States.
In 2017, Big Hit Entertainment, an entertainment agency of BTS, raised its sales and operating profit
by 162.3% and 213.5% from 2016 to KRW 92.4 billion and KRW 32.5 billion, respectively. Hana
Financial Group expects Big Hit Entertainment sales to reach KRW 140 billion and operating profit to
reach KRW 50 billion in 2018. Big Hit Entertainment's 2017 operating profit surpassed the YG (25.2
billion won), JYP (19.5 billion won), and SM (10.9 billion won), respectively, which are classified as
Big 3 for domestic entertainment agencies.
The popularity of K-POP continues to grow, and the range of activities is so wide that if a group has
many fans profitability is easier. As the K-POP market becomes more active, various entertainment
agencies in addition to Korea's three leading entertainment agencies, SM, YG and JYP will continue to
compete for a new debut (Jang 2009). Many new groups continue to prepare for a debut, but most of
the groups are disbanded in a short period of time without getting popular. At this point, research on
strategies to grow into content that meets peoples’ needs and expectations is needed (Park et al., 2013).
Despite of the popularity of K-POP, previous research has focused on case studies or interviews
(Ubonrat and Shin 2007; Cho and Sim 2013). There has been a lack of empirical analysis, especially
about the critical success factor of K-POP. This study, therefore, aims to examine the critical success
factors of K-POP by focusing on the BTS case. So in this research, we will analyze the success factors
using text mining techniques and suggest success strategies that helps new idol groups can grow into
sustainable content.
We will do the text mining approach to analyze social opinions about BTS by crawling social media
data. Specifically, the main issues of the BTS will be derived through data analysis using three text
mining techniques: topic modeling, keyword extraction, and term frequency analysis. Also, we will
conduct interviews with BTS fans, entertainment officials, and industry experts for additional analysis.
Finally, with the derived key factors, we will propose critical success factors based on the text mining
of social opinions and the interview-based additional interpretation of those identified factors.
This study has academic significance in that it uses a mixed research method to derive major success
factors through quantitative research analyzing actual data and qualitative research through interviews.
Also the result of this study has a practical implications in that it suggests the success strategy and the
overseas advance plan of the newborn idol group in the future. In addition, it provides management
insight because it is closely related to the sales and operating profit of the entertainment agency.
Conceptual Background
Previous Research on K-POP and BTS
Twenty-Third Pacific Asia Conference on Information Systems, China 2019
What do people say about K-POP?
Most of previous research on K-POP was conducted through interviews. There were studies about K-
POP culture and success factors, the spread of the Korean Wave, and the relationship between
popularity and social media (Jin and Yoon 2016; Cho and Sim 2013; Leung 2012; Oh and Park 2012).
Table 1. Previous Research on K-POP
Author Method Field Purpose Results
Study how the Korean Social media platform and
Jin and
K-POP Wave phenomenon is social interaction of fans
Yoon Interview
Culture integrated into the culture interplayed and Korean
(2016)
of social media-led wave spread rapidly
Analyze K-POP's global
Changes in the smart
Cho and K-POP success factors and
environment and social
Sim Interview Success explore the impact of
media play a crucial role in
(2013) Factors smart media on K-pop's
spreading the K-POP craze
globalization
An online K-pop fan
culture builds a digital
Study how K-POP is community that
Leung
Interview K-POP Wave built, spread and accepted strengthens the bond
(2012)
globally between the fans and the
connection between fans
and music
There is an empirical
Oh and Social media plays an
relationship between
Park Interview K-POP Sales important role in validating
social media and K-POP
(2012) the K-POP business model.
popularity in Europe
Direct contact between
Ubonrat artists and fans (concerts,
K-POP How Thai youths became
and Shin Interview fan meetings) and fandom
Consumption consumers of K-POP
(2007) activities play an important
role in K-pop consumption
Jin and Yoon (2016) studied how the Korean Wave phenomenon is integrated into the social media-led
culture and confirmed that social interaction with fans through the social media platform quickly spread
the Korean Wave. Leung (2012) conducted a study on how K-POP is spreading and accepting Korean
Wave worldwide, and online K-POP fan culture builds digital community to strengthen the bond
between the fans and strengthen the bond between fans and music. The previous studies related to K-
POP are summarized as Table 1.
As it can be seen from previous studies, most studies related to K-POP have been conducted through
interview methods, and there are insufficient studies to identify the success factors by collecting and
utilizing actual data. Therefore, this study will collect and utilize actual online social media data as well
as interview data about BTS, and through this, we will derive the main success factors of BTS and
suggest the success strategy of the K-POP in the future.
