Studi Perbandingan: Algoritma Random Forest, Naive Bayes Dan Support Vector Machine Dalam Analisis Sentimen Pada Aplikasi Capcut Di Google Play Store

Abstract

CapCut, a highly popular video editing tool, boasts millions of users worldwide across various age groups. Posting reviews on the Google Play Store can provide valuable insights into this application. This study aims to evaluate the effectiveness of three classification algorithms Random Forest, Naïve Bayes, and Support Vector Machine in performing sentiment analysis on Google Play Store reviews of the CapCut application. User reviews are identified and categorized into positive, negative, and neutral labels using sentiment analysis methods. A total of three thousand user review datasets were employed in this investigation. The research procedure involved data preprocessing, feature extraction, and model training. The results show that the Random Forest classification method achieved 83% accuracy, the Naïve Bayes method 70% accuracy, and the Support Vector Machine method 86% accuracy, indicating user sentiment towards the CapCut application. With an accuracy of 0.86, the SVM algorithm is found to yield the best results