Content uploaded by Volkan Aşkun
Author content
All content in this area was uploaded by Volkan Aşkun on Feb 22, 2021
Content may be subject to copyright.
Content uploaded by Volkan Aşkun
Author content
All content in this area was uploaded by Volkan Aşkun on Feb 22, 2021
Content may be subject to copyright.
Please cite this article as: Kirtil, I. G., & Askun, V., Advances in Hospitality and Tourism
Research, https://doi.org/10.30519/ahtr.801690
1
ARTIFICIAL INTELLIGENCE IN TOURISM: A REVIEW AND
BIBLIOMETRICS RESEARCH
İsmail Gökay KIRTIL
1
Demre Dr. Hasan Ünal Vocational School, Akdeniz University, Turkey
ORCID: 0000-0002-3520-9600
Volkan AŞKUN
Demre Dr. Hasan Ünal Vocational School, Akdeniz University, Turkey
ORCID: 0000-0003-2746-502X
ABSTRACT
Artificial Intelligence (AI) came up as an ambiguous concept from
computer sciences and now it is being used in many areas of our
life. It has stimulated academia’s interest due to its alternative
insights into complex problems. Therefore, a bibliometric method
was applied in this study to observe the progress of AI in the
tourism field. A total of 102 papers were collected from Scopus
database. Key factors such as most productive authors,
collaborations and institutions were identified, and research
hotspots were determined using co-occurrence network and most
common author keywords. Progress of AI was visualized with
thematic evolution analysis. Findings indicate that there is a
progressive interest in AI after 2017, and average citations signify
that papers are highly cited. Since this is the first study conducting
a bibliometric on AI in the tourism context, it could be considered
useful for academics and tourism professionals as it provides
general overview of AI, demonstrates research trends and popular
papers.
INTRODUCTION
“I believe there is no deep difference between what can be achieved by a biological brain and
what can be achieved by a computer. It therefore follows that computers can, in theory,
emulate human intelligence, and exceed it.” Stephen Hawking
Information and Communication Technologies (ICTs) in tourism, also
known as e-tourism concept, started a new era in contemporary tourism,
1
Address correspondence to İsmail Gökay Kırtıl, Demre Dr. Hasan Ünal Vocational School, Akdeniz
University, Antalya, Turkey. E-mail: gokay@akdeniz.edu.tr
Keywords
Bibliometric
artificial intelligence
hospitality and tourism
co-citation analysis
co-occurrence analysis
thematic analysis
Advances in Hospitality and Tourism Research (AHTR)
An International Journal of Akdeniz University Tourism Faculty
ISSN: 2147-9100 (Print), 2148-7316 (Online)
Webpage: http://www.ahtrjournal.org/
Article in
press
Article History
Received 29 September 2020
Revised 9 December 2020
Accepted 21 December 2020
Published online 22 Feb. 2021
DOI: 10.30519/ahtr.801690
2
and hospitality industry. ICTs enabled researchers to assess tourist
behavior through intelligent systems much faster and allowed them to deal
with large amount of data coming from both tourists and destination
parties. ICTs also affected tourist behavior radically (Buhalis, 2003) by
changing the way tourists consume, purchase, and share their experiences
(Gretzel et al., 2006). Tourists and service providers had the chance to access
relevant information more accurately, with increased mobility and a greater
decision-making process, eventually, acquiring a more favorable tourism
experience (Gretzel, 2011).
In consideration of advances in ICTs, Artificial Intelligence (AI) is
regarded as the next stage of tourism industry (Bowen & Whalen, 2017;
Gajdošík & Marciš, 2019; Kazak et al., 2020). AI is known for its
sophisticated computing capabilities as it can deal with complex relations
and problems among different concepts (Pannu, 2015) and can easily work
with a big amount of data (Inanc-Demir & Kozak, 2019). Broadly speaking,
an AI system senses external information, understands these, acts in turn to
achieve given goals and learns from its own experiences (Ferràs et al., 2020).
AI functions similar to a human brain as it thinks, learns, makes decisions
and inferences through given data by using intelligent machines. The main
purpose of AI is to enable machines to complete tasks automatically
without needing a human brain (Singh et al, 2020).
Since the late 1990s, AI studies have been applied in tourism
researches to forecast hotel occupancy and tourism demand (Law, 1998,
2000). Afterwards, researchers used AI in different kind of inquiries such as
resource management in tourism companies (Casteleiro-Roca et al., 2018),
examining social media data and online reviews (Kirilenko et al., 2018;
Topal & Uçar, 2018), forecasting tourist flow and arrivals (Zhang et al.,
2020), evaluating tourist satisfaction through facial expression recognition
(González-Rodríguez et al., 2020), and making smart recommendations
(Zheng et al., 2020). AI models are used in tourism studies increasingly
because these models have much more flexibility and they can be used to
estimate non-linear relationships without the limits of traditional methods
(Hadavandi et al., 2011).
Although AI promises entirely alternative solutions for potential and
prospective issues of tourism, with its advanced computing and problem-
solving abilities, there is a lack of academic research on AI in context of
tourism (Gajdošík & Marciš, 2019; Zlatanov & Popesku, 2019). Therefore,
this study adopted a bibliometric method to evaluate the progress, research
themes, and statistical data of AI in tourism field within the scope of data
3
gathered from Scopus database. In line with this purpose, main objectives
of this study are to:
Provide an expanded overview of AI,
Explore the overall theoretical foundation and progress of AI research
in tourism field by focusing on leading contributors (authors, keywords,
publications, and institutions),
Visualize above-mentioned metrics and the evolution of AI, and
Suggest a future research agenda for tourism academicians and
practitioners.
The findings of this study have several useful implications. For social
scientists and tourism researchers interested in AI, the study indicates an
overview of the subject in concern with key studies, authors, collaborations,
and emerging topics. As an emerging and interdisciplinary field, AI can
provide different insights into social sciences and it may help us to
understand complex social issues (Pavaloiu et al., 2017). Particularly in
tourism context, this kind of an insight may provide useful perspectives to
crises and chaotic situations such as global pandemics or disasters (Ritchie,
2004). On the other hand, this study may affect future research trends and
career development of individual researchers (Law et al., 2010). Tourism
managers can also benefit from AI’s abilities such as complex computing
and dealing with large volume of data.
LITERATURE REVIEW
Pritchard (1969) introduced bibliometrics as the application of
mathematical and statistical methods on books and other types of
communications. Bibliometric methods are used to assess the impact of
researchers, institutions, countries, or journals (Cunill et al., 2019) and they
are useful to gain a macroscopic view of large amounts of academic
literature (van Nunen et al., 2018). Bibliometric methods are powerful for
assessing journal performances (Cunill et al., 2019; García-Lillo et al., 2016;
Guzeller & Celiker, 2019; Merigó et al., 2019), evaluating the progress of a
specific field at a given time period (Askun & Cizel, 2019; Dhamija & Bag,
2020; Koseoglu et al., 2016; van Nunen et al., 2018) and especially in
evaluation of international scientific influence of an agent (van Raan, 2003).
Bibliometrics is used across different disciplines and it’s complementary to
traditional methods (Zupic & Čater, 2015). Due to its more objective and
reliable analyzes compared to other qualitative and quantitative reviewing
4
approaches (Aria & Cuccurullo, 2017), scholars are increasingly interested
in bibliometrics as a research method.
Koseoglu et al. (2016) classified bibliometric methods as review
studies, relational techniques, and evaluative techniques. They categorized
systematic reviews, meta-analyzes and qualitative approaches in review
studies; citation, bibliographic, co-word, co-authorship analyzes in relational
techniques; while productivity measures, impact metrics and hybrid metrics
are classified as evaluative techniques. Review studies use basic statistics or
qualitative methods to assess a scientific study. Relational techniques try to
discover the relationships in studies such as structure of the research fields,
new research themes and techniques (Güzeller & Çeli̇ker, 2018), whereas,
evaluative techniques analyze the impact of scholarly work and compare
the performance or scientific contributions of two or more individuals or
groups (Benckendorff & Zehrer, 2013).
Bibliometric methods have been used in tourism, leisure and
hospitality to assess the scientific production of the field. Furthermore,
these were applied in context of different subfields such as smart tourism
(Johnson & Samakovlis, 2019), gastronomy (Okumus et al., 2018), lodging
industry (Köseoglu et al., 2018; Okumus et al., 2019), sustainable tourism
(Ruhanen et al., 2015), rural tourism (Ruiz-Real et al., 2020), wine tourism
(Sánchez et al., 2017), tourism’s economic impact (Comerio & Strozzi, 2019),
social media (Leung et al., 2017), peer to peer studies (Andreu et al., 2020;
Núñez-Tabales et al., 2020), psychological research on tourism (Barrios et
al., 2008), and competitiveness and innovation (Teixeira & Ferreira, 2018).
