Title of Paper

Key attributes of Airbnb experience: Text-mining and sentiment analysis

Author Bios (50 Words for each Author)

Mingming Cheng Dr Mingming Cheng is a lecturer in the Department of Tourism at the University of Otago, New Zealand. Mingming's core research interests and expertise deal with outbound Chinese tourists, sharing economy, data science, social media and inter-disciplinary research. More information can be found in his personal website: https://mingmingcheng.com/

Xin Jin Dr Xin Jin is a senior lecturer in the Department of Tourism, Sports and Hotel Management at Griffith University in Australia. Her two main areas of research interest are convention/exhibition management and destination marketing.

Abstract (150 Words)

Using a “big data” set of online review comments from Airbnb users, this research investigates the attributes forming Airbnb user experience. Both text mining and sentiment analysis were performed to analyze 170,124 review comments from Airbnb users in Sydney, Australia. Three key attributes are identified including location, amenities and host. Price does not emerge as a key attribute. While the analysis suggests a positivity bias in Airbnb users’ comments, noise receives a large number of negative sentiments. Findings also reveal that Airbnb users tend to evaluate Airbnb experience based on a frame of reference derived from past hotel stays. This research offers a more detailed understanding of the sharing economy experience in the case of Airbnb and methodologically contributes to the on-going reconfiguration of social media data by using big data techniques.

This document is currently not available here.

COinS
 

Key attributes of Airbnb experience: Text-mining and sentiment analysis

Using a “big data” set of online review comments from Airbnb users, this research investigates the attributes forming Airbnb user experience. Both text mining and sentiment analysis were performed to analyze 170,124 review comments from Airbnb users in Sydney, Australia. Three key attributes are identified including location, amenities and host. Price does not emerge as a key attribute. While the analysis suggests a positivity bias in Airbnb users’ comments, noise receives a large number of negative sentiments. Findings also reveal that Airbnb users tend to evaluate Airbnb experience based on a frame of reference derived from past hotel stays. This research offers a more detailed understanding of the sharing economy experience in the case of Airbnb and methodologically contributes to the on-going reconfiguration of social media data by using big data techniques.