| Author: | Zhou, Liting |
| Title: | User perceptions of airbnb’s online experiences? An investigation of underlying dimensions through corpus-driven sentiment analysis |
| Advisors: | Witte, Alexandra (SHTM) |
| Degree: | M.Sc. |
| Year: | 2023 |
| Subject: | Tourism -- Computer network resources Airbnb (Firm) Consumer satisfaction Hong Kong Polytechnic University -- Dissertations |
| Department: | School of Hotel and Tourism Management |
| Pages: | xiii, 156 pages : color illustrations |
| Language: | English |
| Abstract: | Airbnb Online Experiences (OEs) are seen as a survival strategy during the pandemic and potential revenue-generating channels in the future, which are even treated as a tool to get competitive advantages in India. However, despite the importance of understanding how tourists perceive them, tourists’ perceptions of India-specific Airbnb OEs are still unknown. Therefore, this research fills the gap by comparatively exploring underlying experiential dimensions of different types of OEs and users’ affective evaluation of identified dimensions. This research uses a mixed research design by combining quantitative corpus analysis and qualitative sentiment analysis. A total number of 6,618 valid customer reviews of 27 India-specific OEs are analyzed. This study categorizes Airbnb’s OEs into four types – “Wellness”, “Virtual Destination Tours”, “Food & Beverage Experiences”, and “History& Culture”. Furthermore, this study finds that eight experiential dimensions general experience dimension, host, guest, experiential Outcomes, practical & technological dimensions, interaction & social connection, culture, and virtual/remote experience dimensions with a total of thirty-three sub-dimensions were perceived as important by guests within the OEs. Next, the findings also indicate that generally positive sentiments underlying “Food & Beverage Experiences” and “History & Culture” OE types. In terms of sentiments regarding experiential dimensions, Experiential outcomes, host, guest, general experience dimension, practical & technological dimensions, and virtual/remote experiences dimensions evoked sentiments of positive valence while culture and interaction & social connection aroused both ambiguous and positive sentiments. This study contributes to the methodology by adopting a corpus-driven sentiment analysis to reduce potential subjective bias. Also, this research furthers the existing understanding of how experiential dimensions can evoke completely different sentiments within the same type of virtual tourism experiences by adopting the WN-Affect framework. It also gives insights into several sectors in the related industry. Virtual tourism service providers and hosts are recommended to pay more attention on types of OEs receiving positive emotional feedbacks; hosts can improve their services by putting extra efforts on experiential dimensions with overall positive acknowledgement by guests; and physical tourism organizations should see virtual tourism products as an effective strategy to attract potential visitors. This study has several limitations. First, guests’ experiential dimensions and sentiment patterns under OEs outside Indian-context would be still unknown. Second, whether sentiment patterns identified towards OEs might be affected by the pandemic cannot be seen. Third, identified ambiguous sentiments and those sentiments that were not identified by WN-Affect lead to the potential bias. Next, post-experience reviews have limitations in representing actual emotional feedbacks guests had during their experiences. Therefore, future research may focus on OEs outside the Indian setting to evaluate the discovered dimensions and sentiment patterns and explore potential differences regarding different cultural contexts. Furthermore, how the pandemic as a timeframe could potentially affect the identification of sentiments, as well as investigations during the post-pandemic era or normal settings, and comparative longitudinal studies in terms of multiple timeframes can be focused on. It is also recommended that to incorporate approaches (e.g., indutive coding) to decrease the constraints brought by sentiment analysis alone based on the WN-Affect framework. |
| Rights: | All rights reserved |
| Access: | restricted access |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| 6865.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 1.77 MB | Adobe PDF | View/Open |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/12417