<![CDATA[The growing popularity of short-form video content across various social media platforms has increased the use of cross-device video editing applications, accessible through smartphones, desktops, and web-based services. CapCut is one of the most widely used applications for creating creative content, and user reviews on the Google Play Store serve as an important indicator for evaluating user experience quality. However, review datasets are often imbalanced, with positive sentiment dominating and neutral sentiment appearing in much smaller proportions, which poses challenges for sentiment classification. This study aims to analyze user sentiment toward CapCut reviews using Support Vector Machine (SVM) and applying the Synthetic Minority Over-sampling Technique (SMOTE) to address data imbalance. The data were collected by scraping reviews from the Google Play Store, resulting in 4,381 cleaned review entries after the data cleaning stage. The reviews then underwent text preprocessing, TF-IDF feature weighting, and model training. The experimental results show that the SVM model achieved an accuracy of 73.54% with a weighted F1-score of 0.736. These findings indicate that SMOTE contributes to improving model performance on minority classes. Overall, this study provides insights into user perceptions of CapCut and highlights the potential of SVM as an effective sentiment classification method for text-based application reviews.]]>
Copyrights © 2026