Researchers apply different types of bibliometrics in their studies.
Benckendorff (2009) examined papers of Australian and New Zealand
researchers published in Annals of Tourism Research and Tourism
Management journals between 1994-2007 by using keyword, citation, co-
citation, and network analyzes. Okumus et al. (2018) analyzed the progress
of food and gastronomy in tourism field between 1976 and 2016, focusing
on most productive journals and institutions, and contributions of countries
to the scientific field. In another study, researchers identified the emerging
themes in tourism and stated that bibliometric studies can enlighten the
unknown patterns in disciplines and support future theory development
(Koseoglu et al., 2016). Virani et al. (2019) examined medical tourism
policies and combined bibliometrics and data visualization techniques.
Another distinctive point of bibliometrics is the visualization of results,
thus, the method increases the comprehension of potential readers in an
emergent area and extends the research scope (Qian et al., 2019).
5
Since bibliometrics is applicable to all scientific areas (Sánchez et al.,
2017), AI can be analyzed with this tool. In terms of AI, to the best of our
knowledge, bibliometrics has been conducted in different disciplines except
tourism and hospitality. For instance, Tran et al. (2019) conducted a research
on AI in health and medicine field. They reached 27,451 published
documents between 1977 and 2018 from Web of Science (WoS) database.
After the year 2002, numbers of AI studies in the health and medicine field
bursts exponentially due to the advances in computing and data storage
capacities. Authors also visualized author and country collaborations and
networks. They revealed that the highest number of papers related to AI
were about robotic surgery, machine learning and artificial neural network,
respectively. Niu et al. (2016) examined 22,072 publications between 1990
and 2014 without delimiting the scientific field. According to this study,
computer science and engineering were the most productive fields in
context of AI, but the AI subject was also used in several other scientific
fields as an interdisciplinary matter. They found that, among 122 countries
that participated in AI research, the most productive ones were the USA,
China, UK, Spain, France, Germany, and Canada, respectively. Chinese
Academy of Sciences was the most productive institution, followed by
Massachusetts Institute of Technology (MIT) and Hong Kong Polytech
University.
Similar to the aforementioned research, Lei & Liu (2019) conducted
a study between 2007-2016 with the keyword ‘artificial intelligence’ but
without delimiting the scientific field. They also found USA was the most
productive country in AI studies, followed by UK and Iran, respectively.
They highlighted that during 10-years period 1,188 articles were published
in 102 research fields. They also emphasized interdisciplinary nature of AI,
with technical methods such as anfis (adaptive network based fuzzy
inference systems), support vector machine (a kind of machine learning),
genetic algorithm and fuzzy logic being the most utilized techniques.
Besides, in terms of research fields, neural network and machine learning
were the most prominent areas. In another research, Shukla et al. (2019)
examined the journal of Engineering Applications of Artificial Intelligence
(EAAI) between years 1988-2018 on both WoS and Scopus indexes. After
2008 the number of publications started to increase significantly.
Distinctively, they divided total citations to total publications (Citations Per
Paper), and they also calculated average citations received by a publication
per year (Citations Per Year) as these are effective metrics to show the
impact of a publication. According to Scopus data, neural networks,
algorithms, genetic algorithm, artificial intelligence, expert systems, fuzzy
6
sets, fuzzy logic were the trendiest author keywords. According to WoS,
developing countries such as Iran, India, Taiwan, and Turkey were among
the top 10 countries that contributes to the EAAI journal, albeit, China was
the top contributor, followed by the USA.
Many bibliometric studies have been conducted in literature to
examine the progress of AI in different scientific fields. There are some
commonalities in these researches such as the prominent countries
regarding scientific production and, in terms of keywords, emerging topics.
Authors divide their researches into time periods to distinguish periodical
emerging different themes to show AI’s rapid progress after spreading into
other disciplines. It is obvious that AI is commonly being studied in the
fields such as engineering, computer sciences, and medical and clinical
studies rather than social sciences. Hence, the current study aims to bridge
this gap in the tourism and hospitality field and to provide some useful
insights into AI’s potential for both academia and practitioners.
Furthermore, this study proposes an AI perspective into the social world’s
complex problems.
RESEARCH METHOD
Analytical Ideology
A research philosophy, which may be assumed as a social paradigm,
represents a scientific interest and guides the entire study (Gunbayi & Sorm,
2018). It helps to enlighten the research problems systematically by
employing necessary tools and methods for research. Therefore, this
research adopted a qualitative way in terms of interpretive paradigm
(Gunbayi & Sorm, 2018) based on the systematic analysis of articles on AI
in tourism through bibliometric analysis using R programming language
(Askun & Cizel, 2020).
R is a free and proper program that provides open source packages,
such as bibliometrix R- package specifically developed for bibliometric and
scientometric studies. Since bibliometrix R- package is an effective, flexible,
and adaptive tool, it is useful for the current study in performing the
bibliometric analyzes (Aria & Cuccurullo, 2017). For data visualization,
ggplot2 library (http://cran.r-project.org/) and VOSviewer were used.
Papers were analyzed by keywords plus, authors’ keywords, and titles,
while network analysis, co-citation, collaboration, co-occurrence analyzes
were performed to analyze keywords. Moreover, author, country, and
7
institution effect in context of tourism was reviewed and discussed to
determine the progress of the field. In general, this study investigates the
most cited papers, collaborations, co-citations, thematic analysis of the field,
keyword co-occurrence, and most common keywords of AI in tourism
studies, respectively.
Most cited papers show the prominent studies in terms of total
citation, local citation, and average citation. Yearly average citation of each
paper was calculated to show paper’s impact. Most cited papers refer the
most significant papers, but most cited papers are not always the most
relevant (Merigó et al., 2019). Therefore, for assessing document quality,
other analyzes are considered necessary.
On the other hand, collaboration analysis was conducted on author
and institution level. Collaboration networks depict the clusters of research
groups consisted of authors and institutions. These networks are distinctive
characteristics of contemporary researches because scholars tend to act as
members of a team rather than individual actors (Glänzel & Schubert, 2005).
Assessing author collaboration networks enlightens the way of analyzed
scientific knowledge among authors and shows prominent scholars,
therefore it gives important insights about the future of the field. AI
collaboration network in tourism was taken from author×author adjacency
matrix which counts collaborating papers.
Co-citation analysis explains groups of papers which are likely to
appear together in reference lists, but which also may have something in
common (Benckendorff & Zehrer, 2013). Co-citation analysis aims to show
the relationship of vast knowledge between documents. If documents are
gathered through two documents, this means they’re connected to each
other and strength of this connection is in accordance with the number of
connected documents. When two different documents compile many
documents, that means there is a strong connection. It can be inferred that
these documents share the same accumulation of knowledge or the same
methodology (Todeschini & Baccini, 2016).
Co-occurrence analysis visualizes network connections and
keywords frequently used in different documents. Creating a co-occurrence
network among keywords, title and abstract of a document enables
delivering a conceptual structure regarding the subject. A more frequently
used keyword is represented by a larger node in the graph. Lines indicate
connections between nodes and their thickness implies the strength of the
relationship. Position and color of the nodes imply different theme clusters,
whilst the distance between nodes asserts inverse proportion. Shorter the
8
distance means greater co-occurrence between keywords, longer the
distance means minor co-occurrence. Hence, conducting thematic evolution
analysis with keyword plus is very useful. Thematic evolution displays the
longitudinal progress of AI and implies the change in time periods. It
visualizes the evolution of the field and enables a smooth progressive
overview of the field. Co-occurrence analysis was conducted by keywords
such as in thematic analysis, but differently, co-occurrence analysis
depends on author keywords. Lastly, most common author keywords
occurrences clarify clusters of each keyword and their occurrences in a
chart-format. Keyword occurrences refer research trends of a scientific field
and may also infer possible future trends.
Data Source
This study’s data were obtained from the peer-reviewed literature database
Scopus. Scopus and WoS are two prominent databases for analyses, and
there is still an ongoing debate upon which one is better. Both databases
offer comprehensive coverage at journal, article and cited reference level
(Norris & Oppenheim, 2007). Before conducting this research, topic words
and keywords were applied to both databases, and as Scopus included
significantly greater number and type of documents than WoS, it was
preferred. Scopus offers articles, book chapters, conference papers, reviews,
notes, and letters, thus providing a broader view of scientific documents.