Research Methodology
The research process of this study is shown in Figure 1. First, we crawl the data and collect the latest
data about BTS on Twitter. After data preprocessing, LDA topic modeling, keyword extraction, and
term frequency analysis of text mining techniques are used to derive the success factors of BTS.
Secondly, we will analyze the code with the data obtained from the interviews of the BTS fans, the
entertainment officials, and the industry experts and derive the success factors of BTS. As a result, we
will derive the key success factors of K-POP through quantitative data analysis of text mining and
qualitative data analysis using interviews. And with the derived results, we will propose a success
strategy for K-POP in the future.
Twenty-Third Pacific Asia Conference on Information Systems, China 2019
What do people say about K-POP?
Figure 1. Research Procedure
Text Mining: LDA Topic Modeling
In this study, we will use topic modeling as the first technique of text mining. Topic modeling is a
procedural probability distribution model that finds a certain pattern in a document or text and finds
potentially meaningful topics. It is an algorithm that probabilistically calculates and extracts the result
as a set of words likely to correspond to the topic on the assumption that the words constituting the
paragraphs are not independent (Blei 2012; X. et al., 2014). After several equations, the probability that
the i-th topic Z of the d-th document is assigned to the j-th is given. As a result, AB is the degree of
relevance of the d-th document to the k-th topic multiplied by the relevance of the n-th word of the d-
th document to the k-th topic (see Equation 1). Therefore, this study aims to analyze the database that
derives the success factors of BTS through LDA topic modeling.
𝒏𝒅,𝒌 +𝒂𝒌 𝒗𝒌,𝒘𝒅,𝒏 +𝜷𝒘𝒅,𝒏
𝐏(𝒁𝒅,𝒊 = 𝐣|𝒁−𝒊 , 𝐰) = ∑𝒌𝒊=𝟏(𝒏𝒅,𝒊 +𝒂𝒊 )
× ∑𝑽
= 𝑨𝑩 … .. Eq 1. LDA Probability Distribution
𝒋=𝟏(𝑽𝒌,𝒋 +𝜷𝒋 )
Text Mining: Keyword Extraction
In this research, we will conduct keyword extraction method as the second technique of text mining. A
keyword extraction method is defined as finding key keywords in the data field. By extracting keywords
that are topics, it can create a basis for understanding trends (M. and A., 2015). The following TF-IDF
weight model is used for keyword extraction (see Equation 2). The TF-IDF weight model evaluates the
relative importance of words within a document and determines the ranking of the most similar
documents. Therefore, a word that has a large TF-IDF value can be used as an important measure to
extract key words (Lee and Kim 2008).
𝑵
𝐓𝐅 − 𝐈𝐃𝐅 = 𝐓𝐅 × 𝐈𝐃𝐅 = 𝑭𝒕,𝒅 ∗ 𝐥𝐨𝐠 (𝒏 ) … .. Eq 2. TF-IDF Weight Model
𝒕
Text Mining: Term Frequency Analysis
In this study, we will use the term frequency analysis technique as the third text mining technique. The
term frequency analysis corresponds to the TF in the TF-IDF model and TF is the abbreviation of the
Term Frequency (see Equation 3). In other words, the number of occurrences of a particular word in
text data can be used to determine which word is most frequently referred to (J Pastizzo and F Carbone
2007). Therefore, in this research, we try to identify the most frequent occurrence of term in Twitter
data by using word frequency analysis method, and derive the success factors of BTS by using
frequency.
𝒏𝒊,𝒋
𝒕𝒇𝒊,𝒋 = ∑ … .. Eq 3. TF Equation
𝒌 𝒏𝒌,𝒋
Data Collection and Analysis Results
Data Collection: Twitter.com
The data for this study were collected from Twitter (Twitter.com) using crawling techniques for BTS
related posts. Twitter is a social network service and a microblog service. It is a platform where people
from all over the world can share and communicate their opinions and thoughts via tweets that are
Twenty-Third Pacific Asia Conference on Information Systems, China 2019
What do people say about K-POP?
shorter than 140 characters. The reason why we choose Twitter for data collection is that the BTS
account, which is now called the best twitter star, has tweets that are selected as the most tweeted
accounts in the world. Therefore, we can analyze and synthesize vast opinions. As a result, we have
collected 20,716 of BTS related data on Twitter since January 1, 2017 through December 31, 2018. Not
only the fans of BTS, but also the members of BTS actively participated and we were able to see many
articles about BTS in public.
