Content area
Full Text
1. Introduction
The size of the online travel industry is steadily expanding and anticipated to reach nearly US$404bn by 2020 (Phocuswright, 2017). Concomitant with the popularity of online booking and global tourism, researchers and industry practitioners are increasingly interested in travelers’ online booking behavior and travel characteristics. The majority of online travel booking studies have focused on attitudinal variables such as traveler perception, platform usability, interface design and technology acceptance by using survey and interview methodological approaches (Chang et al., 2019; Jeon et al., 2019). However, there has been a lack of research endeavors on the behavioral elements of travelers’ online booking (e.g. the number of clicks on booking) and the variations of the elements depending on trip characteristics (e.g. travel distance and the length of stay).
Meanwhile, big data analytics facilitated unprecedented opportunities to understand hospitality and tourism phenomena by enabling the breadth and depth of studies on travel-related online platforms such as Expedia (Lee et al., 2013; Xiang et al., 2015), TripAdvisor (Schuckert et al., 2016; Xiang et al., 2018) and Yelp (Nakayama and Wan, 2019; Yang et al., 2017). Notably, big data analytics is anticipated to overcome the limitations of traditional methodological approaches (e.g. surveys and interviews) that fail to describe one’s actual behavior in research findings. Likewise, the application of machine learning algorithms permits the investigation of past and/or present phenomena and forecast traveler behavior and hidden patterns. The decision tree is one of the popular machine learning algorithms covering both regression and classification as well as the visualization effect and a straightforward interpretation of the result (Bozkir and Sezer, 2011).
The purpose of this study is to identify the significant determinants of travelers’ hotel booking behavior and the interplay with travel distance by analyzing behavioral data from the largest online travel agency (OTA), expedia.com. Overall, this study has three main research objectives. First, the study is to identify significant determinants of online hotel booking behavior – travel distance to the destination country, travel product purchased (e.g. package and individual) and other hotel features (e.g. hotel brand, star rating, distance to the hub, price and popularity). Second, the research is to analyze the result through the decision tree model to uncover the interaction between...