sustainability

Article
Understanding Customer Experience and Satisfaction
through Airline Passengers’ Online Review
Hyun-Jeong Ban and Hak-Seon Kim *
School of Hospitality and Tourism Management, Kyungsung University, 309 Suyoungro, Nam-Gu,
Busan 48434, Korea
* Correspondence: kims@ks.ac.kr; Tel.: +82-51-663-4473




Received: 8 June 2019; Accepted: 18 July 2019; Published: 27 July 2019


Abstract: This study was conducted to understand customer experience and satisfaction through
airline passengers’ online review. To achieve the purpose of this study, the semantic network
analysis was conducted qualitatively by collecting reviews in top 10 airlines selected by Skytrax
(airlinequality.com). In addition, this study quantitatively identified the relationship among six
evaluation factors (seat comfort, staff, food and beverage (F&B), entertainment, ground service,
and value for money), customer satisfaction and recommendation. This study collected 9632 reviews
from the Skytrax. Through a CONCOR (CONvergence of iterated CORrelation) analysis, keywords
were grouped into six clusters (seat comfort, staff, entertainment, ground service, value for money,
and airline brand). Through the linear regression analysis, all evaluation factors except ‘entertainment’
factor significantly had impact on customer satisfaction and recommendation. These results showed
that understanding online review can provide both academic implication and practical implication to
develop sustainable strategy in the airline industry.

Keywords: customer experience; customer satisfaction; online review; skytrax; big data; semantic
network analysis

1. Introduction
Due to fierce competition in the airline industry, the airline company needs to focus on the
passenger’s experience and satisfaction [1]. Customer feedback, in particular, is critical since it is
an outcome measurement for business performance [2]. According to the international air transport
association (IATA) [3], numbers of airline passengers were increasing by about 7% every year since
2015. However, the net profit per airline passenger was decreasing by $10 for 2015, $9 for 2016 and 2017,
and it was estimated at only $7.4 for 2018. This is mainly due to intense competition, and also airline
costs have been rising recently. The major expenses that affect companies in the airline industry are
labor, fuel and other maintenance costs. The airline industry continues to be competitive, even though
many people are traveling by aircraft. The Internet has also created greater price transparency, reducing
margins [4].
Many studies have employed survey methods to measure service quality in the airline
industry [2,5–9]. However, a few recent studies have highlighted the advantages of analyzing online
review data for studying customers’ satisfaction or their experience of the airline [10,11]. Online reviews
are critical since it is a significant source for business growth, performance and improvement of
customer experience, and allow airline companies to conduct two-way communication with airline
passengers [12]. Moreover, electronic word of mouth (eWOM) shared by other airline passengers are
considered trustworthy, fast and widespread [13].
In the previous study, the service quality of airline passenger has been measured in various ways.
Elliott and Roach [5] used on-time performance, baggage handling, food quality, seat comfort, check-in

Sustainability 2019, 11, 4066; doi:10.3390/su11154066 www.mdpi.com/journal/sustainability

Sustainability 2019, 11, 4066 2 of 17

service, and in-flight service as the criteria for evaluating airline service quality. Aksoy et al. [6]
explored the differences in consumer expectations of airline services between passengers on the
Turkish domestic airline and those on four foreign airlines on the same routes. They found that
service expectations differed between the two groups. Gilbert and Wong [7] developed a 26-items
questionnaire incorporating reliability, assurance, facilities, employees, flight patterns, customization,
and responsiveness dimensions to measure and compare the differences in passengers’ expectations of
the desired airline service quality. Significant differences were found among passengers from different
ethnic groups and among passengers who travel for different purposes. However, there were limited
studies on the understanding experience and satisfaction of airline passengers using both qualitative
and quantitative methods to analyze over 9000 online reviews.
The main contribution of this study is the understanding of customer experience and satisfaction
through the airline passengers’ online review. In order to reach the purpose, large amounts of customer
reviews were collected from Skytrax (airlinequality.com). The analysis can be divided into two parts.
One was to analyze the meaning of words extracted from the review data using the semantic network
analysis by qualitative analysis. The other was conducted using the quantitative analysis method to
understand relationships among six evaluation factors, customer satisfaction and recommendation.

2. Literature Review

2.1. Customer Experience, Satisfaction and Online Review

Customer satisfaction is a complex customer experience in the service industry, and can be defined
as an evaluation on which the customers have experienced [14]. Understanding what consumers
expect from a service industry is important in order to provide a standard of comparison against
which consumers judge an organization’s performance regarding the expectation [15]. Service quality
can be defined as a consumer’s overall impression of the relative efficiency of the organization [8].
In addition, customer satisfaction can be defined as experience made on the basis of a specific service
encounter, and it is contributed to customer loyalty, repeat purchase, favorable word-of-mouth (WOM),
and ultimately higher profitability [16].
The customer sets expectations for the product or service and these expectations are becoming
the standard before purchasing. Once the product or service is used, the outcomes or perceptions
are compared to pre-purchase expectations. Consumer generated content contains a variety of media
forms and types [17]. Online reviews that reflect how customers explain and share their experiences in
various forms are a valuable way of figuring out what customers think, and online platforms allow
customers to share experience with information, opinions, and knowledge about products, services
and brands [18]. Customers seek out a variety of information to be confident of their choices, thereby
reducing the perceived risk [19]. Therefore, in this study, data was collected through the online review
written by those who have already experienced it.
Due to the advance of technology, it is easy for customers to post their experience with products
and services on the website [20]. It is especially relevant for service industries because of intangible
characteristics of services. Many studies have demonstrated the strong impact of online customer
reviews. For example, Dellarocas et al. [21] have demonstrated that online review metrics can accurately
forecast movie revenue. Minnema et al. [22] have demonstrated that product returns have a strong
relationship with online customer reviews and the effect of it needs to be considered. The number
of reviews has grown exponentially over the past few decades, and the content of the reviews has
had a significant impact on the repurchase of products [23]. Sotiriadis and van Zyl [24] found that
online reviews and recommendations affect the decision making process of tourists towards tourism
services and WOM has a significant impact on the subjective norms and attitudes towards an airline,
and a customer’s willingness to recommend. Therefore, the online review would be very useful for
airlines to understand their diverse customer base in order to take service improvement strategies
since airlines are inherently multicultural businesses.
Sustainability 2019, 11, 4066 3 of 17

2.2. Skytrax
Skytrax is an airline quality assessment website that performs an online assessment after the
customer directly used each airline [1]. Skytrax has worked for over 150 airlines across the globe,
from the world’s largest airlines through to small domestic carriers and it is a world-recognized brand
that provides professional audit and service benchmarking programs for airlines on product and
service quality. They employ professional auditors to assess the quality of the work done in an airline,
both onboard and in the airport terminals. These evaluations are based on consistent standards [9].
The best airlines in the world highly recognize these quality awards presented by the Skytrax.
When an airline is awarded a ‘Skytrax star-ranking’ or advances to a higher ranking, they immediately
announce this news by publishing press releases and posting it on their websites’ most visible spots.
Both overall star rankings and detailed quality assessment results are publicly available on the Skytrax
website [25]. This data has also been used in various academic studies as shown in Table 1.

Table 1. Academic studies using Skytrax (airlinequality.com).

