<![CDATA[Tourism is a strategic sector contributing to regional economic growth. Although Lumajang Regency offers prominent natural destinations, data-based insights into tourist preferences remain limited. This study analyzed tourist preferences using Google Reviews through a text mining approach that integrated the density-based spatial clustering of applications with noise (DBSCAN) algorithm and lexicon-based sentiment analysis. Data were collected via web scraping from six major destinations, yielding 16,904 reviews, of which 9,800 contained analyzable text. The text data were preprocessed using the term frequency-inverse document frequency (TF–IDF) to generate numerical representations prior to clustering. Using DBSCAN with parameters ε = 0.8 and MinPts = 4, one main cluster comprising 9,353 reviews and 447 outliers was identified. The main cluster was dominated by keywords such as waterfall, beautiful, and scenery, emphasizing the visual appeal of Tumpak Sewu as Lumajang’s tourism icon, while the outliers reflected reviews from international visitors and practical travel information. Sentiment analysis showed that most reviews were positive (68.0%), followed by neutral (24.1%) and negative (7.9%). These findings indicate a predominantly positive perception of Lumajang tourism, though accessibility and facilities require improvement. The study demonstrates the potential of digital review data for developing data-driven tourism management and promotion strategies.]]>
Copyrights © 2025