Data Analysis Method
This chapter describes how to analyze the collected data. As it mentioned earlier in the research
methodology, we will analyze Twitter data by using three text mining techniques. Before analyzing the
data, we first preprocess the data, and then we use the topic modeling method to derive topics from BTS
on Twitter. Second, we will seize the keywords extracted from the collected data by using the keyword
extraction technique. Finally, through the term frequency analysis, we will identify which words are
most frequently mentioned through the frequency of occurrences of words in text data. As a result, the
main success factors of BTS will be derived based on the results obtained through the three techniques.
Analysis Results
First of all, with the result of topic modeling on the total tweets (20,716), we could identify that the
topics related to Loyalty to BTS, Voting for BTS, Loyalty to each BTS Member, ARMY (Fan Club),
BTS Concert and Performance, Awards, Best Idol Group, BTS Music and Performance, BTS
Collaboration, New Challenges of BTS are mentioned. Derived topics and topic-related keywords are
listed in Table 2.
Table 2. Topic Modeling Results
Description
Topic
Label Keywords
01 Loyalty to BTS angel, proud, love, beautiful, thank you, best, talented
02 Voting for BTS bts_twt, retweet, likes, replay, vote, awards, rates
Loyalty to each BTS
03 방탄소년단, 뷔, V, bts_twt, bts, jimin, 정호석, JIN, 태형,
Members
04 ARMY (Fan Club) ARMYs, btsarmy, 아미, mostrequestlive, bts_twt, minniejoons
05 BTS Concert and Tour btsloveyourselftour, btsincitifield, concert, newyork, btsworldtour
06 Awards billboard, pcas, disney, Europe, mpn, bts, epiphany
07 Best Idol Group kpop, idol, thegroup, bts_bighit, US, universe
BTS Music and
08 뷔, 방탄소년단, rap, music, jimmy, thesong
Performance
09 BTS Collaboration bts_collab, vt_cosmetics, moment, making, g8, g6
New Challenges of
10 rts, game, kpop, wanted, movie, tweet, heart,
BTS
Secondly, the analyze results through the keyword extraction as the second analysis technique, the top
30 key keywords are listed in order of the weight value which indicates the TF-IDF value. The top 30
extracted keywords are listed in Table 3 and it can be visually identified which words are important.
Next to each keyword, weights are displayed, and the keywords are interpreted as words. As a result, it
can be seen that not only keywords related to the BTS Member (8 keywords) are mentioned most
frequently but also weights are also highly distributed. Followed by BTS Member related keywords,
there are 6 keywords related to BTS Concert in the top 30, and weights are also generally higher. In
conclusion, it can be understood that the core keyword in the BTS related Twitter data are the member
Twenty-Third Pacific Asia Conference on Information Systems, China 2019
What do people say about K-POP?
name of BTS and the concert of BTS. It means fans like each member of the BTS, and they also like
BTS’s songs and performances at the concert.
Table 3. Keyword Extraction Results
Keyword Weight Interpretation Keyword Weight Interpretation
BTS
01 태형_뷔 1
Member
16 boxy_smile 0.455 Performance
BTS
02 milktae_jpg 0.771 17 concert_venue 0.445 BTS Concert
Member
03 shoot_dance 0.583 Performance 18 security_guard 0.433 BTS Concert
BTS family_membe
04 뷔_ 0.534
Member
19
rs
0.433 BTS Member
05 eng_vers 0.516 BTS Music 20 rap_duo 0.412 BTS Music
06 btschartdata 0.498 SNS Voting 21 finger_heart 0.391 Performance
btsloveyourselfto
07 0.498 BTS Concert 22 concerts 0.385 BTS Concert
ur
08 fanart 0.498 Fandom 23 슈가 0.368 BTS Member
09 lip_bite 0.498 Performance 24 fan_account 0.355 Fandom
music_nomina
10 universe 0.498 Idol Group 25 0.326 Awards
tions
11 boys 0.498 Idol Group 26 monster_army 0.279 Fandom
BTS
12 taehyung 0.497 27 bts_collab 0.270 Collaboration
Member
attendance_recor
13 0.464 BTS Concert 28 prince 0.266 BTS Member
d
14 solo_spotlight 0.458 BTS Concert 29 heart_attack 0.262 BTS Member
15 btsvotingteam01 0.385 SNS Voting 30 idol 0.257 Idol Group
Lastly, the term frequency analysis was used as the third analysis technique. It is a part of corresponding
to the TF (Term Frequency) of the TF-IDF model, and it can grasp how frequently a specific word
appears in the data. The results are listed in Table 4, top 30 frequency were selected. For each word, the
interpretation is displayed along with the frequency of occurrences in the table. As you can see from
the results table, we could find that ‘bts_twt’ has the highest TF value (1871), which is the official
account of BTS on Twitter. This means BTS members and fans are constantly communicating globally
through online official accounts.