Author Year Title

Relationship between passengers’ satisfaction and service
Adeniran, A., Fadare, S. O. [26] 2018 quality in Murtala Muhammed international airport,
Lagos, Nigeria
Jeong, E. Y. [27] 2017 Analyze of airlines online-reviews: Focusing on Skytrax
A critical evaluation of US airlines’ service quality
Yayla-Kullu, H. M., Tansitpong, P. [28] 2013
performance: Lower costs compared to satisfied customers
Pérezgonzález, J. D., Gilbey, A. [29] 2011 Predicting Skytrax airport rankings from customer reviews
Lohmann, G., Albers, S., Koch, B., From hub to tourist destination—An explorative study of
2009
Pavlovich, K. [30] Singapore and Dubai’s aviation-based transformation
Towards a means of consistently comparing airline
Mason, K. J., Morrison, W. G. [31] 2008 business models with an application to the ‘low cost’
airline sector
The effects of individual dimensions of airline service
Park, J. W., Robertson, R., Wu, C. L. [32] 2006
quality: Findings from Australian domestic air passengers
Competitive advantage of low-cost carriers: Some
Gillen, D., Lall, A. [33] 2004
implications for airports

2.3. Big Data Analysis

Big Data represents a new era in data exploration and utilization [34]. This is occurring because of
new sources of data, and since the very beginning of the Internet, users have been keeping generating
data on the Internet. The big data intensifies the need for sophisticated statistics and analytical skills.
The big data technologies are providing unprecedented opportunities for statistical inference on
massive analysis, but they also bring new challenges to be addressed, especially when compared to the
analysis of carefully collected smaller data sets [16].
A semantic network analysis is becoming its own research paradigm as well as a method for
analysis of the big data. The semantic network analysis is a method of identifying and analyzing
relationships between words to describe a part of a connected network [35]. The semantic network
analysis, as a method of qualitative textual analysis, provides a strong theoretical and methodological
foundation with which to describe the semantic nature of the online tourism domain [36–38]. Centrality
and proximity were employed to measure the structure of the semantic network and to compare the
differences between two semantic structures in the Jo and Kim’s [39] study. The similarities matrix
generated in the text analysis can be used as input into multidimensional scaling to assess both the
content and structure of the semantic network [40,41].
While in this study, the similarities of the top 100 frequent words were conducted by CONCOR
analysis, and the methodology for visualizing data are vital for understanding the semantic network
of words [42]. The network can be visualized and verified, and the visual representation makes it easy
Sustainability 2019, 11, 4066 4 of 17

to see at a glance the structure of the network or the associativity between nodes [43]. The approach
and visualization for the semantic network analysis of this study was performed by Ucinet 6.0.

3. Methodology

3.1. Data Collection

The data was collected by web crawling, and the web crawler written in the Python 2.7. The server
operating system is the Ubuntu 16.04 LTS from Skytrax (airlinequality.com). Several types of data were
collected including the airline brand, type of aircraft, location of departure, destination, type of traveler,
seat type, date flown, overall satisfaction rating along with additional ratings of seat comfort, staff,
F&B, inflight entertainment, ground service, value for money, and the textual reviews with title and
content with recommendation [25]. The overall satisfaction rating score was defined as the customer
satisfaction in this study.
Initially, 62249 reviews were collected in 100 airlines, and there were 9632 reviews in the top
10 airlines presented by Skytrax (airlinequality.com). This study analyzes the top 10 airlines efficiently
in terms of time and cost. The top 10 airlines are Singapore Airlines, Qatar Airways, All Nippon
Airways, Emirates, EVA Air, Cathay Pacific Airways, Lufthansa, Hainan Airlines, Garuda Indonesia
and Thai Airways. The period of collected reviews is from January 2011 to March 2019. The size of
data is 9632 reviews and a total of 1,290,047 words are extracted [25].
The one-way analysis of variance (ANOVA) was used to analyze the mean difference between the
average customer satisfaction with individual brands, and the method of Duncan was used for the
post-hoc test. As shown in Table 2, Emirates has the largest number of reviews (1754) because it has
the highest market share in the airline industry; it was followed by Qatar (1415), Lufthansa (1395),
Cathay Pacific Airways (1173), and Singapore (1047).

Table 2. Reviews and average ratings according to airline brands.

Rank Brands Reviews Customer Satisfaction Std. Std. Error

1 Singapore Airlines 1047 7.501 c 2.7292 0.0844
2 Qatar Airways 1415 7.477 c 2.5277 0.0672
3 All Nippon Airways 466 7.765 bc 2.4989 0.1158
4 Emirates 1754 5.766 e 3.2187 0.0768
5 EVA Air 526 8.030 ab 2.3012 0.1002
6 Cathay Pacific Airways 1173 6.908 d 2.9557 0.0863
7 Lufthansa 1395 6.652 d 3.0144 0.0806
8 Hainan Airlines 347 8.238 a 2.2909 0.1228
9 Garuda Indonesia 726 8.301 a 2.0979 0.0778
10 Thai Airways 783 6.796d 2.8231 0.1009
Total/Avg. 9632 7.058 2.8972 0.0295
* F = 86,101, p < 0.001, a,b,c,d,e : Ducan Post-hoc test.

The results of the analysis of variance among customer satisfaction with brands showed the
following results, in decreasing order of satisfaction level: Garuda Indonesia (8.301), Hainan Airways
(8.238), EVA Air (8.030), All Nippon Airways (7.765), Singapore Airlines (7.501), Qatar Airways
(7.477), Cathay Pacific Airways (6.908), Thai Airways (6.796), Lufthansa (6.652) and Emirates (5.766).
The results showed that there were significant differences between the customer satisfaction and
numbers of reviews.

3.2. Qualitative Data Analysis

The analysis of this study, as shown in Figure 1, carried out the following process to identify
the frequency and importance of keywords for the recognition of airline passengers [44]. The first
stage was going to be collecting the review data online and refining the collected text. As for the
 Sustainability 2019, 11, 4066 5 of 17

data analysis, Ucinet 6.0 packaged with Netdraw, which is a visualization tool. The Netdraw is
an illustration
Sustainability program
2019, 11, to REVIEW
x FOR PEER express how the frequent words are related and make an impact 5 ofwith
17
certain interaction. It is mainly concentrated on the semantic network analysis about the extracted top
frequent keywords
100 frequent keywords fromfrom
raw raw
data.data.
After that,that,
After the the
centrality analysis
centrality of of
analysis Freeman’s degree
Freeman’s has
degree hasbeen
been
worked
workedout outfor
forrefining
refiningthe
themeanings
meaningsofofconnection
connectionbetween
betweenthethetop
top100
100frequent
frequentwords.
words.

Qualitative
Figure1.1.Qualitative
Figure researchprocess.
research process.

AccordingtotoLiLiand
According andSun
Sun[45],
[45],different
differentmeasures
measuressuch
suchasascentrality
centralityand
andproximity
proximitycan
canbebederived
derived
totomeasure
measurethethestructure
structureofofthe
thesemantic
semanticnetwork
networkand
andtotocompare
comparedifferences
differencesbetween
betweentwo
twosemantic
semantic
structures. The last step of the analysis is CONCOR analysis. It helps segments of
structures. The last step of the analysis is CONCOR analysis. It helps segments of upper frequent upper frequent
words and visualizes the segmentation of higher frequent words so that frequently
words and visualizes the segmentation of higher frequent words so that frequently used words used words belong
to anyto
belong particular category
any particular and other
category and words belongbelong
other words to anytogroup.
any group.
3.3. Quantitative Data Analysis
3.3. Quantitative Data Analysis
For the quantitative analysis, the evaluation data on six factors (seat comfort, staff, F&B,
For the quantitative analysis, the evaluation data on six factors (seat comfort, staff, F&B,
entertainment, ground service, value for money) with customer satisfaction and recommendation
entertainment, ground service, value for money) with customer satisfaction and recommendation
from Skytrax (airlinequality.com) were collected. Using the SPSS (IBM, Armonk, NY, USA) Statistics
from Skytrax (airlinequality.com) were collected. Using the SPSS (IBM, Armonk, NY, USA) Statistics
program, it conducted a linear regression analysis and verified the impact of each factor and the
program, it conducted a linear regression analysis and verified the impact of each factor and the
customer satisfaction. In addition, the relationship between customer satisfaction and recommendation
customer satisfaction. In addition, the relationship between customer satisfaction and
was analyzed. For verification, a research model was established as shown in Figure 2.
recommendation was analyzed. For verification, a research model was established as shown in Figure
There were thirteen hypotheses suggested as follows:
2.
Hypotheses 1-1. Seat comfort positively influences the customer satisfaction of the airline.
Hypotheses 1-2. Staff positively influences the customer satisfaction of the airline.
Hypotheses 1-3. F&B positively influences the customer satisfaction of the airline.
Hypotheses 1-4. Entertainment positively influences the customer satisfaction of the airline.
Hypotheses 1-5. Ground service positively influences the customer satisfaction of the airline.

