Assessment of Hotel Performance and Guest Satisfaction through eWOM: Big Data for Better Insights

ABSTRACT

Nowadays, many customers prefer to browse over the internet to meet their travel and hotel stay needs. Electronic-word-of-mouth (eWOM) acts as an important source of information online. Customers choose hotels on the basis of comments posted online by the fellow travelers. Our aim is to predict hotel performance and guest satisfaction using eWOM and hotel features embedded in hotel eWOMs. Using a big dataset of 2,25,582 hotel eWOM from tripadvisor.com, we predict overall hotel performance and guest satisfaction using eWOM and hotel features embedded in hotel eWOM. The uniqueness of this study is lies in using technical and hotel features embedded in eWOM and investigates their influence on guest satisfaction. The relative importance of each feature with respect to guest experience has also been estimated.

KEYWORDS:

• Big data
• eWOM
• hotel performance
• guest satisfaction
• sentiment analysis
• econometric models
• machine learning
• feature importance