Since the current research is a systematic analysis, aiming to position
and synthesize studies about a specific research question, it uses organized,
transparent, repeatable procedures in each step of the process (Littell et al.,
2008). It utilizes purposeful sampling method and criterion sampling
technique that are commonly used in qualitative research methods (Palys,
2008), in which keywords are sampling criterions. To create the dataset for
analysis “artificial intelligence” was searched in author keywords or in
abstract, whilst, “tourism” was searched in topic or in abstract. An advanced
search was conducted without limiting to year, document type or language
criterions. Finally, papers published between 2003 and 2020 were
downloaded from Scopus on August 15, 2020. A total of 102 papers were
analyzed including 52 articles, 35 conference papers, 8 reviews, 5 book
chapters, 1 note and 1 letter. A remarkable number of conference papers
indicates that there is a growing interest to this field, while gathering other
scientific sources ensure data diversity.
In the next step, bibliographical data (e.g., papers, authors, titles,
keywords, references) were downloaded in CSV format, in line with
9
bibliometric methods proposed by Cobo et al. (2011) and Börner et al.
(2005). Figure 1 displays the rapid growth of AI studies in tourism field,
especially in recent years. There were only 3 papers published in 2003, and
until 2017 there wasn’t much attention to this subject. However, after 2017
the number of studies has grown significantly (annual growth rate: 8.36%).
The advancements in computer science and the proliferation of Internet
may have affected the authors’ tendency on AI, albeit these advances
enabled much faster reach for data.
Figure 1. Annual Scientific Production
RESULTS AND DISCUSSION
Data of this research consisted of 263 authors and 102 publications, with 610
different keywords that authors used to classify their documents. Average
citation per paper (16.58) and annual average citation per paper (3.04)
denote that papers are highly cited and they’re gaining importance
gradually. Single-authored papers were conducted by 18 authors, whereas
multi-authored papers were conducted by 245 authors (TAm). There are 18
single-authored papers and 84 multi-authored (TPm) ones. In that case, there
is a predominant collaboration upon studies, as shown by author per paper
(2.58) and co-author per paper (2.8) metrics. Because of the complex nature
of interactions among authors, structure and strength of collaborations
cannot be easily determined. In this case, Collaboration Index (CI) can be an
effective tool to overcome that concern. CI can be calculated by a formula
from Ajiferuke et al.’s (1988) study:
34323
8
6
32323
5
21
24
10
0
5
10
15
20
25
30
2003 2005 2007 2009 2011 2013 2015 2017 2019
Documents
10
𝐶𝐼 = 𝑇𝐴𝑚
𝑇𝑃
𝑚
= 2.92
Papers examined in this study (total of 102) have received 1,691
citations which means a number of 16.58 citations per paper. Total citations
are related with the visibility of a paper, and also roughly imply the quality
and impact of a study. Thus, the increasing amount of citations on open
access journals’ papers may provide a better interpretation to that case
(Chiu & Ho, 2007). The current study was conducted from 76 different
sources, and number of 38.61 citations per paper demonstrates that, studies
in tourism upon AI will gradually enhance their academic efficiency.
Most Cited Papers and Collaborations
Table 1 shows the most influential 15 papers in tourism field regarding AI.
This table shows the title, total citation, local citation, and annual average
citation of the papers. Akehurst’s (2009) study on improving user generated
content and web blogs received 218 citations and became the most cited
paper, whilst, it is in the sixth place in terms of annual average citation
(19.8). The paper authored by Borràs et al. (2014), which analyzed
conference papers presented upon intelligent e-tourism field focusing on
different types of interfaces and the usage of AI techniques, came in the
second place with 214 citations, but on the fourth in annual average citation
(35.7). In terms of annual average citation, on the other hand, Buhalis &
Sinarta’s (2019) research upon how tourism brands’ instant interaction with
customers’ enhances technology and social media was in first place with 56
citations within one year period. The second place was Song et al.’s (2019)
research with 40 citations within one year period, upon determining the
complexity of tourism demand and different forecasting methods. Buhalis
et al.’s (2019) research concerning examples on information-based tourism
industry’s effects on intelligent settings such as AI, was in the third place
with 38 citations within a year. Besides, Cho’s (2003) research on forecasting
the nature of tourist traffic and changes in tourism demand hits 168 citations
in total, but with an annual average citation of 9.9 demonstrating that,
recent studies are arousing more interest among researchers.
Figure 2 displays the AI collaboration network patterns in tourism
between years 2003-2020. Leading 30 authors, collaboration of minimum
one paper, and papers that show the strongest connections were taken into
consideration in this analysis. Lines and their thickness indicate the
presence of different collaborations.
11
Table 1. Most Cited Papers
References
Journal
Title
Year
TC
LC
C/Y
1
Akehurst, G.
Service Business
User generated content: the use of blogs for tourism
organisations and tourism consumers
2009
218
3
19.8
2
Borràs ,J., Moreno,
A., Valls, A.
Expert Systems with
Applications
Intelligent tourism recommender systems: A survey
2014
214
4
35.7
3
Cho, V.
Tourism Management
A comparison of three different approaches to tourist
arrival forecasting
2003
168
8
9.9
4
Cambria, E. Speer, R.
Havasi, C., Hussain,
A.
2010 AAAI Fall
Symposium Series
SenticNet: A Publicly available semantic resource for
opinion mining
2010
146
0
14.6
5
Goh, C., Law, R.
Tourism Management
Incorporating the rough sets theory into travel demand
analysis
2003
99
4
5.8
6
García-Crespo, A., et
al.
Expert Systems with
Applications
Sem-Fit: A semantic based expert system to provide
recommendations in the tourism domain
2011
68
3
7.6
7
Yu, G., Schwartz, Z.
Journal of Travel
Research
Forecasting short time-series tourism demand with
artificial ıntelligence models
2006
56
7
4.0
8
Buhalis, D., Sinarta,
Y.
Journal of Travel &
Tourism Marketing
Real-time co-creation and nowness service: lessons from
tourism and hospitality
2019
56
4
56.0
9
Hadavandi, E., et al.
Tourism Management
Tourist arrival forecasting by evolutionary fuzzy systems
2011
54
3
6.0
10
Felfernig, A., et al.
OGAI Journal
A short survey of recommendation technologies in travel
and tourism
2006
49
1
3.5
11
Goh, C., Law, R.
Journal of Travel &
Tourism Marketing
The methodological progress of tourism demand
forecasting: A review of related literature
2011
44
4
4.9
12
Song, H., Qiu, R.T.R.,
Park, J.
Annals of Tourism
Research
A review of research on tourism demand forecasting:
Launching the Annals of Tourism Research Curated
Collection on tourism demand forecasting
2019
40
0
40.0
13
Buhalis, D., et al..
Journal of Service
Management
Technological disruptions in services: lessons from
tourism and hospitality
2019
38
4
38.0
14
Kim, K., Park, O.,
Yun, S., Yun, H.
Technological
Forecasting and Social
Change
What makes tourists feel negatively about tourism
destinations? Application of hybrid text mining
methodology to smart destination management
2017
31
1
10.3
15
Lu,L., Cai, R.,
Gursoy, D.
International Journal
of Hospitality
Management
Developing and validating a service robot integration
willingness scale
2019
26
3
26.0
TC: Total citation, LC: Local citation, C/Y: Total citation/Years
Figure 2 reveals that there were 7 different author collaborations. The
greatest author collaboration was consisted of Moreno, Borràs, Valls,
Anton-Clavé, Flor, Isern, Russo, Pérez, and, these were in different
universities of Spain. These authors’ book chapter about recommender
systems on geographical information systems regarding tourism
destinations, and, Moreno, Borràs and Valls’ article in 2014 with 214
citations influenced this primacy. In another collaboration, Buhalis from
Bournemouth University, UK, published 3 different papers in 2019. Buhalis
et al.’s (2019) paper with 38 citations, and Volchek et al.’s (2019) research
upon tourists visiting five different London museums (13 citations) were
among the influential ones. Webster from USA and Ivanov from Bulgaria
12
have published four different papers since 2018 and became most
productive and collaborative authors, although these papers got only 16
citations. Hadavandi and Ghanbari have collaborated in two studies. The
research in ninth place at Table 1, which offers a solution regarding tourist
arrival forecasting (54 citations) and the conference paper on the same topic
(4 citations) affected that collaboration. Bouslama, Ayachi, and Amor from
Tunusia contributed to literature by presenting two different conference
papers in Spain and Serbia.