Figure 2. Quantitative research model.

 Figure 1. Qualitative research process.

According to Li and Sun [45], different measures such as centrality and proximity can be derived
Sustainability 2019, 11, 4066 6 of 17
to measure the structure of the semantic network and to compare differences between two semantic
structures. The last step of the analysis is CONCOR analysis. It helps segments of upper frequent
Hypotheses 1-6. Value for
words and visualizes themoney positively influences
segmentation of higher the customer
frequent satisfaction
words of the
so that airline. used words
frequently
belong to any particular category and other words belong to any group.
Hypotheses 2-1. Seat comfort positively influences the recommendation of the airline.
Hypotheses 2-2. Staff
3.3. Quantitative Data positively
Analysis influences the recommendation of the airline.
Hypotheses
For the2-3. F&B positively
quantitative influences
analysis, the the recommendation
evaluation data onof the
six airline.
factors (seat comfort, staff, F&B,
entertainment, ground service, value for money) with customer satisfaction and recommendation
Hypotheses 2-4. Entertainment positively influences the recommendation of the airline.
from Skytrax (airlinequality.com) were collected. Using the SPSS (IBM, Armonk, NY, USA) Statistics
Hypotheses
program, it 2-5. Ground aservice
conducted linearpositively influences
regression theand
analysis recommendation
verified theofimpact
the airline.
of each factor and the
Hypotheses 2-6. Value for money positively influences the recommendation of the airline. satisfaction and
customer satisfaction. In addition, the relationship between customer
recommendation was analyzed. For verification, a research model was established as shown in Figure
Hypotheses
2. 3. Customer satisfaction positively influences the recommendation of the airline.

.
Figure 2. Quantitative research model.
Figure 2. Quantitative research model.
4. Results

4.1. Frequency Analysis

To find the words most frequently used in customer reviews, Table 3 lists the top 100 frequent words
associated with the airline experience and their corresponding frequency percentage. These words
reflect the airline experience, including the airline brands such as ‘emirates’, ‘lufthansa’, ‘qatar’, ‘cathay’
and ‘singapore’ that have the frequency rank of 5, 7, 9, 10, 11, and words related with seat comfort like
‘seat’, ‘leg’, ‘legroom’ and ‘comfort’ also possess high occurrence. The words describing F&B, such as
‘food’, ‘meal’, ’drink’ and ’breakfast’ also belong to the top 100 frequency words. There are words
for ground service such as ‘lounge’, ‘luggage’, ‘ticket’ and ‘bag’. The words related with staff and
service, such as ‘crew’, ‘staff’, ‘attitude’, ‘smile’ and ‘cabin’ appeared many times, and words in the
value of money, such as ‘money’, ‘price’ and ‘cost’ were also included in the comments on their airline
experience. The distribution of major upper words is shown in Figure 3, and the result of visualizing
the network that reflects the frequency is Figure 4.

Table 3. Top 100 frequent words from the online airline review.

Rank Word Freq % Rank Word Freq % Rank Word Freq %

1 custom 4308 19.43% 35 quality 54 0.24% 69 price 20 0.09%
2 review 4225 19.05% 36 plane 52 0.23% 70 frankfurt 19 0.08%
3 airway 1747 7.88% 37 entertain 50 0.22% 71 world 19 0.08%
4 service 915 4.12% 38 standard 50 0.22% 72 notch 19 0.08%
5 emirate 816 3.68% 39 journey 48 0.21% 73 gate 19 0.08%
6 airline 790 3.56% 40 aircraft 47 0.21% 74 room 19 0.08%
7 lufthansa 694 3.13% 41 airport 45 0.20% 75 day 18 0.08%
8 flight 633 2.85% 42 choice 42 0.18% 76 attitude 18 0.08%
9 cathay 562 2.53% 43 money 40 0.18% 77 wine 18 0.08%
10 qatar 555 2.50% 44 drink 40 0.18% 78 baggage 18 0.08%
11 singapore 491 2.21% 45 check 39 0.17% 79 job 17 0.07%
12 crew 458 2.06% 46 value 37 0.16% 80 movie 17 0.07%
13 experience 408 1.84% 47 expectation 33 0.14% 81 system 17 0.07%
Sustainability 2019, 11, 4066 7 of 17

Sustainability 2019, 11, x FOR PEER REVIEW Table 3. Cont. 7 of 17

Rank Word Freq % Rank Word Freq % Rank Word Freq %

19 cabin 319 1.43% 53 space 29 0.13% 87 wifi 15 0.06%
14 garuda 372 1.67% 48 smile 33 0.14% 82 cost 17 0.07%
20 ana 298 1.34% 54 lounge 29 0.13% 88 bus 15 0.06%
15 seat 352 1.58% 49 way 32 0.14% 83 rate 17 0.07%
2116 staff
indonesia 263
352 1.18%
1.58% 55
50 legroom
ground 3229 0.13%
0.14% 89
84 member
dinner 1615 0.06%
0.07%
2217 food air 222
337 1.00%
1.52% 56
51 onboard
leg 3028 0.12%
0.13% 90
85 upgrade
future 1615 0.06%
0.07%
18 eva 321 1.44% 52 premium 30 0.13% 86 route 16 0.07%
2319 class
cabin
154
319
0.69%
1.43%
57
53
a380
space 29
26 0.11%
0.13%
91
87
kind
wifi 15
15 0.06%
0.06%
2420 mealana 116
298 0.52%
1.34% 58
54 level
lounge 2926 0.11%
0.13% 92
88 option
bus 1515 0.06%
0.06%
2521 staff
economy 263
111 1.18%
0.50% 55
59 legroom
dubai 2924 0.13%
0.10% 89
93 member
turn 1515 0.06%
0.06%
22 food 222 1.00% 56 onboard 28 0.12% 90 upgrade 15 0.06%
2623 time
class 102
154 0.46%
0.69% 60
57 a350
a380 2624 0.10%
0.11% 94
91 selection
kind 1514 0.06%
0.06%
2724 busy
meal 101
116 0.45%
0.52% 61
58 comfort
level 2623 0.10%
0.11% 95
92 average
option 1514 0.06%
0.06%
25 economy 111 0.50% 59 dubai 24 0.10% 93 turn 15 0.06%
28 hainan 99 0.44% 62 luggage 23 0.10% 96 carry 14 0.06%
26 time 102 0.46% 60 a350 24 0.10% 94 selection 14 0.06%
2927 attendant
busy 92
101 0.41%
0.45% 63
61 ticket
comfort 2323 0.10%
0.10% 97
95 departure
average 1414 0.06%
0.06%
3028 hainan
hour 7199 0.44%
0.32% 62
64 luggage
board 2322 0.10%
0.09% 96
98 carry
travel 1414 0.06%
0.06%
29 attendant 92 0.41% 63 ticket 23 0.10% 97 departure 14 0.06%
3130 inflight
hour 6471 0.28%
0.32% 65
64 lay
board 2222 0.09%
0.09% 99
98 home
travel 1413 0.05%
0.06%
3231 trip
inflight 6364 0.28%
0.28% 66
65 bag
lay 2221 0.09%
0.09% 100
99 people
home 1313 0.05%
0.05%
3332 passenger
trip
5763 0.28%
0.25% 66
67 bag
breakfast 21
21 0.09%
0.09% 100 people 13 0.05%
33 passenger 57 0.25% 67 breakfast 21 0.09%
3434 star
star 5454 0.24%
0.24% 68
68 care
care 2121 0.09%
0.09%

