Automated analysis ot tourist sentiments towards sustainability and health secure practices

Abstract

Existing literature on sentiment analysis of hotel experiences has primarily emphasized market intelligence. However, the impact of health-related and sustainable practices on consumer emotions remains underexplored, despite their importance in post-COVID tourism. Thus, this study employs a deep learning-based method to analyze sentiment and emotions in hotel reviews. By leveraging advanced neural network architectures we are able to accurately capture the semantic meaning and contextual nuances of textual data. Our research examines 2,801 online reviews from 20 hotels listed on Booking.com, aiming to uncover the underlying sentiment (ranging from very negative to very positive) and specific emotions (joy, neutral, sadness, surprise, anger, fear, and disgust) conveyed in the reviews. The significance of communicated health and sustainability practices is reflected in the number of comments mentioning them: 157 comments (4.5% of the sample) mention health-related practices, and 23 comments (almost 1% of the sample) mention sustainable practices. Positive sustainability-related comments discussed a range of topics, nonetheless insufficiently covering all aspects of sustainability in hospitality. Notably, negative comments were also made where hotels were not implementing some simple yet highly contentious topics such as use of plastic. The most common negative comment regarding health was the lack of enforcement of COVID-19 measures, especially social distancing, conversely reviewers praised sound implementation of COVID-19 regulations. Comments expressing joy and neutral emotions appeared in comments with higher overall review ratings than those expressing anger. Further nuanced automated sentiment analyses of online content in tourism is necessary to better understand evolving consumer demand and to further improve sustainability and wellness within hospitality and tourism.