Figure 2. Author collaboration network
Prominent Countries and Institutions
In terms of institutions there are 38 different countries and the most prolific
countries considering the number of papers were Spain (36), China (33),
USA (24), UK (21), and Iran (14), respectively. Moreover, China (TC: 100)
has got six corresponding authorships, whereas Hong Kong (TC: 320), Iran
(TC: 72), Spain (TC: 294) and UK (TC: 210) got four of it. Accordingly, top
20 institutions that collaborated in at least one research were taken into
consideration, and Figure 3 illustrates this collaboration among institutions
(bolder line means more collaboration), likewise, total number of papers
were expressed with the size of figure. There were five different
collaboration groups. First of all, the research that mostly affected the
collaboration of Bournemouth University (UK), De Montfort University
13
(UK), The Ohio State University (USA), University of Portsmouth (UK),
University of Delaware (USA), and Florida State University (USA) was the
one in which Buhalis was the corresponding author. Bournemouth
University (UK) from the same group has linked with another connection
to University of Surrey (UK) and The Hong Kong Polytechnic University
(Hong Kong) through Buhalis’ research on five different London museums.
The collaboration of Ball State University (USA) and Varna University of
Management (Bulgaria) was due to Craig Webster and Stanislav H.
Ivanov’s researches. Besides, Hadavandi and Ghanbari’s research affected
the cooperation of the group Sharif University of Technology (Iran),
University of Tehran (Iran), Iran University of Technology (Iran).
Figure 3. Institution collaboration network
Co-Citation Analysis
Figure 4 and Table 2 demonstrate the intellectual structure of AI in tourism
field. Betweenness centrality (BC) in Table 2 is an advanced metrics which
shows the importance of a node to create the shortcuts among other nodes,
and also indicates the degree of influence of the communication between
nodes (Freeman, 1977). Adapting from Guns et al. (2011) to calculate BC as
follows;
Pkj gives the number of the shortest paths that connects k and j edges,
whereas Pkj(i) gives the number of the shortest paths passes through i edge.
V, is the number of edges in the graph.
14
Figure 4. Co-citation paper network
Table 2. Co-citation paper network overview
Cl
References
Title
Year
BC
C
C/Y
B
Witt, S. F., Witt,
C. A.
Forecasting tourism demand: A review of
empirical research
1995
32.71
1,118
44,72
B
Law, R.
Back-propagation learning in improving the
accuracy of neural network-based tourism
demand forecasting
2000
20.36
407
20,35
B
Cho, V.
A comparison of three different approaches to
tourist arrival forecasting
2003
13.00
434
25,52
B
Li, G., Song, H.,
Witt, S.F.
Recent developments in econometric modeling
and forecasting
2005
7.93
531
35,40
R
Tussyadiah, I. P,
Park, S.
Consumer evaluation of hotel service robots
2018
1.66
74
37,00
G
Bangwayo-Skeete
P. F., Skeete, R.
W.
Can Google data improve the forecasting
performance of tourist arrivals? Mixed-data
sampling approach
2015
1.5
200
40,00
G
Gunter, U.,
Önder, I.
Forecasting city arrivals with Google Analytics
2016
1.5
72
18,00
R
Tung, V. W. S. T.,
Au, N.
Exploring customer experiences with robotics in
hospitality
2018
1.33
65
32,50
R
Tung, V. W. S. T.,
Law, R.
The potential for tourism and hospitality
experience research in human-robot interactions
2017
0.61
94
31,33
R
Huang, M., Rust,
R. T.
Artificial intelligence in service
2018
0.39
343
171,5
Cl: cluster, BC: betweenness centrality, C: citation, B:blue, R:red, G:green
15
Hereunder Witt and Witt's research upon forecasting tourism
demand through empirical data was the most prominent research with total
1,118 citations and also got the most powerful BC degree (32.71). In the same
group set, Law’s study implying the importance of neural networks in
tourism demand forecasting came in the second place in terms of BC (20.36)
and received 407 citations. In the red group set, studies were conducted
after 2017 and research topics were directly upon artificial intelligence,
robotics. Huang and Rust’s theoretical research received 343 citations in a
short period of time, and this particularly implies that this research will be
efficient in the field. In the green group set, there were studies upon
forecasting tourist behavior through utilizing data sources such as Google,
and co-citation researches were mainly upon forecasting.
Thematic and co-occurrence Analysis
Figure 5 shows a thematic evolution of two different periods. This
progressive illustration is derived from the breakthrough in 2018 (Figure 1).
There were 68 documents analyzed in 2003-2018 period, whereas 34
documents in 2019-2020. Themes are more likely to occur in four areas in a
period of more than a year, thus, interpretation is required due to the high
number of publications after this sudden breakthrough. First period’s
keyword plus number was 478 but second period’s was 178. Among these
keywords to filter the most frequently used ones, minimum 3 occurrence
threshold were preferred. According to Cahlik’s (2000) specification,
concepts emerged at top-right side of the chart are defined as motor themes,
and they’re highly centralized and intense. In other word, these concepts
imply importance for the research field and they simply illustrate the
progress. In period of 2003-2018 forecasting theme’s sub-dimensions were
tourism demand, fuzzy systems, fuzzy inference and time series analysis,
whereas expert systems theme’s sub-dimensions were intelligent agents
and semantics. In 2019-2020 period the emerging theme was Big Data.
The concepts in the bottom-right are highly centralized with low
density, and they’re called as basic and transversal themes. Besides implying
importance for the field, these concepts are in relation to the common
themes that interact with different fields of knowledge. In the period 2003-
2018, under the theme of artificial intelligence, emerging sub-dimensions
were knowledge management, semantic web, e-tourism, www; whereas
under recommender systems theme, electronic commerce, knowledge-
based systems, and intelligent systems emerged. In period of 2019-2020
there were no emerging themes in same theme zone.
16
Figure 5. Thematic Evolution
Concepts emerging in the bottom-left have low centrality and low
density, and they are called emerging or declining themes. These concepts are
considered as underdeveloped and marginal. In 2003-2018 period, tourism
and data mining themes appeared, and the emerging sub-themes of data
mining were learning systems, artificial intelligence techniques, tourism
management, and forecasting method. In 2019-2020 period artificial
intelligence theme emerged with tourism development, robotics,
forecasting method, tourism as its sub-themes. Regarding this period,
themes showed up in a relatively shorter time period. It is considered
17
beneficial to evaluate these emerging themes as influencers of prospective
studies, and how they will change or transform with future studies. The
2019-2020 Thematic Evolution map signifies a breakthrough in terms of AI’s
effect on tourism, and because of its themes are highly mentioned in the
scientific field, it can be interpreted that these are industry’s primary
contemporary demands from AI technologies.
Concepts emerging at the top-left side have low centrality but high
density, and they are called high developed and isolated themes. These notions
constitute highly developed and isolated themes, thus have limited
importance for the research field. User interfaces and decision-making were
emerging themes in 2003-2018 period. User interfaces theme’s sub-
dimension was web services, whereas geographic information systems,
information systems, decision support systems were the sub-dimensions of
decision-making theme. In the period of 2019-2020, there was no emerging
theme in that zone. Mostly, up to the year 2018 forecasting and expert
systems themes were boosting themes, but after 2019 Big Data took that
place. Similarly, until 2018 artificial intelligence theme was dominant in
tourism field, additionally demonstrated a strong cooperation with other
fields of study. After 2019 artificial intelligence theme and its sub-themes
displayed weak progress against Big Data. In this context AI’s effect on
tourism may gain progress regarding its collaboration with Big Data.
Finally, the researchers interested in tourism, AI, and data mining themes
between 2003-2018 also showed interest to AI in 2019-2020 period.
Presented as in Figure 6 the result of the analysis was consisted of six
clusters. Inherently, artificial intelligence keyword had the greatest number
of nodes (27). Secondly, tourism keyword in green cluster was formed by
13 nodes. Table 3 details prominent 30 author keywords in documents,
demonstrating the interactions between keywords and clusters.
18
Figure 6. Co-occurrence author keywords network analysis.
(Note: Visualization was produced in VOSviewer software. Size of a node is proportional to number of
appearances of the keyword, that is, larger the size, higher the occurrence of the papers in authors’ keywords.
The general distance between the nodes provide information about their relationship to each other. The shorter
distance between nodes, the stronger their relationship. The relevance of terms is determined by counting the
number of times terms occur in keywords. Colors are used to distinguish different clusters.)
