Open Collections

Sim, Hong-Jiang

University of British Columbia

The COVID-19 pandemic has wreaked financial devastation for airlines. As travel restrictions eases, it is important for airlines to gather and assess the opinions of passengers in order to meet new expectations. With the availability of online electronic platforms, it has become feasible to collect customer reviews on a large scale. Collecting, processing and deciphering textual data remain a challenge to most. This thesis details the collection of textual review data from SkyTrax and Trip Advisor. Subsequently, it demonstrates the use of quantitative sentiment scoring of the textual reviews, exploratory data analysis and statistical modelling to ascertain factors that affect passenger perceptions of airlines services. Specifically, we developed a linear mixed model with passengers and airline as random to rank various airlines with respect to sentiment scores, while confirming fixed factors which affect these scores. The application of this model allows for interpretability to the expectations and recommendations made by various passengers of airlines, measures the performance of airlines with respect to customer experience.

Text

2023-01-19

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/83687

Computer Science

University of British Columbia

UBCO

http://creativecommons.org/licenses/by-nc-nd/4.0/