1000
900
800
700
600
500
400
300
200
100
0
inflight

aircraft
experience

cabin

time

drink

ground
class

space
meal
economy

quality
plane
hour

money

smile
crew

seat
service

staff

airport

premium

lounge
expectation
food

attendant

star

entertain

leg
value

Figure 3. Distribution of major 30 words frequency.

Figure 3. Distribution of major 30 words frequency.
Sustainability 2019, 11, 4066 8 of 17
Sustainability 2019, 11, x FOR PEER REVIEW 8 of 17

Figure 4. Keywords visualization of network analysis.

Figure 4. Keywords visualization of network analysis.
4.2. Semantic Network Analysis
4.2. Semantic Network Analysis
The semantic network analysis identifies the relationship between words and expresses the
The semantic
connection betweennetwork
them. The analysis identifies
centrality the relationship
and CONCOR analysis of between
keywords words
wereand expresses
performed for the
the
connection between them. The centrality and CONCOR analysis of keywords were
semantic network analysis. In the review of airlines, among the top 100 frequent words, the results of performed for the
semantic
an analysisnetwork analysis.
of the degree andIneigenvector
the review centrality
of airlines,ofamong
the wordsthe top 100 frequent
are described words,
in Table 4. the results
of anThe
analysis of the degree and eigenvector centrality of the words are
degree centrality is a simple centrality measure that counts how many neighbors described in Table 4. a node
has, and refers to the degree to which a word has many connections and becomes central,node
The degree centrality is a simple centrality measure that counts how many neighbors a andhas,
the
and
morerefers to the degree
connections it has,tothe
which a word
greater has many
its impact connections
on other words,and the becomes central, and
more dominant it canthebemore
[46].
connections
The it has,
eigenvector the greater
centrality its impact
extends on other
the concept words, the
of connective more dominant
centrality it can be
by considering not[46].
onlyThethe
eigenvector
number of wordscentrality extendsbut
connected, the concept
also of connective
how important centrality
a connected by considering
relationship is. Thus, notprestige
only the is
number of words connected, but also how important a connected relationship is. Thus,
a useful indicator for finding the most influential central node in the network [47]. It is sometimes used prestige is a
useful indicator for finding the most influential central node in the network [47]. It
to measure a node’s influence in the network. It performs matrix calculations to determine adjustments. is sometimes used
to measure
The resulta node’s influence
is that words in the
related withnetwork. It performs
airline brands recorded matrix
a highcalculations
rank in degree to determine
centrality,
adjustments.
such as ‘singapore’, ‘emirates’, ‘qatar’, ‘cathay’ and ‘lufthansa’. However, words related to airline
brandsThehave
result is that
a low rank words
in therelated with airline
eigenvector brands
centrality. The recorded a high rank
words related with in degree
staff, suchcentrality,
as ‘crew’,
such as ‘singapore’,
‘attendant’ and ‘service’‘emirates’, ‘qatar’, ‘cathay’
were frequently found in and
the‘lufthansa’.
online review. However, words
However, thoserelated
words to are
airline
not
brands have a low rank in the eigenvector centrality. The words related with
significantly connected with other words, according to the degree and eigenvector centrality analysis.staff, such as ‘crew’,
‘attendant’
In addition,and ‘service’
‘class’ were frequently
and ‘economy’ found in athe
have recorded online
higher review.
rank However,
in degree thosethan
centrality words are not
frequency
significantly connected with other words, according to the degree and eigenvector
and eigenvector centrality. ‘money’, ‘value’ and ‘expectation’ are frequently found but they are not centrality analysis.
In addition,words.
centralized ‘class’ and ‘economy’ have recorded a higher rank in degree centrality than frequency
and eigenvector centrality. ‘money’, ‘value’ and ‘expectation’ are frequently found but they are not
centralized words.
Sustainability 2019, 11, 4066 9 of 17

Table 4. Comparison of keywords frequency and centrality analysis.

Frequency Degree Eigenvector Frequency Degree Eigenvector

Freq Rank Coef. Rank Coef. Rank Freq Rank Coef. Rank Coef. Rank
custom 4308 1 2.645 1 0.610 1 time 102 26 0.318 15 0.004 20
review 4225 2 2.595 2 0.610 2 busy 101 27 0.185 26 0.002 29
airway 1747 3 1.144 3 0.365 3 hainan 99 28 0.073 46 0.020 15
service 915 4 0.388 10 0.009 17 attendant 92 29 0.062 50 0.001 33
emirate 816 5 0.450 7 0.141 4 hour 71 30 0.254 21 0.003 25
airline 790 6 0.603 5 0.117 8 inflight 64 31 0.050 57 0.000 63
lufthansa 694 7 0.333 12 0.126 7 trip 63 32 0.047 60 0.001 34
flight 633 8 0.723 4 0.010 16 passenger 57 33 0.226 24 0.003 26
cathay 562 9 0.401 9 0.134 6 star 54 34 0.032 70 0.000 64
qatar 555 10 0.414 8 0.139 5 quality 54 35 0.055 55 0.001 35
singapore 491 11 0.513 6 0.100 9 plane 52 36 0.132 30 0.001 36
crew 458 12 0.322 13 0.003 21 entertain 50 37 0.087 41 0.001 37
experience 408 13 0.088 39 0.001 32 standard 50 38 0.015 86 0.000 65
garuda 372 14 0.306 17 0.073 10 journey 48 39 0.036 68 0.000 66
seat 352 15 0.348 11 0.004 18 aircraft 47 40 0.077 44 0.001 38
indonesia 352 16 0.273 20 0.073 11 airport 45 41 0.145 28 0.002 30
air 337 17 0.308 16 0.066 13 choice 42 42 0.108 34 0.001 39
eva 321 18 0.247 22 0.065 14 money 43 43 0.029 73 0.000 67
cabin 319 19 0.288 19 0.003 22 drink 44 44 0.129 32 0.001 40
ana 298 20 0.242 23 0.072 12 check 45 45 0.098 35 0.001 41
staff 263 21 0.141 29 0.002 27 value 46 46 0.016 85 0.000 68
food 222 22 0.171 27 0.002 28 expectation 47 47 0.005 97 0.000 69
class 154 23 0.320 14 0.003 23 smile 48 48 0.020 80 0.000 70
meal 116 24 0.214 25 0.003 24 way 49 49 0.063 49 0.001 42
economy 111 25 0.292 18 0.004 19 ground 50 50 0.060 52 0.001 43

The CONCOR analysis analyzes the connection of the relationship and patterns between words
to see their similarity, and the greater the similarity of the connection relationship patterns, the greater
the degree of structural equivalence of the two words. It forms clusters that include keywords with
similarities [10]. One of the most natural methods is the cluster analysis, which is a statistical technique
that binds them into similarity groups based on interrelationships [48]. In other words, the CONCOR
analysis is a method of repeatedly analyzing correlations to find appropriate levels of similarity groups.
This study identifies the blocks of nodes according to the correlation coefficient of the matrices of the
concurrent keywords and forms clusters that include keywords with similarities [18]. The keywords
extracted from the frequency histogram according to the frequency ranking were used and a [keywords
× keywords] matrix were constructed. To visualize the analysis results, NetDraw in UCINET 6.0
program was used. The nodes are presented as blue squares and their sizes indicate their frequency;
the network shows the connectivity between them.
The result of the CONCOR analysis has been shown in Figure 5 with visibility. There are six groups
that were intricately interwoven with each other. After looking at the words in the group, the group
was named as seat comfort, staff, entertainment, airline brand, value for money and ground service.
Interestingly, the CONCOR results were very similar to the evaluation factors (seat comfort, staff, F&B,
entertainment, ground service and value for money) presented by Skytrax (airlinequality.com).
Sustainability 2019, 11, 4066 10 of 17
Sustainability 2019, 11, x FOR PEER REVIEW 10 of 17

Figure 5. Visualization of CONvergence of iterated CORrelation (CONCOR) analysis.

