Foreign Visitors’ Dining Experiences in Asian Restaurants Operating in Istanbul

Abstract

The purpose of this study is to reveal the dimensions of dining experience by modeling the foreign\rvisitors reviews for Asian restaurants operating in Istanbul, Turkey. In the study, the latent\rDirichlet allocation (LDA) algorithm, sentiment analysis, dimensional salience and valence\ranalysis (DSVA), and lexicon salience and valence analysis (LSVA) were used as text mining\rmethods to analyze 3,843 online English reviews for Asian restaurants on TripAdvisor. Five\rdimensions were found for the Asian restaurants experience: authenticity, staff, sushi, service, and\rview. According to the dimensional salience analysis, the authenticity forms the core of the Asian\rrestaurants experience. As a result of the dimensional valence analysis, the staff has a highly\rpositive valence and the service has a highly negative valence dimension. As a result of lexicon\rsalience analysis based on the SVM estimation, the most salience term was food, while the least\rsalience term was Bosphorus. Bosphorus, delicious, and amazing were determined as the terms\rwith the highest positive valence by lexicon valence analysis, respectively. The results may also\rbe of interest to various gastronomy stakeholders interested in the Asian restaurant experience\rfrom a customer perspective.