Leveraging Association Rule Mining in Travelers’ Hotel Selection Preferences
| aut.embargo | No | en_NZ |
| aut.thirdpc.contains | No | en_NZ |
| dc.contributor.advisor | MacDonell, Stephen G. | |
| dc.contributor.advisor | Leong, Paul | |
| dc.contributor.advisor | Pears, Russel | |
| dc.contributor.author | Zhu, Xuan | |
| dc.date.accessioned | 2020-05-20T02:09:10Z | |
| dc.date.available | 2020-05-20T02:09:10Z | |
| dc.date.copyright | 2020 | |
| dc.date.issued | 2020 | |
| dc.date.updated | 2020-05-20T00:15:35Z | |
| dc.description.abstract | In seeking to explore and understand the hotel preferences of travelers, this study applies the association rule mining (ARM) method to the database of a hotel property management system (PMS). In particular, this study considers the associations between travelers’ hotel selection behavior and demographic factors, travel types, and hotel attributes. Most of the prior research literature that has addressed travelers’ hotel preferences is based on customers’ responses to questionnaires. Potential deficiencies in such a research methodology include the following: 1. Questionnaire responses are based on a ‘virtual’ hotel booking scenario, where the potential customer ranks the importance of a list of possible influential hotel selection factors, and the resulting indications may not accurately reflect the actual purchasing propensity of the travelers; 2. Questionnaires typically include hotel selection factors such as “Hotel Staff Service Attitude” and “Hotel Facilities Cleanliness” which cannot be known by customers when they first book a hotel. Therefore, the conclusion of such studies cannot fully reflect travelers’ preferences when booking a hotel for the first time. This study explores customers’ behavioral tendency by investigating the decisions reflected in hotel customers’ actual purchase behavior as recorded in a hotel management software database (i.e., their reservations), and eliminates the ‘unknowable’ influencing factors of the first hotel booking choice. This study screens, transforms, and preprocesses the original data in the PMS database, retaining and generating factors that have a potential impact on the customers’ hotel selection behavior. Lift is used to filter the rules that are positively associated to the choice of hotels. The results show that the ARM method can effectively identify the rules of strong correlation and can be used to explain hotel customer behavior and – potentially – predict hotel booking trends. For instance, the analysis has revealed that travel purpose influences travelers' length of stay, and when traveling in pairs, customers are less price sensitive when selecting hotels. The study thus concludes that the factors considered are indeed influential in customers’ hotel selection. The study further suggests the impact of excavated association rules on hotel marketing strategies and market segmentation strategies. | en_NZ |
| dc.identifier.uri | https://hdl.handle.net/10292/13349 | |
| dc.language.iso | en | en_NZ |
| dc.publisher | Auckland University of Technology | |
| dc.rights.accessrights | OpenAccess | |
| dc.subject | Association rules mining | en_NZ |
| dc.subject | Data mining | en_NZ |
| dc.subject | Traveler behavior | en_NZ |
| dc.subject | Customer preference | en_NZ |
| dc.subject | Hotel selection preference | en_NZ |
| dc.title | Leveraging Association Rule Mining in Travelers’ Hotel Selection Preferences | en_NZ |
| dc.type | Thesis | en_NZ |
| thesis.degree.grantor | Auckland University of Technology | |
| thesis.degree.level | Masters Theses | |
| thesis.degree.name | Master of Philosophy | en_NZ |