Table 3. Most common keyword occurrences
R
Keywords
C
Co
Oc
R
Keywords
C
Co
Oc
1
artificial intelligence
1
27
46
16
human-robot interaction
1
4
4
2
tourism
2
13
26
17
recommender system
3
4
4
3
robots
2
10
6
18
tourism marketing
3
4
3
4
big data
3
10
8
19
internet of things
3
4
3
5
machine learning
3
9
13
20
competitiveness
4
4
3
6
forecasting
1
6
6
21
overtourism
4
4
3
7
social media
2
6
3
22
sustainability
4
4
3
8
hospitality
2
6
4
23
tourism demand forecasting
5
4
4
9
personalization
2
6
4
24
information technology
6
4
4
10
service automation
1
5
3
25
review
1
3
3
11
recommender systems
2
5
6
26
marketing
2
3
4
12
automation
2
5
3
27
smart tourism
3
3
8
13
digital economy
4
5
3
28
neural network
5
3
3
14
tourism demand
1
4
5
29
deep learning
5
3
3
15
robotics
1
4
6
30
e-tourism
6
3
5
R: rank, C: cluster, Co: Author keyword co-occurrences links, Oc: Author keyword occurrences
19
CONCLUSION
This bibliometric research provides a systematic overview of AI in tourism
studies. It highlights the scientific proliferation of AI by scanning the most
popular papers, collaborations, research hotspots, and advancements. To
the best of the authors’ knowledge, this current research is among the first
to evaluate and demonstrate the progress of AI in the context of tourism.
Therefore, this study fills this gap by enlightening the prominent aspects of
AI. As AI had a long journey since it was conceptualized by McCarthy et al.
in 1955, it can be said that it has just completed its incubation period and
that it is now ready to transform the society as a game-changer.
This study focuses on AI’s evolution in tourism field, but more
importantly, aims to draw attention to its potential effects on social sciences.
Even though AI is still regarded as a complicated subject, its roots are
embedded in early mathematics, economics, philosophy, and psychology
(Russell & Norvig, 2016). Therefore AI should not be evaluated as mere
mathematical equations regarding computer and data science, but also as
an economic and societal contribution to humankind (Pavaloiu et al., 2017;
Tussyadiah & Miller, 2019). Results of the current study concerning popular
keyword occurrences, support this reflection as there were both numeral
(digital economy, forecasting, big data, etc.) and human-driven
(recommender systems, sustainability, personalization, etc.) keywords
regarding AI.
Due to its interdisciplinary nature, adoption of AI has potential to
drive innovation across sectors and provide social welfare for countries
around the world (Perrault et al., 2019). According to McKinsey Global's
report (Chui et al., 2018), in terms of tourism industry, AI can double what
is achievable using a traditional analytic method(s) and enable a growth
between 7% to 11.6% of total revenue, making tourism and travel industry
the biggest potential beneficiary of AI among industries. Besides, this
study’s results upon most cited papers and co-citation networks
demonstrate that, AI is predominantly being used for forecasting, demand
analysis, and recommender systems. In addition to that, tourism industry
benefits from AI in different settings such as sentiment analysis with
Natural Language Processing, augmented reality, virtual reality, robotics in
hospitality and service, intelligent chatbots etc. AI improves
personalization and accurate recommendations in tourism which is related
to main goals of the industry (Mich, 2020). However, businesses and
industries come across some challenges adopting AI. IBM and O’Reilly’s
20
Report (Thomas, 2019) underlined these challenges and classified them in
five themes as follows:
Lack of Understanding: Businesses should carefully analyze their
needs and problems. They should check the applicability of AI to their
concern. Because of its spreading popularity, there is a misperception that
AI will fix any kind of problem.
Getting a Handle on Data: Lack of data, too much data or bad data are
constraints for businesses in integrating their workflow to AI. For
implementing AI successfully there is a strict need to accurate and good
data.
Lack of Relevant Skills: Skills needed for AI experts are utterly different
than current software engineers. This is a continuous relearning process as
the machine learning algorithm learns from the training data. There is also
a need for skilled AI programmers.
Trust: AI recommendations or decisions should be traceable in order
to ensure businesses to see what their AI is doing. In doing so, businesses
can avoid the risks of bias. Transparency in process is also another
requirement for ethical AI.
Culture and Business Model Change: As AI enables deduction from
unstructured vast amount of data, businesses should adapt their systems
with new technologies AI brings in.
To overcome these challenges, IBM and O’Reilly (Thomas, 2019)
propose a guiding strategy, called the AI Ladder, which suggests
operationalizing AI throughout the business (infuse), building and scaling
AI with trust and transparency (analyze), creating a business-ready analytics
foundation (organize), making data simple and accessible (collect). Similarly,
Samara et al. (2020) conducted a broad literature review and summarized
AI challenges in tourism as; technical challenges, financial and business
challenges, regulatory challenges, and socio-ethical challenges. Technical,
financial, and business challenges refer to data quality and accuracy,
ensuring lack of bias, and cost concerns. Regulatory issues imply the way
data is collected and processed, referring to the role of governments
regarding the safety and privacy of tourism businesses Big Data. Socio-
ethical challenges are comprised of acceptance of AI in routine of tourism
and the fear of job losses.
Briefly, these concerns regarding AI’s implementation to businesses
remind the progress and misperceptions of e-tourism along with 2000’s.
21
Just as in e-tourism, AI systems are already being used in tourism industry
for a while without realizing these are reflections of AI. AI’s effect on
automation, rule-based jobs, and auto-tasks are inevitable. But if the
industry manages AI properly, it will augment the jobs rather than
eliminating them, and it will bring new opportunities and businesses
altogether. Besides, interaction between AI systems and tourism and travel
industry largely depends on tourism professionals’ skills, thus, human
workforce will remain valuable and essential in conducting a healthy AI-
industry interaction (Cain et al., 2019). Moreover, developing countries may
operate AI systems without having large industrial networks, and gain a
competitive advantage by utilizing these in tourism context. Throughout its
effect on decision-making process, AI can be useful in terms of
underdeveloped and developing touristic destinations as it can assist and
ease tourist decisions and recommendations.
By its very nature, tourism industry is fragile to local or global risks
and complexities. These complexities can either be human-made disasters,
natural catastrophes, or global epidemics such as SARS, COVID-19 viruses.
Further to that, Gretzel et al. (2020) called for transformative research and
argued that COVID-19 may act as a breaking point, challenging current
paradigms, just as Kuhn (1962) articulated in ‘The structure of scientific
revolutions’. COVID-19 is changing conditions rapidly nowadays, and
therefore exhibits a powerful uncertainty. For example, a very small change
in one parameter (e.g., length of lockdowns, travel restrictions) might create
very different outcome on many variables (Zenker & Kock, 2020).
Accordingly, Pappas (2019) asserts that because of its mostly reductionist
approach, tourism and travel research paid less attention to chaos and
complexity theories. In doing so, he assumes tourist decision-making
processes as complex patterns, and suggests that complexity cognizance can
help understanding rapidly changing dynamics. Therefore, it is suggested
that AI techniques can be applied to diverse complex problems (Corchado
& Lees, 1998), herein particularly chaotic problems of tourism industry. In
this study, it is suggested that AI tools (e.g., machine learning, neural
networks, deep learning, natural language processing) may broaden
tourism industry’s perspectives to contemporary problems without the
restriction of traditional methods.
However, integrating AI into tourism realm is a nuanced
phenomenon. Tussyadiah (2020) points out application of AI and intelligent
automation in tourism and travel industry is expected to increase in near
future. Therefore, she sheds light on AI-tourism relationship and suggests
a guideline for future researches:
22
Designing Beneficial Artificial Intelligence: AI systems should be
designed and developed to enhance tourism experiences by intelligent
automation. This relationship can be considered as a mutual relationship
that both parties interact due to the progress between each other. Technical
issues such as privacy of tourists’ personal data, eliminating bias, bugs,
cyber-attacks, and other security concerns appear in designing and
implementing beneficial artificial intelligence.
Facilitating Adoption: AI brings some acceptance concerns to tourism
field. Tourism businesses, employees and tourists’ attitude towards
technology will shape this adoption, thus, barriers to adoption should be
carefully understood and facilitators should be encouraged. This notion is
evaluated in a broader viewpoint in Ivanov & Webster’s (2017) study.
Authors discussed adoption process in scope of robots, AI, and service
automation in tourism and travel industry. They focused on costs of AI and
implied that company characteristics and culture, technology costs, degree
of technological complexity, customer’s attitudes and characteristics, and
safety characteristics affect cost side of AI adoption.
Assessing the Impacts of Intelligent Automation in Tourism: Positive and
negative effects of AI need to be deliberately evaluated with respect to host
community, tourists, and tourism professionals, namely the industry.