As it can be seen on top 10 high frequencies, the frequency of terms about BTS Group is significantly
higher, and the terms about the BTS Concert and fan club (ARMY) also have a high frequency. As a
result of the data analysis, it can be confirmed that the fans mention a lot about the concert and words
indicating BTS themselves through the official SNS account. Also, since the fan-related words such as
the fan club (btsarmy) show high frequency, it can be seen that the fandom activity is very active.
Table 4. Term Frequency Results
Term Freq. Interpretation Term Freq. Interpretation
01 bts_twt 1871 BTS Group 16 jimmyfallon 104 Fandom
02 btsxcitifield 1606 BTS Concert 17 btsvotingteam 100 SNS Voting
03 bts 1076 BTS Group 18 army 87 Fandom
04 뷔 516 BTS Member 19 제이홉 83 BTS Member
05 방탄소년단 447 BTS Group 20 태형 81 BTS Member
06 btsincitifield 349 BTS Concert 21 thesong 78 BTS Music
07 pcas 274 Awards 22 taehyung 78 BTS Member
08 citifield 252 BTS Concert 23 rm 76 BTS Member
09 btsarmy 243 Fandom 24 mostrequestliv 74 BTS Music
Twenty-Third Pacific Asia Conference on Information Systems, China 2019
What do people say about K-POP?
e
10 minniejoons 203 Fandom 25 Jungkook 73 BTS Member
btsloveyourselfto
11 201 BTS Concert 26 bts_bighits 71 Kpop Idol
ur
12 idol 195 Kpop Idol 27 concert 71 BTS Concert
13 지민 160 BTS Member 28 정호석 71 BTS Member
14 thegroup 142 Kpop Idol 29 themusicvideo 47 BTS Music
15 newyork 113 Fandom 30 mpn 42 Awards
Using the three text mining techniques, we analyzed 20,716 Tweets related to BTS and found the results.
The result of keyword extraction and term frequency analysis were mapped on 10 topics derived from
the topic modeling. The 10 topics are Loyalty to BTS, Voting for BTS, Loyalty to each BTS Members,
ARMY(Fan Club), BTS Concert and Tour, Awards, Best Idol Group, BTS Music and Performance,
BTS Collaboration, and New Challenges of BTS. So the main success factors of BTS, which were
commonly found through three text mining techniques, are Voting for BTS, Loyalty to each BTS
Member, ARMY (Fan Club), BTS Concert and Tour, BTS Music and Performance, and Best Idol Group
(7 factors). Therefore, the overall interpretation of mapped result is that we could find that ARMY,
which is a fan club name of BTS, is actively promoting its presence by using twitter. Which means the
official online fan account is active all over the world. Also, we could identify the interest and
enthusiasm for the music, performance, and concert of the BTS and loyalty to each member of the BTS.
It also proves that BTS is the most successful idol group, and it shows that BTS has formed worldwide
fandom. The popularity is proved once again through the awards.
As a result, the main issues of BTS are: most of all, the global fandom (ARMY) is formed and it proves
they are the most powerful and dedicated fandom in the world by active activity. Second, ARMYs (BTS
fan club) plays the important role on twitter which promotes to people for voting when music awards
voting are ongoing. Third, fans of BTS have great loyalty to each BTS members. So they keep
mentioning on twitter about each member’s name and this gives an effect of spreading the name of BTS.
Fourth, BTS keep communicating face-to-face with fans through concert and a world tour. Fifth, BTS
shows good music and stage that captured the hearts of fans. Sixth, BTS is recognized worldwide that
they are the best idol group not only they are good at dancing and singing, but also they have good
manners and shows kindness to everyone. Lastly, BTS was awarded at many global music awards, such
as in ‘Billboard’ and ‘PCAs’ and so on, it proves they are successful globally.
