<![CDATA[Tourism is a strategic sector that plays an important role in supporting regional economic development, including in the Special Region of Yogyakarta as one of Indonesia’s leading tourist destinations. High tourism activity has led to an increasing number of visitor reviews on online platforms such as TripAdvisor, which can be utilized to understand tourists’ perceptions and satisfaction levels. The large volume of reviews makes manual analysis ineffective, thus requiring an automated, computation-based approach. This study aims to analyze the sentiment of tourism reviews in Yogyakarta City using the Support Vector Machine (SVM) algorithm with the addition of the Linear Discriminant Analysis (LDA) dimensionality reduction method. The dataset consists of 5,000 tourism reviews that have undergone preprocessing stages, including data cleaning, case folding, tokenization, normalization, stopword removal, and stemming. Text feature representation was performed using the TF-IDF method. Model evaluation was conducted using an 80:20 split between training and testing data, with performance measured using accuracy, precision, recall, and F1-score metrics. The results indicate that the application of LDA to the SVM model improves the balance of classification performance, particularly in precision (77.71%) and F1-score (79.5%), despite a slight decrease in accuracy. Sentiment classification results are dominated by positive sentiment (92.4%), reflecting generally favorable tourist perceptions of destinations in Yogyakarta. This study is expected to serve as a reference for tourism managers in evaluating service quality and facilities based on visitor opinions.]]>
Copyrights © 2026