Dynamics of AI integrated destination and community may be rapidly
changed, so the ratio of labor-automation should be carefully planned.
Ethical concerns also arise in terms of human-robot interaction; thus,
probable harms of intelligent systems must be minimized.
Creating a Sustainable Future: AI systems and intelligent automation
should prevent prospective future problems of tourism. Along with
governmental policy support, AI systems can be designed to reduce the
negative effects of automation in industry; and provide a sustainable
development through tourism. Since intelligent automation may diminish
socialization between tourism partners (e.g., tourist, employee), beneficial
AI implementation acts a vital role in maintaining human values and
responsible use behaviors among partners. Regarding AI-tourism
relationship, both technical and social aspects are critical to create a
sustainable future.
This study aims to contribute to scientific field of AI in tourism
context by providing the hotspots and progress, and furthermore highlights
the importance of AI for changing tourism complexities. Focusing on its
interdisciplinary characteristic, AI can be an effective tool for tourism
23
stakeholders (e.g., tourist, tourism employees, destinations, governance
actors) in adaption of new solutions to contemporary concerns.
Finally, this bibliometric study has some shortcomings. First, the
current study was conducted on documents incorporated in Scopus
database, hence, future studies could use Google Scholar as a data collection
database. Second, the data source of this study was limited to only tourism-
related documents. Future research could examine the progress of AI in
other fields or apply inclusive bibliometrics to different disciplines to
review the evolution. Considering the limitations of bibliometrics,
systematic reviews and content analyses of most cited papers can be
conducted to gain deeper understanding of AI in different fields. Lastly, this
study was conducted upon keywords. Therefore, conducting different
bibliometric techniques in other languages could provide a valuable
evaluation upon AI.
REFERENCES
Ajiferuke, I., Burell, Q., & Tague, J. (1988). Collaborative coefficient: A single measure of
the degree of collaboration in research. Scientometrics, 14(5–6), 421–433.
Akehurst, G. (2009). User generated content: The use of blogs for tourism organisations
and tourism consumers. Service Business, 3(1), 51–61.
https://doi.org/10.1007/s11628-008-0054-2
Andreu, L., Bigne, E., Amaro, S., & Palomo, J. (2020). Airbnb research: an analysis in
tourism and hospitality journals. International Journal of Culture, Tourism, and
Hospitality Research, 14(1), 2–20. https://doi.org/10.1108/IJCTHR-06-2019-0113
Aria, M., & Cuccurullo, C. (2017). Bibliometrix: An R-tool for comprehensive science
mapping analysis. Journal of Informetrics, 11(4), 959–975.
https://doi.org/10.1016/j.joi.2017.08.007
Askun, V., & Cizel, R. (2019). Kompleks problem çözme üzerine R programı ile bir
bibliyometrik analiz. Mediterranean Journal of Humanities, 9(1), 37–47.
https://doi.org/10.13114/mjh.2019.445
Askun, V., & Cizel, R. (2020). Twenty years of research on mixed methods. Journal of Mixed
Methods Research, 1(1), 28–43. https://doi.org/10.14689/jomes.2020.1.2
Barrios, M., Borrego, A., Vilaginés, A., Ollé, C., & Somoza, M. (2008). A bibliometric study
of psychological research on tourism. Scientometrics, 77(3), 453–467.
https://doi.org/10.1007/s11192-007-1952-0
Benckendorff, P. (2009). Themes and trends in Australian and New Zealand tourism
research: A social network analysis of citations in two leading journals (1994-2007).
Journal of Hospitality and Tourism Management, 16(1), 1–15.
https://doi.org/10.1375/jhtm.16.1.1
Benckendorff, P., & Zehrer, A. (2013). A network analysis of tourism research. Annals of
Tourism Research, 43, 121–149. https://doi.org/10.1016/j.annals.2013.04.005
Börner, K., Chen, C., & Boyack, K. W. (2005). Visualizing knowledge domains. Annual
Review of Information Science and Technology, 37, 179–255.
https://doi.org/10.1002/aris.1440370106
Borràs, J., Moreno, A., & Valls, A. (2014). Intelligent tourism recommender systems: A
24
survey. Expert Systems with Applications, 41, 7370–7389.
https://doi.org/10.1016/j.eswa.2014.06.007
Bowen, J., & Whalen, E. (2017). Trends that are changing travel and tourism. Worldwide
Hospitality and Tourism Themes, 9(6), 592–602. https://doi.org/10.1108/WHATT-09-
2017-0045
Buhalis, D. (2003). eTourism: Information technology for strategic tourism management. London:
Pearson Education.
Buhalis, D., Harwood, T., Bogicevic, V., Viglia, G., Beldona, S., & Hofacker, C. (2019).
Technological disruptions in services: Lessons from tourism and hospitality.
Journal of Service Management, 30(4), 484–506. https://doi.org/10.1108/JOSM-12-
2018-0398
Buhalis, D., & Sinarta, Y. (2019). Real-time co-creation and nowness service: Lessons from
tourism and hospitality. Journal of Travel & Tourism Marketing, 36(5), 563–582.
https://doi.org/10.1080/10548408.2019.1592059
Cahlik, T. (2000). Search for fundamental articles in economics. Scientometrics, 49(3), 389–
402.
Cain, L. N., Thomas, J. H., & Alonso, M. (2019). From sci-fi to sci-fact: The state of robotics
and AI in the hospitality industry. Journal of Hospitality and Tourism Technology,
10(4), 624–650. https://doi.org/10.1108/JHTT-07-2018-0066
Casteleiro-Roca, J.-L., Gomez-Gonzalez, J. F., Calvo-Rolle, J. L., Jove, E., Quintian, H.,
Acosta Martin, J. F., … Mendez-Perez, J. A. (2018). Prediction of the energy demand
of a hotel using an artificial intelligence-based model. In J. F. de Cos Juez, J. R.
Villar, E. A. de la Cal & A. Herrero (Eds.), Hybrid Artificial Intelligent Systems (Vol.
1, pp. 586–596). Cham: Springer. https://doi.org/10.1007/978-3-319-92639-1
Chiu, W. T., & Ho, Y. S. (2007). Bibliometric analysis of tsunami research. Scientometrics,
73(1), 3–17. https://doi.org/10.1007/s11192-005-1523-1
Cho, V. (2003). A comparison of three different approaches to tourist arrival forecasting.
Tourism Management, 24(3), 323–330. https://doi.org/10.1016/S0261-5177(02)00068-7
Chui, M., Manyika, J., Miremadi, M., Henke, N., Chung, R., Nel, P., & Malhotra, S. (2018).
Notes from the AI frontier: Insights from hundreds of use cases. In McKinsey&
Company. Retrieved September 10, 2020, from
https://www.mckinsey.com/~/media/mckinsey/featured%20insights/artificial%20
intelligence/notes%20from%20the%20ai%20frontier%20applications%20and%20v
alue%20of%20deep%20learning/notes-from-the-ai-frontier-insights-from-
hundreds-of-use-cases-discussion-paper.ashx.
Cobo, M. J., López-Herrera, A. G., Herrera-Viedma, E., & Herrera, F. (2011). Science
mapping software tools: Review, analysis, and cooperative study among tools.
Journal of the American Society for Information Science and Technology, 62(7), 1382–
1402. https://doi.org/10.1002/asi.21525
Comerio, N., & Strozzi, F. (2019). Tourism and its economic impact: A literature review
using bibliometric tools. Tourism Economics, 25(1), 109–131.
https://doi.org/10.1177/1354816618793762
Corchado, J. M., & Lees, B. (1998). Cognitive models for integrating artificial intelligence
approaches. AII Workshop on Knowledge Discovery. Glasgow, UK.
Cunill, O. M., Salvá, A. S., Gonzalez, L. O., & Mulet-Forteza, C. (2019). Thirty-fifth
anniversary of the International Journal of Hospitality Management: A
bibliometric overview. International Journal of Hospitality Management, 78, 89–101.
https://doi.org/10.1016/j.ijhm.2018.10.013
Dhamija, P., & Bag, S. (2020). Role of artificial intelligence in operations environment: A
25
review and bibliometric analysis. TQM Journal, 32(4), 869–896.
https://doi.org/10.1108/TQM-10-2019-0243
Ferràs, X., Hitchen, E. L., Tarrats-Pons, E., & Arimany-Serrat, N. (2020). Smart tourism
empowered by artificial intelligence: The case of Lanzarote. Journal of Cases on
Information Technology, 22(1), 1–13. https://doi.org/10.4018/JCIT.2020010101
Freeman, L. C. (1977). A set of measures of centrality based on betweenness. Sociometry,
40(1), 35- 41. https://doi.org/10.2307/3033543
Gajdošík, T., & Marciš, M. (2019). Artificial intelligence tools for smart tourism
development. In R. Silhavy (Ed.), Artificial Intelligence Methods in Intelligent
Algorithms (Vol. 985, pp. 392–402). Cham: Springer. https://doi.org/10.1007/978-3-
030-19810-7_39
García-Lillo, F., Úbeda-García, M., & Marco-Lajara, B. (2016). The intellectual structure of
research in hospitality management: A literature review using bibliometric
methods of the journal International Journal of Hospitality Management.
