Understanding Customer Perception and Brand Equity in the Hospitality Sector: Integrating Sentiment Analysis and Topic Modeling

This study integrates sentiment analysis and topic modeling techniques to investigate customer perception and brand equity within the hospitality sector. Drawing upon data from Booking.com in Ho Chi Minh City, Vietnam, an array of machine learning (ML) and deep learning (DL) models, including logistic regression (LR), random forest (RF), multinomial Naive Bayes (NB), Dense, long short-term memory (LSTM), and convolutional neural network (CNN), were employed. Notably, the Dense model exhibited a notable degree of accuracy, with the best-performing model achieving an accuracy rate of 0.95. Furthermore, the Dense model showcased an impressive F1-score of 0.97, underscoring its efficacy in sentiment classification. Using latent Dirichlet allocation (LDA), topic modeling analysis identified significant themes encompassing noise levels, room conditions, facilities, staff performance, and location advantages. These findings enhance our comprehension of the variables that shape brand equity and provide tangible implications for hotel managers seeking to enhance service quality, elevate customer satisfaction, and fortify their brand’s competitive position. Overall, this study highlights the effectiveness of data-driven approaches in attaining insights into customer perceptions and serves as a guiding resource for future research endeavors in hospitality.