Artificial Intelligence in Hospitality: A Content and Sentiment Analysis of Guest Reviews. The Case of Las Vegas

Petrovic, Lea (2025) Artificial Intelligence in Hospitality: A Content and Sentiment Analysis of Guest Reviews. The Case of Las Vegas. [Laurea magistrale], Università di Bologna, Corso di Studio in Tourism economics and management [LM-DM270] - Rimini, Documento full-text non disponibile
Il full-text non è disponibile per scelta dell'autore. (Contatta l'autore)

Abstract

Artificial intelligence (AI) has become increasingly integrated into hotel service delivery, yet not much is known about how guests perceive these technologies in real review environments. This study analyzed 778 AI-mention reviews across three major platforms from eleven Las Vegas hotels (2022-2025) to examine sentiment patterns, compare AI-mention reviews with matched controls, and identify the thematic contexts in which AI appears in guest discourse. A mixed-methods approach was used combining NLP with statistical analysis and following several steps (sentiment classification, propensity-score matching to compare AI-mention reviews with paired non-AI reviews, topic modelling through BERTopic. Overall sentiment toward AI was mixed but slightly positive (43.7% positive, 39.3% negative), though this average concealed large contextual differences. Alexa, Ivy and Rose AI systems were evaluated positively (60-77% positive), while automation in operational areas such as housekeeping, maintenance, and check-in kiosks generated strongly negative sentiment (76-84%). Platform differences were significant, with Google Reviews showing the most favorable attitudes and TripAdvisor the most adverse. Comparative analysis showed that AI-mention reviews were modestly more positive than matched controls, with effects strongest on Google Reviews. Content analysis revealed that most AI references (56.4%) appeared within general hotel evaluations rather than as standalone assessments, indicating that AI functions primarily as a backbone to hotel experience. Explicit naming of AI systems was rare (6%), occurring mostly in positive contexts. The findings suggest that AI perception in hospitality is context dependent. Guests respond favorably to interactive, value-adding systems and negatively when AI appears within narratives of operational dissatisfaction. Implementation quality and integration with broader service performance play a decisive.

Abstract
Tipologia del documento
Tesi di laurea (Laurea magistrale)
Autore della tesi
Petrovic, Lea
Relatore della tesi
Scuola
Corso di studio
Ordinamento Cds
DM270
Parole chiave
Artificial intelligence, Las Vegas, Content analysis, Sentiment Analysis, Transformers, BERT
Data di discussione della Tesi
16 Dicembre 2025
URI

Altri metadati

Gestione del documento: Visualizza il documento

^