Future Research Plan
In the future research, we will further interpret and validate those main issued identified from the text
mining approach. For this purpose, we are going to apply interview-based qualitative approach: fans,
agency, and industry experts. We will conduct in-depth interviews with actual BTS fans about the
reasons why fans like BTS and why fans keep going on fan activities. As a result of text mining methods
shows one of the main factors of success of BTS is that ARMYs’ (fan club) dedicated activities and
powerful reactions, this interview will be an important part to derive key success factors of BTS. We
will also interview the representatives of BTS 's agency and related people to understand the strategy
that they have developed as a global idol group. We can then find the strategies how BTS could
acknowledge their faces to the world. Lastly, we will interview experts related to the entertainment
industry to find out the main factors of success. This interview part could give us another side of
opinions of success of BTS. Based on the combination of the text mining analysis results and the
interview analysis results, we will propose meta-inferences regarding the social opinions about K-POP
(i.e., BTS), which may suggest the critical success factors. This study thus contributes to the literature
by proposing the mixed methods approach and identifying the critical success factors for K-POP and
digital contents business in general.
References
Blei, D. M. 2012. “Probabilistic topic models,“ Communication of the ACM (55:4), pp. 77-84.
Twenty-Third Pacific Asia Conference on Information Systems, China 2019
What do people say about K-POP?
Cho, Byung-Chul and Sim, Hichul. 2013. “Success Factor Analysis of K-POP and A Study on
sustainable Korean Wave - Focus on Smart Media based on Realistic Contents,” The Journal of the
Korea Contents Association (13:5), pp. 90-102.
Choung, Jae-Min. 2010. “An Analysis of Youth Fandom Based on the Ground Theory,” Studies on
Korean Youth (21:3), pp. 91-119.
Fu, Kang and Shon, Seunghye. 2018. “A Comparative Analysis of Composition and Activities of Idol
Group Fandom among Chinese Fans- Grounded theory analysis of TFBOYS and BTS fan cafe
member interviews,” Media, Gender&Culture (33:2), pp. 5-43.
J Pastizzo, Matrhew and F Carbone, Robert. 2007. “Spoken word frequency counts based on 1.6 million
words in American English,” Behavior Research Methods (39:4), pp. 1025-1028.
Jang, Gyu-Soo. 2009. “Overseas advance instance and strategic study of Korean pop music,” The
Journal of Global Contents (9:6), pp. 217-238.
Jang, Won ho and Song, Jung Eun. 2017. “The Influences of K-pop Fandom on Increasing Cultural
Contact: With the Case of Philippine Kpop Convention, Inc.,” Journal of the Korea Entertainment
Industry Association (18:2), pp. 29-56.
Jin, D. Y., and Yoon, K. 2016. “The social mediascape of transnational Korean pop culture: Hallyu 2.0
as spreadable media practice,” New Media & Society (18:7), pp. 1277–1292.
Lee, S. and Kim, H. 2008. “News Keyword Extraction for Topic Tracking,” 2008 Fourth International
Conference on Networked Computing and Advanced Information Management, pp. 554-559.
Leung, Sarah. 2012. “Catching the K-Pop Wave: Globality in the Production, Distribution, and
Consumption of South Korean Popular Music,” Senior Capstone Projects, pp. 149.
M. Habibi and A. Popescu-Belis. 2015. “Keyword Extraction and Clustering for Document
Recommendation in Conversations,” IEEE/ACM Transactions on Audio, Speech, and Language
Processing (23:4), pp. 746-759.
Oh, Ingyu and Park, Gil-Sung. 2012. “From B2C to B2B: Selling Korean Pop Music in the Age of New
Social Media,” KOREA OBSERVER (43:3), pp. 365-397.
Park, Cheol., Kim, In Kiw., and Jeon, Youn. 2013. “Policy Research: The Development Strategy of K-
POP through the Improvement of Digital Music Distribution in Korea,” Korea Business Review
(17:3), pp. 251-265.
Ubonrat, Siriyuvasak and Shin, Hyunjoon. 2007. “Asianizing K‐pop: production, consumption and
identification patterns among Thai youth,” Inter‐Asia Cultural Studies (8:1), pp. 109-136.
X. Cheng, X. Yan, Y. Lan and J. Guo. 2014. “BTM: Topic Modeling over Short Texts,” IEEE
Transactions on Knowledge and Data Engineering (26:12), pp. 2928-2941.
Twenty-Third Pacific Asia Conference on Information Systems, China 2019
