Measuring Digital Destination Imagery: The Role of Naïve Bayes Sentiment Modelling in Evaluating Visitor Satisfaction from Google Maps Reviews

Universitas Muhammadiyah Magelang, Magelang, Indonesia

^* Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

Hot springs in Tempuran District are an important regional tourist attraction. Therefore, this study aims to measure public perception of five hot spring destinations, namely Umbul Banyu Roso, Tirta Madu, Lintang Water Park, Ngasinan, and Tirta Sambara, as well as identify problems that affect the visitor experience. The approach used is sentiment analysis, with data collected in the form of reviews from Google Maps. The data is then analyzed using the Naive Bayes Classifier Algorithm through the RapidMiner application. The results of the analysis showed that Tirta Sambara Hot Springs dominated with perfect positive sentiment (1,000) and the model curation reached 100%, while Ngasinan Hot Springs was dominated by negative reviews (0.650), and Tirta Madu showed balanced sentiment (0.500 positive and 0.500 negative). The implications of these results are reinforced by Word Cloud visualizations, which show that although tourist attractions are rated as “Convenient” and “Cheap”, negative sentiment is often driven by ethical issues such as “Illegal fees” and operational issues such as “Difficult access” and “Poor maintenance”, implying the need for management interventions focused on improving the integrity of services and infrastructure.