Figure 5. Visualization of CONvergence of iterated CORrelation (CONCOR) analysis.
To make it easier to see which words belong to each group, the words grouped in the cluster
To make it easier to see which words belong to each group, the words grouped in the cluster
and the ones to be noted are listed in Table 5. As shown in Table 5, the group name was chosen as
and the ones to be noted are listed in Table 5. As shown in Table 5, the group name was chosen as
follows considering the characteristics of the words. Airline brand segmentation refers to the airline
follows considering the characteristics of the words. Airline brand segmentation refers to the airline
brand, which is in the top 10 airlines. The Thai Airways is the only airline company, which did not
brand, which is in the top 10 airlines. The Thai Airways is the only airline company, which did not
appear in the airline brand group. It means that the Thai Airways is not in the top 100 frequent words.
appear in the airline brand group. It means that the Thai Airways is not in the top 100 frequent words.
The group of seat comfort has ‘hour’, ‘comfort’, ‘legroom’, ‘time’, ‘seat’, ‘route’, ‘plane’ and ‘room’.
The group of seat comfort has ‘hour’, ‘comfort’, ‘legroom’, ‘time’, ‘seat’, ‘route’, ‘plane’ and ‘room’.
The entertainment group contained two concepts. One is F&B, and the other one is entertainment
The entertainment group contained two concepts. One is F&B, and the other one is
activity. There are ‘food’, ‘meal’, ‘breakfast’, ‘drink’, ‘dinner’ and ‘wine’ for eating and drinking.
entertainment activity. There are ‘food’, ‘meal’, ‘breakfast’, ‘drink’, ‘dinner’ and ‘wine’ for eating and
However, ‘entertain’, ‘premium’, ‘economy’ and ‘class’ could be related to activities, therefore there
drinking. However, ‘entertain’, ‘premium’, ‘economy’ and ‘class’ could be related to activities,
are no other words for actual activity except ‘movie’. In the group of staff, there were ‘smile’ and
therefore there are no other words for actual activity except ‘movie’. In the group of staff, there were
‘kind’, which is directly connected with service attitude. So, ‘smile’ and ‘kind’ can describe the service
‘smile’ and ‘kind’, which is directly connected with service attitude. So, ‘smile’ and ‘kind’ can describe
staff working in the top 10 airlines. The other interesting word is ‘bus’ in the group of ground service.
the service staff working in the top 10 airlines. The other interesting word is ‘bus’ in the group of
‘ticket’, ‘lounge’ and ‘bag’ are usually represented by ground service. However, the number of flights,
ground service. ‘ticket’, ‘lounge’ and ‘bag’ are usually represented by ground service. However, the
airlines is diverse, and therefore the airport is crowded, so passengers need to get on the plane using
number of flights, airlines is diverse, and therefore the airport is crowded, so passengers need to get
a bus. Therefore, airline passengers may feel uncomfortable about boarding the plane. It means
on the plane using a bus. Therefore, airline passengers may feel uncomfortable about boarding the
transportation for boarding also can affect the satisfaction, because passengers wrote down their airline
plane. It means transportation for boarding also can affect the satisfaction, because passengers wrote
experience mentioning this segmentation. In addition, this service can improve passengers’ perceived
down their airline experience mentioning this segmentation. In addition, this service can improve
achievement and power compared to others [49].
passengers’ perceived achievement and power compared to others [49].
Sustainability 2019, 11, 4066 11 of 17

Table 5. Result of CONCOR analysis.

Extracted Words Significant Words

flight/seat/time/hour/way/leg/plane/journey/
legroom/comfort/luggage/board/day/system/ flight/seat/time/hour/way/leg/plane/journey/
Seat Comfort
room/trip/airport/frankfurt/route/option/ legroom/comfort/board/room/route/option
departure/travel/job/home
crew/cabin/staff/attendant/inflight/star/standard/ crew/cabin/staff/attendant/star/standard/
Staff aircraft/smile/onboard/a380/level/a350/ smile/level/notch/attitude/kind/
notch/attitude/kind/selection/average/carry selection/average/carry
service/food/class/meal/economy/busy/quality/
service/food/class/meal/economy/quality/
entertain/drink/
Entertainment entertain/drink/premium/breakfast/
check/premium/breakfast/world/wine/
wine/movie/dinner
baggage/movie/dinner/future/member
experience/choice/passenger/lounge/space/
experience/passenger/lounge/space/bag/
Ground Service bag/care/lay/ground/rate/people/turn/gate/
lay/ground/date/bus/ticket
expectation/bus/ticket
Value for Money custom/review/air/value/price/money/cost custom/money/value/price/cost
airway/emirates/airline/lufthansa/cathay/qatar/ airway/emirates/airline/lufthansa/cathay/
Airline Brand singapore/garuda/indonesia/eva/ana/ qatar/singapore/garuda/indonesia/eva/
hainan/dubai/upgrade ana/hainan/dubai

4.3. Linear Regression Analysis

Table 6 displays the results of the linear regression analysis with ‘Customer Satisfaction (CS)’ as
the dependent variable and it has six independent variables: ‘Seat comfort (SC)’, ‘Staff (S)’, ‘Food &
Beverage (FB)’, ‘Entertainment (E)’, ‘Ground service (GS)’ and ‘Value for money (VM)’. The overall
variance explained by the six predictors was 76.2% (R2 = 0.762) and the standard error of the estimated
value was calculated as 1.412. All factors are significant at the p < 0.001 level except ‘Entertainment
(E, β = −0.003, p = 0.617)’ factor. Five factors are positively related to the customer satisfaction, according
to their standardized coefficient values: ‘Seat comfort (SC, β = 0.080, p < 0.001)’, ‘Staff (S, β = 0.176,
p < 0.001)’, ‘Food & beverage (FB, β = −0.102, p < 0.001)’, ‘Value for money (VM, β = 0.603, p < 0.001)
and ‘Ground service (GS, β = 0.042, p < 0.001)’. Therefore, hypothesis 1-1, 1-2, 1-3, 1-5 and 1-6 were
supported, however hypothesis 1-4 was rejected. In order to estimate the possible correlations among
the predictors, a multicollinearity statistic was conducted. The tolerance level is less than 10.00, and the
variance inflation factor (VIF) of the predictors were between 10.00 and 100.00, respectively, that is,
the predictors were not significantly correlated to each other. Therefore, based on unstandardized β,
the regression equation can be expressed as:

CS = −0.843 + 0.172SC* + 0.349S* + 0.203FB* − 0.005E + 0.058GS* + 1.347VM*

The ‘Value for money’ factor holds the highest standardized coefficients, which mean this
experience aspect of the passenger is the most important factor associated with customer satisfaction
significantly. Reviews like “It’s value for money and it’s best if you compare to other airlines based
on travel duration and price.” and “First time trying the premium economy of Singapore Airlines,
must I say I was really impressed by the offering and the value for money.” are related to the airline
experience based upon ‘Value for money’ attributes.
Sustainability 2019, 11, 4066 12 of 17

Table 6. Results of linear regression analysis (Dependent variable: Customer Satisfaction).

