<![CDATA[The tourism industry is a rapidly growing sector that significantly contributes to the economy, including Indonesia. One of the popular tourist destinations in Indonesia is Senggigi, located on the island of Lombok. This destination offers high natural and cultural appeal. In the tourism industry, hotels are crucial as primary accommodations for travelers to stay and rest. Tourist reviews on hotel services greatly influence potential visitor’s decisions in selecting the right accommodation. Therefore, sentiment analysis of hotel reviews is essential for understanding customer satisfaction levels and assisting hotel managers in improving service quality. This research applies a comparative quantitative approach using Decision Tree and Support Vector Machine (SVM) algorithms. The dataset consists of 6,920 hotel reviews collected from TripAdvisor platforms through web scraping techniques. Data preprocessing included data cleaning, case folding, tokenization, stop word removal, and stemming to enhance classification performance. Sentiment labels were categorized into positive, neutral, and negative classes. Model performance was evaluated using multiple metrics, including accuracy, precision, recall, and F1-score, to ensure a comprehensive assessment. The word frequency distribution reveals that that accommodation experience and room quality play a crucial role in customer satisfaction. Positive sentiment is dominated by adjectives like great, nice, and beautiful, reflecting pleasant experiences. Negative sentiment is expressed more politely through phrases such as not good or not very nice. Neutral sentiment tends to be descriptive without strong emotional expression. In terms of model performance, SVM outperformed the Decision Tree model, achieving an accuracy of 90%, precision of 91%, recall of 90%, and an F1-score of 85%. In comparison, the Decision Tree achieved an accuracy of 87%, precision of 84%, recall of 87%, and an F1-score of 85%. These findings demonstrate the superior capability of SVM in handling complex and diverse textual data. This study contributes academically by strengthening empirical evidence on the effectiveness of machine learning–based sentiment analysis in the tourism domain and practically by providing actionable insights for hotel managers to improve service quality and customer satisfaction.]]>
Copyrights © 2026
