Sentiment Analysis of Indonesia's Economic Acceleration Program 2025 Using Support Vector Machine

Abstract

Indonesia's Economic Acceleration Program 2025 is a government policy to accelerate national economic growth in response to the global economic slowdown. The implementation of this program has generated a variety of public responses on social media. This study aims to analyze public sentiment towards the Indonesian Economic Acceleration Program in the launch and implementation phases and identify differences in sentiment distribution in both phases. Data in the form of TikTok comments was collected through web scraping and then processed through preprocessing, lexicon-based sentiment labeling validated using manually labeled samples, TF-IDF feature representation, and classification using the Support Vector Machine. Model evaluation was carried out using 10-fold cross-validation. The results of the study showed that SVM provided superior performance to the comparison model. In the launch phase, SVM achieved an accuracy of 81%, while in the implementation phase it achieved an accuracy of 79.5% with superior performance in all evaluation metrics. The distribution of sentiment in the launch phase was dominated by neutral sentiment by 67.1%, while in the implementation phase the proportion of negative sentiment increased to 46.38%. These results show that there is a difference in the distribution of sentiment between the launch and implementation phases, so they can be an input in understanding the public's response to the Indonesian Economic Acceleration Program.