Unstandardized Standardized
Collinearity Statistics
Model Coefficient Coefficient t Sig.
B Std. Error Beta Tolerance VIF
(Constant) −0.843 0.049 −17.253 0.000
Seat Comfort (SC) 0.172 0.016 0.080 10.828 0.000 0.449 2.227
Staff (S) 0.349 0.016 0.176 21.643 0.000 0.375 2.670
Food & Beverage (FB) 0.203 0.016 0.102 12.420 0.000 0.366 2.733
Entertainment (E) −0.005 0.011 −0.003 −0.501 0.617 0.676 1.480
Ground Service (GS) 0.058 0.007 0.042 8.026 0.000 0.890 1.123
Value for money (VM) 1.347 0.016 0.603 84.612 0.000 0.486 2.057
* Dependent variable: Customer Satisfaction (CS); R2 = 0.762; F = 5147.275; p < 0.001.

Table 7 displays the results of the linear regression analysis with ‘Recommendation (R)’ as the
dependent variable and it has six independent variables: ‘Seat comfort (SC)’, ‘Staff (S)’, ‘Food &
Beverage (FB)’, ‘Entertainment (E)’, ‘Ground service (GS)’ and ‘Value for money (VM)’. The overall
variance explained by the six predictors was 60.5% (R2 = 0.605) and the standard error of the estimated
value was calculated as 0.2773. All factors are significant at the p < 0.001 level except ‘Entertainment
(E, β = −0.015, p = 0.053)’ factor. Five factors are positively related to the recommendation, according
to their standardized coefficient values: ‘Seat comfort (SC, β = 0.045, p < 0.05)’, ‘Staff (S, β = 0.197,
p < 0.05)’, ‘Food & beverage (FB, β = 0.040, p < 0.05)’, ‘Ground service (GS, β = 0.014, p < 0.05) and
‘Value for money (VM, β = 0.577, p < 0.05)’0. Therefore, hypothesis 2-1, 2-2, 2-3, 2-5 and 2-6 was
supported, however hypothesis 2-4 was rejected. Therefore, based on unstandardized β, the regression
equation can be expressed as:

R = −0.325 + 0.015SC* + 0.059S* + 0.012FB* − 0.004E + 0.003GS* + 0.196VM*

Table 7. Results of linear regression analysis (Dependent variable: Recommendation).

Unstandardized Standardized
Collinearity Statistics
Model Coefficient Coefficient t Sig.
B Std. Error Beta Tolerance VIF
(Constant) −0.325 0.010 −33.886 0.000
Seat Comfort (SC) 0.015 0.003 0.045 4.4741 0.000 0.449 2.227
Staff (S) 0.059 0.003 0.197 18.794 0.000 0.375 2.670
Food & Beverage (FB) 0.012 0.003 0.040 3.807 0.000 0.366 2.733
Entertainment (E) −0.004 0.002 −0.015 −1.936 0.053 0.676 1.480
Ground Service (GS) 0.003 0.001 0.014 2.032 0.042 0.890 1.123
Value for Money (VM) 0.196 0.003 0.577 62.722 0.000 0.486 2.057
* Dependent variable: Recommendation (R); R2 = 0.605; F = 2451.601; p < 0.001

To verify the part corresponding to hypothesis 3, the third linear regression analysis was
performed. The result is shown in Table 8 with a recommendation as the dependent variable and it has
independent variables as customer satisfaction. The overall variance explained by the predictor was
69.4% (R2 = 0.694) and the standard error of the estimated value was calculated as 0.244. The customer
satisfaction is significant and positively related to the recommendation, according to their standardized
coefficient values: β = 0.833, p < 0.001). Therefore, hypothesis 3 was supported.
Sustainability 2019, 11, 4066 13 of 17

Table 8. Results of linear regression (Dependent variable: Recommendation).

Standardized
Unstandardized Coefficient
Model Coefficient t Sig.
B Std. Error Beta
(Constant) −0.158 0.007 −24.173 0.000
Customer Satisfaction (CS) 0.127 0.001 0.833 147.592 0.000
* Dependent variable: Recommendation(R); R2 = 0.694; F = 21783.469; p < 0.001.

5. Discussion
This study was conducted to enhance the customer’s experience, satisfaction and recommendation
by qualitatively and quantitatively analyzing the reviews of airline passengers. For the airline
passengers’ review data analysis, the first process is extracting keywords by text mining and the second
is grouping them using the CONCOR analysis. In addition, the study conducted three consecutive linear
regression analyses to understand the relationship between evaluation factors, customer satisfaction,
and recommendations presented on the customer review website. Interestingly, six clusters (airline
brand, seat comfort, staff, entertainment, ground service and value for money) derived from qualitative
semantic network analysis were very similar to the six evaluation factors (seat comfort, staff, F&B,
entertainment, ground service and value for money) that Skytrax is asking customers to evaluate on
the website. Therefore, this study quantitatively analyzed the impact relationships among the six
evaluation factors of Skytrax, customer satisfaction and recommendation.
The following implications can be suggested by combining qualitative and quantitative analysis:
First of all, the group representing the highest beta coefficient was ‘Value for money’ in the linear
regression analysis, and the related words were ‘value’, ‘price’, ‘money’ and ‘cost’ through the semantic
network analysis. The group contains fewer words than the other group. Even though customers are
not frequently mentioning the words related with ‘Value for money’ on the online review, it is still the
most important factor to figure out the customer experience of airlines. According to Brochado, Rita,
Oliveira, & Oliveira [12], ‘Value for money’ is the key factor as a criterion for positive and negative
eWOM. In other words, the airline passengers who classify ‘Value for money’ as very good or excellent
also provide positive eWOM about the airlines’ seats, staff, entertainment and food. In addition,
Rajaguru [50] examined the effect of value for money and service quality on customer satisfaction and
behavioral intention. The results of this study show the same results as most prior studies show that
Value for Money has the greatest impact on customer satisfaction and recommendation. Therefore,
the airline should focus its most essential tangible and intangible resources on the value for money.
The second highest beta value was ‘Staff’ in the linear regression analysis, and the related words
were ‘crew’, ‘cabin’, ‘staff’, ‘attendant’, ‘smile’, ‘kind’ and ‘attitude’ through the semantic network
analysis. Service performing by staff was an essential key factor to create a good image in the service
industry and can still be seen as a part of the company that must be managed at all times to keep up
the image of the company. Therefore, it is important to improve the attitude of employees through
systematic service training. In addition, providing an environment to enhance employee satisfaction to
produce better service to customers can be another way.
Third, the entertainment group derived from the CONCOR analysis includes two factors
(Entertainment, F&B) of passenger evaluation factors on the Skytrax website. Based on the words
relating with ‘entertainment’ were mentioned in the online review text such as ‘service’, ‘food’,
‘class’, ‘meal’, ‘economy’, ‘quality’, ‘entertain’, ‘drink’, ‘premium’, ‘breakfast’, ‘wine’, ‘movie’ and
‘dinner’ through the semantic network analysis. As the result of regression analysis, evaluation
factor on the Skytrax ‘Entertainment’ had no impact on customer satisfaction and recommendation.
However, the other evaluation factor ‘F&B’ had impact on customer satisfaction and recommendation.
In particular, the results related to F&B are significant, and have been found in the recent study as
a critical customer satisfaction factor in the Tourism industry [38,51,52].
Sustainability 2019, 11, 4066 14 of 17

