Analyzing Customer Satisfaction Through Online Reviews Using Topic Modeling and Linguistic Model: A Study of a Hotel Chain in Madrid

Shu, Ziwei; Sánchez Figueroa, María Cristina

Publicación:
Analyzing Customer Satisfaction Through Online Reviews Using Topic Modeling and Linguistic Model: A Study of a Hotel Chain in Madrid

dc.contributor.author	Shu, Ziwei
dc.contributor.author	Sánchez Figueroa, María Cristina
dc.contributor.editor	Springer
dc.date.accessioned	2025-07-04T16:54:43Z
dc.date.available	2025-07-04T16:54:43Z
dc.date.issued	2025
dc.description	The registered version of this article, first published in “Systems and Technologies (SIST, volume 439)", is available online at the publisher's website: Universidad nacional de Educación a Distancia, https://10.1007/978-981-96-3081-3_20 La versión registrada de este artículo, publicado por primera vez en “Systems and Technologies (SIST, volume 439)", está disponible en línea en el sitio web del editor: Universidad nacional de Educación a Distancia, https://10.1007/978-981-96-3081-3_20
dc.description.abstract	Abstract With the widespread use of social media and review platforms, Electronic Word of Mouth (eWOM) has a greater reach and influence than traditional Word of Mouth (WOM), becoming a vital information source for customers and companies. Online reviews, a representative type of eWOM, are essential for guiding customer hotel choices and helping hotel managers gain valuable insights for necessary improvements. Mining online reviews and sentiment analysis are popular research areas, where sentiment analysis typically classifies the overall sentiment of customer reviews as positive, neutral, or negative. This study presents an approach for analyzing hotel customer satisfaction using the Latent Dirichlet Allocation (LDA) model and the 2-tuple linguistic model. The LDA model, widely used for topic modeling, is employed to identify the topics that matter more to customers and influence their satisfaction. The 2-tuple linguistic model is applied in sentiment analysis to address information loss when converting compound scores into sentiment labels, offering a clearer and more intuitive representation of sentiment while retaining the information. The proposed model analyzes 26,740 TripAdvisor reviews from twenty hotels in Madrid that are part of the same hotel chain. The results highlight the importance of considering variations in hotel scores and rankings across different languages, as identical rankings do not always correspond to the same sentiment scores from English-speaking and Spanish-speaking reviews, and vice versa. This study contributes to online reviews research and hotel management by offering a clearer and more accurate expression of customer review sentiment through 2-tuple values, and the most frequent topics from reviews.	en
dc.description.version	versión publicada
dc.identifier.citation	Shu, Ziwei; Sanchez Figueroa, Maria Cristina: "Analyzing Customer Satisfaction Through Online Reviews Using Topic Modeling and Linguistic Model: A Study of a Hotel Chain in Madrid. Smart Innovation", Systems and Technologies (SIST,volume 439), https://10.1007/978-981-96-3081-3_20
dc.identifier.doi	https://10.1007/978-981-96-3081-3_20
dc.identifier.issn	2190-3018, eISSN 2190-3026
dc.identifier.uri	https://hdl.handle.net/20.500.14468/29241
dc.journal.issue	439
dc.journal.title	Smart Innovation, Systems and Technologies
dc.journal.volume	2
dc.language.iso	en
dc.page.final	330
dc.page.initial	315
dc.publisher	Springer
dc.relation.center	Facultad de Ciencias Económicas y Empresariales
dc.relation.department	Economía aplicada y estadística
dc.rights	info:eu-repo/semantics/closedAccess
dc.rights.uri	Atribución-NoComercial-SinDerivadas 4.0 Internacional
dc.subject	53 Ciencias Económicas
dc.subject	33 Ciencias Tecnológicas
dc.title	Analyzing Customer Satisfaction Through Online Reviews Using Topic Modeling and Linguistic Model: A Study of a Hotel Chain in Madrid	en
dc.type	artículo	es
dc.type	journal article	en
dspace.entity.type	Publication
relation.isAuthorOfPublication	ecf9447f-9fba-4d53-acbc-29d1e49274e8
relation.isAuthorOfPublication.latestForDiscovery	ecf9447f-9fba-4d53-acbc-29d1e49274e8