International Journal of Hospitality Management, 52, 121–130.
https://doi.org/10.1016/j.ijhm.2015.10.007
Glänzel, W., & Schubert, A. (2005). Analysing scientifc networks through co-authorship. In
H. F. Moed, W. Glänzel, & U. Schmoch (Eds.), Handbook of quantitative science and
technology research (pp. 257–276). Berlin: Springer.
González-Rodríguez, M. R., Díaz-Fernández, M. C., & Pacheco Gómez, C. (2020). Facial-
expression recognition: An emergent approach to the measurement of tourist
satisfaction through emotions. Telematics and Informatics, 51.
https://doi.org/10.1016/j.tele.2020.101404
Gretzel, U. (2011). Intelligent systems in tourism. A social science perspective. Annals of
Tourism Research, 38(3), 757–779. https://doi.org/10.1016/j.annals.2011.04.014
Gretzel, U., Fesenmaier, D. R., & O’Leary, J. T. (2006). The transformation of consumer
behaviour. In D. Buhalis & C. Costa (Eds.), Tourism Business Frontiers: Consumers,
Products and Industry (pp. 9–18). Oxford: Elsevier. https://doi.org/10.1016/b978-0-
7506-6377-9.50009-2
Gretzel, U., Fuchs, M., Baggio, R., Hoepken, W., Law, R., Neidhardt, J., … Xiang, Z. (2020).
e-Tourism beyond COVID-19: a call for transformative research. Information
Technology and Tourism, 22(2), 187–203. https://doi.org/10.1007/s40558-020-00181-3
Gunbayi, I., & Sorm, S. (2018). Social paradigms in guiding social research design: The
functional, interpretive, radical humanist and radical structural paradigms.
International Journal on New Trends in Education and Their Implications, 9(2), 57-76.
Guns, R., Liu, Y. X., & Mahbuba, D. (2011). Q-measures and betweenness centrality in a
collaboration network: A case study of the field of informetrics. Scientometrics,
87(1), 133–147. https://doi.org/10.1007/s11192-010-0332-3
Guzeller, C. O., & Celiker, N. (2019). Bibliometrical analysis of Asia Pacific Journal of
Tourism Research. Asia Pacific Journal of Tourism Research, 24(1), 108–120.
https://doi.org/10.1080/10941665.2018.1541182
Güzeller, C. O., & Çeli̇ker, N. (2018). Bibliometric analysis of tourism research for the
period 2007-2016. Advances in Hospitality and Tourism Research, 6(1), 1–22.
https://doi.org/10.30519/ahtr.446248
Hadavandi, E., Ghanbari, A., Shahanaghi, K., & Abbasian-Naghneh, S. (2011). Tourist
arrival forecasting by evolutionary fuzzy systems. Tourism Management, 32(5),
1196–1203. https://doi.org/10.1016/j.tourman.2010.09.015
Inanc-Demir, M., & Kozak, M. (2019). Big data and its supporting elements: implications
for tourism and hospitality marketing. In M. Sigala, R. Rahimi, & M. Thelwall
26
(Eds.), Big Data and Innovation in Tourism, Travel, and Hospitality: Managerial
Approaches, Techniques, and Applications (pp. 213-223). Singapore: Springer.
https://doi.org/10.1007/978-981-13-6339-9
Ivanov, S., & Webster, C. (2017). Adoption of robots, artificial intelligence and service
automation by travel, tourism and hospitality companies – a cost-benefit analysis.
In V. Marinov, M. Vodenska, M. Assenova & E. Dogramadjieva (Eds.) Traditions
and Innovations in Contemporary Tourism (pp. 190-203). UK: Cambridge Scholars
Publishing.
Johnson, A. G., & Samakovlis, I. (2019). A bibliometric analysis of knowledge development
in smart tourism research. Journal of Hospitality and Tourism Technology, 10(4), 600–
623. https://doi.org/10.1108/JHTT-07-2018-0065
Kazak, A. N., Chetyrbok, P. V., & Oleinikov, N. N. (2020). Artificial intelligence in the
tourism sphere. IOP Conference Series: Earth and Environmental Science, 421(4).
https://doi.org/10.1088/1755-1315/421/4/042020
Kirilenko, A. P., Stepchenkova, S. O., Kim, H., & Li, X. (Robert). (2018). Automated
sentiment analysis in tourism: Comparison of approaches. Journal of Travel
Research, 57(8), 1012–1025. https://doi.org/10.1177/0047287517729757
Köseoglu, M. A., Okumus, F., Putra, E. D., Yildiz, M., & Dogan, I. C. (2018). Authorship
trends, collaboration patterns, and co-authorship networks in lodging studies
(1990–2016). Journal of Hospitality Marketing and Management, 27(5), 561–582.
https://doi.org/10.1080/19368623.2018.1399192
Koseoglu, M. A., Rahimi, R., Okumus, F., & Liu, J. (2016). Bibliometric studies in tourism.
Annals of Tourism Research, 61, 180–198. https://doi.org/10.1016/j.annals.2016.10.006
Kuhn, T. S. (1962). The structure of scientific revolutions. Chicago: University of Chicago
press.
Law, R. (1998). Room occupancy rate forecasting: A neural network approach. International
Journal of Contemporary Hospitality Management, 10(6), 234–239.
https://doi.org/10.1108/09596119810232301
Law, R. (2000). Back-propagation learning in improving the accuracy of neural network-
based tourism demand forecasting. Tourism Management, 21(4), 331–340.
https://doi.org/10.1016/S0261-5177(99)00067-9
Law, R., Leung, R., & Buhalis, D. (2010). An analysis of academic leadership in hospitality
and tourism journals. Journal of Hospitality and Tourism Research, 34(4), 455–477.
https://doi.org/10.1177/1096348010370866
Lei, Y., & Liu, Z. (2019). The development of artificial intelligence: A bibliometric analysis,
2007-2016. Journal of Physics: Conference Series, 1168(2). https://doi.org/10.1088/1742-
6596/1168/2/022027
Leung, X. Y., Sun, J., & Bai, B. (2017). Bibliometrics of social media research: A co-citation
and co-word analysis. International Journal of Hospitality Management, 66, 35–45.
https://doi.org/10.1016/j.ijhm.2017.06.012
Littell, J. H., Corcoran, J., & Pillai, V. (2008). Systematic reviews and meta-analysis. New York:
Oxford University Press.
McCarthy, J., Minksy, M., Rochester, L., & Shannon, C. E. (1955). A proposal for the
Dartmouth summer research project on artificial intelligence. AI magazine, 27(4),
12-12. Retrieved December 5, 2020, from http://www-
formal.stanford.edu/jmc/history/dartmouth.pdf
Merigó, J. M., Mulet-Forteza, C., Valencia, C., & Lew, A. A. (2019). Twenty years of Tourism
Geographies: A bibliometric overview. Tourism Geographies, 21(5), 881–910.
https://doi.org/10.1080/14616688.2019.1666913
27
Mich, L. (2020). Artificial intelligence and machine learning. In Z. Xiang, M. Fuchs, U.
Gretzel, & W. Höpken (Eds.), Handbook of e-Tourism. Springer Nature Switzerland.
https://doi.org/10.1007/978-3-030-05324-6_25-1
Niu, J., Tang, W., Xu, F., Zhou, X., & Song, Y. (2016). Global research on artificial
intelligence from 1990-2014: Spatially-explicit bibliometric analysis. ISPRS
International Journal of Geo-Information, 5(5), 1–19.
https://doi.org/10.3390/ijgi5050066
Norris, M., & Oppenheim, C. (2007). Comparing alternatives to the Web of Science for
coverage of the social sciences’ literature. Journal of Informetrics, 1(2), 161–169.
https://doi.org/10.1016/j.joi.2006.12.001
Núñez-Tabales, J. M., Solano-Sánchez, M. Á., & Caridad-y-López-del-Río, L. (2020). Ten
years of Airbnb phenomenon research: A bibliometric approach (2010–2019).
