Identifying attributes of public transport services for urban tourists: A data-mining method
Introduction
Cities with effective and extensive Public Transport (PT) networks are potentially more attractive to tourists (Hall et al., 2017). The contribution of PT to urban tourism should motivate transport planners and academic researchers to deepen their understanding of tourists' needs to improve the provision of PT services for this population (Hall et al., 2017) and to develop marketing and management tools to more effectively promote the use of PT (Le-Klähn and Hall, 2015). However, as Thompson and Schofield (2007) noted, transport service providers have often neglected the issue of tourist satisfaction with PT. More recently, this claim was strengthened by Hall et al. (2017, p. 141), who noted, “Local transport planners need to develop strategies to become more tourist-friendly and recognize that tourists from other countries may have a different public transport service culture and system.”
The current work aims to bridge the gaps in our knowledge regarding tourist satisfaction with PT, and deepen our understanding of Quality of Service attributes for tourist PT provision (from now on – PT-QoS). Increasing the use of PT by tourists is important for tourists, locals, and transport authorities. First, traveling on PT can enhance tourists' satisfaction with the destination by offering a way to meet local people and experience the destination at close hand (Thompson and Schofield, 2007). Second, tourism income can provide funding for PT development and service improvement (Albalate and Bel, 2010). Third, PT has clear environmental benefits over the use of private cars (Hall et al., 2017).
This study offers a new approach for using data mining to identify tourist needs through items posted in Question and Answer (Q&A) forums in TripAdvisor (TA). TA has become the most popular social media platform for tourism-related research (Chang et al., 2020; Gal-Tzur et al., 2019; Taecharungroj and Boonyanit, 2019). Taecharungroj and Boonyanit (2019) reviewed previous works on text mining from online reviews and found that of 16 recent relevant works, eleven were based on TA as a single source, two studies used TA plus other sources (Google Maps, Expedia, Yelp), and just three used only other sources (Dadoo, Airbnb, and Yelp). This phenomenon in the research community reflects the vast popularity of TA and its active Q&A forums among users. In May 2020, TA's Europe forums had more than 21 M posts covering over 3 M topics (TripAdvisor, 2020). By comparison, the competitor site Lonely Planet had about 184 K posts in its Europe forums over the same period (Lonely Planet, 2020). Many tourists even favor TA over official information channels (Mkono and Tribe, 2017). In addition, transport is one of the prominent themes addressed in TA forums (Edwards et al., 2017). For all these reasons, we consider TA's Q&A forums the most efficient way to capture the population of tourists seeking PT-related information.
The present study applies a data mining method to a database of items posted in TA Q&A forums to probe tourists' attitudes towards PT-QoS attributes. We address the following research questions: (1) What PT-QoS attributes are more important to tourists planning to visit urban destinations? (2) How different are these PT-QoS attributes across destinations, tourists' region of origin, seasons of the year, and the year the question is posted?
Section snippets
Quality of service of urban public transport for tourists – PT-QoS
The QoS of public transport can be defined by a wide range of attributes. These attributes, in turn, can be ascertained from (1) stated preference models, based on hypothetical behaviors of potential users, and (2) revealed preferences based on actual use (Paulley et al., 2006). One of the most recent studies addressing PT-QoS attributes highlights their crucial role in determining QoS level, thus significantly affecting the intention to use PT (de Oña, 2020). Researchers have employed
Research design
We use text mining to analyze TA content and identify recurring QoS-PT attributes. Our first task was, therefore, to select a text-mining method for this work. General text-mining methods, such as Support Vector Machine or Random Forest, are suitable for classifying general information (Gal-Tzur et al., 2014) but are less suited to the focused search employed in the present study (identifying specific PT-QoS attributes). On the other hand, Bekkerman and Gavish (2011) argued that Phrase-Based
Results
As described earlier, a total of 8905 questions (37.4% of the 23,797 questions in the dataset, most posted between 2008 and 2018) were classified into one or more of the four PT-QoS attributes. Each record used for the content analysis included the year, the destination city, the origin of the person posting the question (Asia, Europe, North America, and Oceania), the season of posting, and the question's content. Table 3 presents the descriptive statistics for each destination, region of
Discussion
This study analyzed already-known PT-QoS attributes using the PBC method for a content analysis of questions harvested from TA city Q&A forums. The study aims to explore the intensity to which each PT-QoS attribute is addressed in these questions, as well as possible differences between destinations, tourists' origin, seasons and years. Four PT-QoS attributes were identified as most important to potential urban tourists: Pricing and ticketing, Accessibility, Trip duration, and Service
Limitations and future research
This study has several limitations. First, the study focuses on TA Q&A forums for seven tourist destinations in Europe and the United States, all of which commonly employ the English language. Hence, this study does not represent the entire population of tourists. In particular, our sample does not represent tourists who do not use TA Q&A forums; tourists traveling to the developing world; and tourists coming from places whose residents tend not to have a good English command. Relatedly, the
Declaration of competing interests
None.
Acknowledgments
The authors would like to thank the Israel Ministry of Science, Technology, and Space for supporting this research, and also two anonymous reviewers for their valuable suggestions.
References (44)
- et al.
Tourism and urban public transport: holding demand pressure under supply constraints
Tour. Manag.
(2010) - et al.
Using deep learning and visual analytics to explore hotel reviews and responses
Tour. Manag.
(2020) - et al.
A novel transit rider satisfaction metric: rider sentiments measured from online social media data
J. Public Transp.
(2013) - et al.
The quality of service desired by public transport users
Transp. Policy
(2011) - et al.
Service quality attributes affecting customer satisfaction for bus transit
J. Public Transp.
(2007) - et al.
User satisfaction and the organization of local public transport: evidence from European cities
Transp. Policy
(2013) - et al.
The potential of social media in delivering transport policy goals
Transp. Policy
(2014) - et al.
Key factors for successful leisure and tourism public transport provision
J. Transp. Geogr.
(2007) - et al.
What do hotel customers complain about? Text analysis using structural topic model
Tour. Manag.
(2019) - et al.
Ground travel mode choices of air arrivals at regional destinations: the significance of tourism attributes and destination contexts
Res. Transp. Econ.
(2010)
Automating a framework to extract and analyse transport related social media content: the potential and the challenges
Transp. Res. Part C
Visitor users vs. non-users of public transport: the case of Munich, Germany
J. Destin. Mark. Manag.
Analysis of visitor satisfaction with public transport in Munich
J. Public Transp.
Identifying emerging hotel preferences using emerging pattern mining technique
Tour. Manag.
Big data in tourism research: a literature review
Tour. Manag.
Factors affecting the design of tourism bus services
Ann. Tour. Res.
The demand for public transport: the effects of fares, quality of service, income and car ownership
Transp. Policy
Exploring the capacity of social media data for modelling travel behaviour: opportunities and challenges
Transp. Res. Part C
Hotel location and tourist activity in cities
Ann. Tour. Res.
Analysing TripAdvisor reviews of tourist attractions in Phuket, Thailand
Tour. Manag.
An investigation of the relationship between public transport performance and destination satisfaction
J. Transp. Geogr.
Service encounters, experiences and the customer journey: defining the field and a call to expand our lens
J. Bus. Res.
Cited by (17)
Method for multiple attribute group decision making with single-valued neutrosophic number and applications to service quality evaluation of urban public transport
2023, International Journal of Knowledge-Based and Intelligent Engineering SystemsDesigning a model for selecting, ranking and optimising service quality indicators using meta-heuristic algorithms
2023, International Journal of Data Mining, Modelling and Management