Fourth, ‘seat’ recorded rank 15 in the top 100 frequent words, and related words were ‘flight’, ‘seat’,
‘time’, ‘hour’, ‘way’, ‘leg’, ‘plane’, ‘journey’, ‘legroom’, ‘comfort’, ‘board’, ‘room’, ‘route’ and ‘option’
through semantic network analysis. In addition, ‘Seat comfort’ had impact on customer satisfaction
and recommendation through linear regression analysis. The airline industry has the characteristic
of sharing a narrow space with many people, therefore ‘Seat comfort’ can have a significant impact
on customer satisfaction and recommendation. According to the meta-analysis conducted by Lim
& Tkaczynski [53], seating comfort is among the most frequent items mentioned in airline service
quality studies. The current results also confirmed that ‘Seat comfort’ is an important dimension in
airline industry.
Lastly, words related to ground service are ‘lounge’, ‘bag’, ‘bus’ and ‘ticket’. If the ‘Seat comfort’
is the indoor physical environment, the ‘Ground service’ can be the outdoor physical environment
to provide a comfortable environment outside of the aircraft. So the airlines need to take care of
the condition of the lounge. Especially, the lounge is the space used while waiting, and the time for
waiting will be a chance to provide impressive service to the passengers. F&B provided in the lounge is
absolutely important to have quality rest in the lounge. Ground services are shown in linear regression
results that have a significant positive influence on both customer satisfaction and recommendation.
Airline companies can satisfy passengers and create a positive image by paying attention to the quality
of their baggage claim service, the ticketing service required to board the aircraft, and the quality of
service provided on limousine buses.
This study presents the academic implication that the study has extended its application area of
semantic network analysis. While given the significance of the airline segment in the tourism industry,
this study empirically explores among airline experience, customer satisfaction and recommendation
by big data analytics. Along the way, the airline has the opportunity to gain an understanding of
factors on the review web site, so as to infiltrate into this market and create corresponding marketing
strategies for their strong advantages. Understanding online reviews as a manifestation of passengers’
experiences can help airlines to identify the main attributes required to achieve positive post-purchase
behaviors and to minimize negative intentions. Thus, the online reviews not only provide an efficient
way for airlines to collect feedback from airline passengers, but also provide an opportunity to discover
how to generate positive intent after the experience. To create a high customer rating and a positive
eWOM, airlines should consider ‘Seat comfort’, ‘Staff’, ‘F&B’, ‘Ground service’ and ‘Value for money’.
Among them, ‘Value for money’ was the most influential attribute in the regression analysis. These key
factors may be used to examine the customer satisfaction or to test theoretical models to have a better
understanding of airline passenger behavior.
In practice, the analysis of online reviews can be used as a diagnostic tool by managers since
customer feedback is important for airlines to improve services and products, and to take action
regarding service. The analysis also provides the level of importance of these service attributes so
airlines can allocate their resources accordingly. Online review analyses can provide reliable satisfaction
assessment for airlines. Airlines can also use this method to analyze their competitors’ passenger
feedbacks so that they can benchmark themselves against competitors in terms of customer satisfaction.
These reviews can be used for sustainable strategic marketing decisions against competitors.
However, this study shows limitations in the area of the study as it focuses on airlines that are
mainly handled by the Skytrax (airlinequality.com). Therefore, in future studies, big data analysis using
social media data reviews, which is known all around the world, will be a better way to understand
consumer trends. Secondly, the collected text was analyzing based on the frequency of individual
words, therefore, it is difficult to understand the additional meaning of words. In future studies, further
analysis of positives and negatives, and sentimental analysis regarding airline recognition is expected
to be carried out to better understand the customer’s thoughts and to present stronger strategies to the
airline industry.

Author Contributions: H.-J.B. and H.-S.K. designed the research model, analyzed online review data and wrote
the paper.
Sustainability 2019, 11, 4066 15 of 17

Funding: This research was supported by Kyungsung University Research Grants in 2019 [grant number
KSU-Grants2019]. Additionally, this work was supported by the Ministry of Education of the Republic of Korea
and the National Research Foundation of Korea (NRF-2016S1A5A2A03928029).
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Siering, M.; Deokar, A.V.; Janze, C. Disentangling consumer recommendations: Explaining and predicting
airline recommendations based on online reviews. Decis. Support Syst. 2018, 107, 52–63. [CrossRef]
2. Li, W.; Yu, S.; Pei, H.; Zhao, C.; Tian, B. A hybrid approach based on fuzzy AHP and 2-tuple fuzzy linguistic
method for evaluation in-flight service quality. J. Air Transp. Manag. 2017, 60, 49–64. [CrossRef]
3. International Air Transport Association. Industry Statistics Fact Sheet. Available online: https://www.iata.
org/pressroom/facts_figures/fact_sheets/Documents/fact-sheet-industry-facts.pdf (accessed on 20 May 2019).
4. Beers, B. Investopedia. Available online: https://www.investopedia.com/ask/answers/040715/what-are-
major-expenses-affect-companies-airline-industry.asp (accessed on 3 July 2019).
5. Elliott, K.M.; Roach, D.W. Service quality in the airline industry: Are carriers getting an unbiased evaluation
from consumers? J. Prof. Serv. Mark. 1993, 9, 71–82. [CrossRef]
6. Aksoy, S.; Atilgan, E.; Akinci, S. Airline services marketing by domestic and foreign firms: Differences from
the customers’ viewpoint. J. Air Transp. Manag. 2003, 9, 343–351. [CrossRef]
7. Gilbert, D.; Wong, R.K. Passenger expectations and airline services: A Hong Kong based study. Tour. Manag.
2003, 24, 519–532. [CrossRef]
8. Park, J.W.; Robertson, R.; Wu, C.L. The effect of airline service quality on passengers’ behavioural intentions:
A Korean case study. J. Air Transp. Manag. 2004, 10, 435–439. [CrossRef]
9. Xu, X.; Liu, W.; Gursoy, D. The impacts of service failure and recovery efforts on airline customers’ emotions
and satisfaction. J. Travel Res. 2018, 58, 1034–1051. [CrossRef]
10. Zhang, Y.; Cole, S.T. Dimensions of lodging guest satisfaction among guests with mobility challenges:
A mixed-method analysis of web-based texts. Tour. Manag. 2016, 53, 13–27. [CrossRef]
11. Liau, B.Y.; Tan, P.P. Gaining customer knowledge in low cost airlines through text mining. Ind. Manag.
Data Syst. 2014, 114, 1344–1359. [CrossRef]
12. Saha, G.C.; Theingi. Service quality, satisfaction, and behavioural intentions: A study of low-cost airline
carriers in Thailand. Manag. Serv. Qual. Int. J. 2009, 19, 350–372.
13. Brochado, A.; Rita, P.; Oliveira, C.; Oliveira, F. Airline passengers’ perceptions of service quality: Themes in
online reviews. Int. J. Contemp. Hosp. Manag. 2019, 31, 855–873. [CrossRef]
14. Hunt, J.D. Image as a factor in tourism development. J. Travel Res. 1975, 13, 1–7. [CrossRef]
15. Walker, J.; Baker, J. An exploratory study of a multi-expectation framework for services. J. Serv. Mark. 2000,
14, 411–431. [CrossRef]
16. Tao, S.; Kim, H.S. Cruising in Asia: What can we dig from online cruiser reviews to understand their
experience and satisfaction. Asia Pac. J. Tour. Res. 2019, 24, 514–528. [CrossRef]
17. Gretzel, U. Consumer generated content–trends and implications for branding. E Rev. Tour. Res. 2006,
4, 9–11.
18. Kim, H.S.; Noh, Y. Elicitation of design factors through big data analysis of online customer reviews for
washing machines. J. Mech. Sci. Technol. 2019, 33, 2785–2795. [CrossRef]
19. Ban, H.J.; Kim, H.S. A study on the TripAdvisor review analysis of restaurant recognition in Busan 1:
Especially concerning English reviews. Culin. Sci. Hosp. Res. 2019, 25, 1–11.
20. Browning, V.; So, K.K.F.; Sparks, B. The influence of online reviews on consumers’ attributions of service
quality and control for service standards in hotels. J. Travel Tour. Mark. 2013, 30, 23–40. [CrossRef]
21. Dellarocas, C.; Zhang, X.M.; Awad, N.F. Exploring the value of online product reviews in forecasting sales:
The case of motion pictures. J. Interact. Mark. 2007, 21, 23–45. [CrossRef]
22. Minnema, A.; Bijmolt, T.H.; Gensler, S.; Wiesel, T. To keep or not to keep: Effects of online customer reviews
on product returns. J. Retail. 2016, 92, 253–267. [CrossRef]
23. Mudambi, S.M.; Schuff, D. What makes a helpful review? A study of customer reviews on Amazon.com.
MIS Q. 2010, 34, 185–200. [CrossRef]
Sustainability 2019, 11, 4066 16 of 17