Sustainability, 12(15). https://doi.org/10.3390/su12156205
Okumus, B., Koseoglu, M. A., & Ma, F. (2018). Food and gastronomy research in tourism
and hospitality: A bibliometric analysis. International Journal of Hospitality
Management, 73, 64–74. https://doi.org/10.1016/j.ijhm.2018.01.020
Okumus, F., Köseoglu, M. A., Putra, E. D., Dogan, I. C., & Yildiz, M. (2019). A bibliometric
analysis of lodging-context research from 1990 to 2016. Journal of Hospitality and
Tourism Research, 43(2), 210–225. https://doi.org/10.1177/1096348018765321
Palys, T. (2008). Purposive sampling. In L. M. Given (Ed.), The Sage encyclopedia of qualitative
research methods (pp. 697–698). California: Sage.
Pannu, A. (2015). Artificial intelligence and its application in different areas. International
Journal of Engineering and Innovative Technology, 4(10), 79–84.
Pappas, N. (2019). UK outbound travel and Brexit complexity. Tourism Management, 72, 12–
22. https://doi.org/10.1016/j.tourman.2018.11.004
Pavaloiu, A., Köse, U., & Boz, H. (2017). How to apply artificial intelligence in social
sciences. IASOS - Congress of International Applied Social Sciences, (September). Uşak,
Turkey. Retrieved September 10, 2020, from
https://www.researchgate.net/publication/325398286_How_to_Apply_Artificial_I
ntelligence_in_Social_Sciences
Perrault, R., Shoham, Y., Brynjolfsson, E., Clark, J., Etchemendy, J., Grosz Harvard, B., …
Mishra, S. (2019). The AI Index 2019 Annual Report. In AI Index Steering Committee,
Human-Centered AI Institute. Stanford. Retrieved September 10, 2020, from
https://hai.stanford.edu/sites/g/files/sbiybj10986/f/ai_index_2019_report.pdf
Pritchard, A. (1969). Statistical bibliography or bibliometrics. Journal of Documentation,
25(4), 348–349.
Qian, J., Law, R., & Wei, J. (2019). Knowledge mapping in travel website studies: A
scientometric review. Scandinavian Journal of Hospitality and Tourism, 19(2), 192–209.
https://doi.org/10.1080/15022250.2018.1526113
Ritchie, B. W. (2004). Chaos, crises and disasters: A strategic approach to crisis management
in the tourism industry. Tourism Management, 25(6), 669–683.
https://doi.org/10.1016/j.tourman.2003.09.004
Ruhanen, L., Weiler, B., Moyle, B. D., & McLennan, C. J. (2015). Trends and patterns in
sustainable tourism research: A 25-year bibliometric analysis. Journal of Sustainable
Tourism, 23(4), 517–535. https://doi.org/10.1080/09669582.2014.978790
Ruiz-Real, J. L., Uribe-Toril, J., Valenciano, J. de P., & Gázquez-Abad, J. C. (2020). Rural
tourism and development: Evolution in scientific literature and trends. Journal of
Hospitality and Tourism Research, 1–25. https://doi.org/10.1177/1096348020926538
Russell, S. J., & Norvig, P. (2016). Artificial intelligence: A modern approach (3rd ed.). Harlow:
28
Pearson Education.
Samara, D., Magnisalis, I., & Peristeras, V. (2020). Artificial intelligence and big data in
tourism: A systematic literature review. Journal of Hospitality and Tourism
Technology, 11(2), 343-367. https://doi.org/10.1108/JHTT-12-2018-0118
Sánchez, A. D., de la Cruz Del Río Rama, M., & García, J. Á. (2017). Bibliometric analysis
of publications on wine tourism in the databases Scopus and WoS. European
Research on Management and Business Economics, 23(1), 8–15.
https://doi.org/10.1016/j.iedeen.2016.02.001
Shukla, A. K., Janmaijaya, M., Abraham, A., & Muhuri, P. K. (2019). Engineering
applications of artificial intelligence: A bibliometric analysis of 30 years (1988–
2018). Engineering Applications of Artificial Intelligence, 85, 517–532.
https://doi.org/10.1016/j.engappai.2019.06.010
Singh, S. K., Rathore, S., & Park, J. H. (2020). BlockIoTIntelligence: A blockchain-enabled
intelligent IoT architecture with artificial intelligence. Future Generation Computer
Systems, 110, 721–743. https://doi.org/10.1016/j.future.2019.09.002
Song, H., Qiu, R. T. R., & Park, J. (2019). A review of research on tourism demand
forecasting. Annals of Tourism Research, 75, 338–362.
https://doi.org/10.1016/j.annals.2018.12.001
Teixeira, S. J., & Ferreira, J. J. D. M. (2018). A bibliometric study of regional competitiveness
and tourism innovation. International Journal of Tourism Policy, 8(3), 214–243.
https://doi.org/10.1504/IJTP.2018.094483
Thomas, R. (2019). The AI ladder: Demystifying AI challenges. In: IBM and O’Reilly.
Retrieved September 10, 2020, from https://www.oreilly.com/online-
learning/report/The-AI-Ladder.pdf.
Todeschini, R., & Baccini, A. (2016). Handbook of bibliometric indicators: Quantitative tools for
studying and evaluating research. Weinheim, Germany: Wiley-VCH.
Topal, I., & Uçar, M. K. (2018). In tourism, using artificial intelligence forecasting with
Tripadvisor data: Year of Turkey in China. International Conference on Artificial
Intelligence and Data Processing (IDAP), 1–5. Retrieved February 15, 2020, from
https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8620874
Tran, B., Vu, G., Ha, G., Vuong, Q.-H., Ho, M.-T., Vuong, T.-T., … Ho, R. (2019). Global
evolution of research in artificial intelligence in health and medicine: A
bibliometric study. Journal of Clinical Medicine, 8(360).
https://doi.org/10.3390/jcm8030360
Tussyadiah, I. (2020). A review of research into automation in tourism: Launching the
Annals of Tourism Research Curated Collection on artificial intelligence and
robotics in tourism. Annals of Tourism Research, 81, 102883.
https://doi.org/10.1016/j.annals.2020.102883
Tussyadiah, I., & Miller, G. (2019). Perceived impacts of artificial intelligence and responses
to positive behaviour change intervention. In Information and Communication
Technologies in Tourism 2019 (pp. 359–370). Springer International Publishing.
https://doi.org/10.1007/978-3-030-05940-8
van Nunen, K., Li, J., Reniers, G., & Ponnet, K. (2018). Bibliometric analysis of safety culture
research. Safety Science, 108, 248–258. https://doi.org/10.1016/j.ssci.2017.08.011
van Raan, A. F. J. (2003). The use of bibliometric analysis in research performance
assessment and monitoring of interdisciplinary scientific developments. TATuP -
Zeitschrift Für Technikfolgenabschätzung in Theorie Und Praxis, 12(1), 20–29.
https://doi.org/10.14512/tatup.12.1.20
Virani, A., Wellstead, A., & Howlett, M. P. (2019). Where is the policy? A bibliometric
29
review of the state of policy research on medical tourism. Global Health Research and
Policy, 5, 1–16. https://doi.org/10.2139/ssrn.3445235
Volchek, K., Liu, A., Song, H., & Buhalis, D. (2019). Forecasting tourist arrivals at
attractions: Search engine empowered methodologies. Tourism Economics, 25(3),
425-447. https://doi.org/10.1177%2F1354816618811558
Zenker, S., & Kock, F. (2020). The coronavirus pandemic-A critical discussion of a tourism
research agenda. Tourism Management, 81, 104164.
https://doi.org/10.1016/j.tourman.2020.104164
Zhang, B., Li, N., Shi, F., & Law, R. (2020). A deep learning approach for daily tourist flow
forecasting with consumer search data. Asia Pacific Journal of Tourism Research,
25(3), 323–339. https://doi.org/10.1080/10941665.2019.1709876
Zheng, W., Liao, Z., & Lin, Z. (2020). Navigating through the complex transport system: A
heuristic approach for city tourism recommendation. Tourism Management, 81,
104162. https://doi.org/10.1016/j.tourman.2020.104162
Zlatanov, S., & Popesku, J. (2019). Current applications of artificial intelligence in tourism
and hospitality. In International Scientific Conference on Information Technology and
Data Related Research (pp. 84-90), Sinteza, Romania.
Zupic, I., & Čater, T. (2015). Bibliometric methods in management and organization.
Organizational Research Methods, 18(3), 429–472.
https://doi.org/10.1177/1094428114562629
