<![CDATA[Rapid advances in digital technology have led to an increase in the amount of text data available online, including user reviews of mobile applications. The Loklok application, as a popular entertainment platform, is one source of review data that is rich in user opinions. This research focuses on performing sentiment analysis on user reviews of the Loklok application by employing the Support Vector Machine (SVM) algorithm alongside the Term Frequency-Inverse Document Frequency (TF-IDF) method for feature extraction. The review dataset was sourced from the Kaggle platform and underwent several text preprocessing steps, including data cleaning, tokenization, stopword elimination, and stemming. The evaluation results indicate that the SVM model, combined with TF-IDF, achieved an accuracy of 86%, a precision of 88%, a recall of 86%, and an F1-score of 87%. Classification performance tends to be better for positive sentiment classes compared to negative ones, due to data imbalance. This finding demonstrates that the combination of TF-IDF and SVM methods is effective in classifying user review sentiment and can serve as a foundation for decision-making in the development of digital applications.]]>
Copyrights © 2025