24. Sotiriadis, M.D.; VanZyl, C. Electronic word-of-mouth and online reviews in tourism services: The use of
twitter by tourists. Electron. Commer. Res. 2013, 13, 103–124. [CrossRef]
25. Skytrax. World’s Top 100 Airlines 2018. Available online: https://www.worldairlineawards.com/worlds-top-
100-airlines-2019/ (accessed on 20 March 2019).
26. Adeniran, A.; Fadare, S.O. Relationship between passengers’ satisfaction and service quality in murtala
muhammed international airport, Lagos, Nigeria. Int. J. Res. Ind. Eng. 2018, 7, 349–369.
27. Jeong, E.Y. Analyze of Airlines Online-reviews: Focusing on Skytrax. J. Tour. Leis. Res. 2017, 2, 261–276.
28. Yayla-Kullu, H.M.; Tansitpong, P. A critical evaluation of US airlines’ service quality performance: Lower
costs vs. satisfied customers. J. Manag. Strategy 2013, 4, 1. [CrossRef]
29. Pérezgonzález, J.D.; Gilbey, A. Predicting Skytrax airport rankings from customer reviews. J. Air Transp. Manag.
2011, 5, 335–339.
30. Lohmann, G.; Albers, S.; Koch, B.; Pavlovich, K. From hub to tourist destination—An explorative study of
Singapore and Dubai’s aviation-based transformation. J. Air Transp. Manag. 2009, 15, 205–211. [CrossRef]
31. Mason, K.J.; Morrison, W.G. Towards a means of consistently comparing airline business models with
an application to the ‘low cost’airline sector. Res. Transp. Econ. 2008, 24, 75–84. [CrossRef]
32. Park, J.W.; Robertson, R.; Wu, C.L. The effects of individual dimensions of airline service quality: Findings
from Australian domestic air passengers. J. Hosp. Tour. Manag. 2006, 13, 161–176. [CrossRef]
33. Gillen, D.; Lall, A. Competitive advantage of low-cost carriers: Some implications for airports. J. Air
Transp. Manag. 2004, 10, 41–50. [CrossRef]
34. Davenport, T.H.; Barth, P.; Bean, R. How big data is different. MIT Sloan Manag. Rev. 2012, 54, 43.
35. He, W.; Xu, G. Social media analytics: Unveiling the value, impact and implications of social media analytics
for the management and use of online information. Online Inf. Rev. 2016, 40. [CrossRef]
36. Han, S.I.; Kim, T.H.; Lee, J.H.; Kim, H.S. A study on the application of SNS big data to the industry in the
fourth industrial revolution. Culin. Sci. Hosp. Res. 2017, 23, 1–10.
37. Oh, C.; Kim, H.S. Understanding of nutrition labelling use and related factors among Korean adults. Culin. Sci.
Hosp. Res. 2018, 24, 16–22.
38. Kim, Y.J.; Ban, H.J.; Kim, H.S. An exploratory study on the semantic network analysis of Busan tourism:
Using Google web and news. Culin. Sci. Hosp. Res. 2019, 25, 126–134.
39. Jo, A.; Kim, H.S. A comparison of Starbucks between South Korea and U.S.A. through big data analysis.
Culin. Sci. Hosp. Res. 2018, 23, 195–205.
40. Shuting, T.; Kang, B.; Kim, H.S. Understanding the food hygiene of cruise through the big data analytics
using the web crawling and text mining. Culin. Sci. Hosp. Res. 2018, 24, 34–43.
41. Kim, H.S.; Yim, H.R. An exploratory study on the semantic network analysis of ‘Culinary Science &
Hospitality Research’ through the Google Scholar. Culin. Sci. Hosp. Res. 2018, 24, 1–10.
42. Shuting, T.; Kim, H.S. A study of comparison between cruise tours in China and USA through big data
analytics. Culin. Sci. Hosp. Res. 2017, 23, 1–10. [CrossRef]
43. Oh, C.; No, J.K.; Kim, H.S. Dietary pattern classifications with nutrient intake and body composition changes
in Korean elderly. Nutr. Res. Pract. 2014, 8, 192–197. [CrossRef]
44. Hong, J.W.; Park, S.B. The identification of marketing performance using text mining of airline review data.
Mob. Inf. Syst. 2019, 19. [CrossRef]
45. Li, H.; Sun, J. Forecasting business failure: The use of nearest-neighbour support vectors and correcting
imbalanced samples—Evidence from the Chinese hotel industry. Tour. Manag. 2012, 33, 622–634. [CrossRef]
46. Kim, H.S. A semantic network analysis of big data regarding food exhibition at convention center. Culin. Sci.
Hosp. Res. 2017, 23, 257–270.
47. Bonacich, P. Some unique properties of eigenvector centrality. Soc. Netw. 2007, 29, 555–564. [CrossRef]
48. Lee, Y.J.; Yoon, J.H. Exploring ways to utilize SNS big data in the tourism sector. Int. J. Tour. Hosp. Res. 2014,
28, 5–14.
49. Kim, J.J.; Kim, K.; Hwang, J. Self-enhancement driven first-class airline travelers’ behavior: The moderating
role of third-party certification. Sustainability 2019, 11, 3285. [CrossRef]
50. Lim, S.S.; Tkaczynski, A. Origin and money matter: The airline service quality expectations of international
students. J. Hosp. Tour. Manag. 2017, 31, 244–252. [CrossRef]
51. Choi, H.Y.; Kwak, G.H.; Kim, H.S. A positioning study of national food: In perspective of Korean, American,
Chinese food tourists. Culin. Sci. Hosp. Res. 2017, 23, 86–94. [CrossRef]
Sustainability 2019, 11, 4066 17 of 17

52. Kim, H.S. An exploratory study on the semantic network analysis of food tourism through the big data.
Culin. Sci. Hosp. Res. 2017, 23, 22–32.
53. Rajaguru, R. Role of value for money and service quality on behavioural intention: A study of full service
and low cost airlines. J. Air Transp. Manag. 2016, 53, 114–122. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).