Sentiment Analysis of Public Responses Regarding The Use of Electric Cars in Indonesia with Support Vector Machine and Random Forest Methods

Abstract

The diminishing use of fossil fuels has encouraged the search for alternative energy sources, one of which is the electric car. However, public acceptance of electric cars in Indonesia is not widely understood. This study aims to analyze public sentiment towards electric cars based on data from X social media. The dataset used consists of 3,450 data, which is analyzed using two machine learning methods, namely Support Vector Machine (SVM) and Random Forest. The research was conducted in three scenarios: SVM kernel comparison, Random Forest performance evaluation with various numbers of n-estimators (1, 10, 100), and performance comparison between the two methods. The experimental results show that Random Forest with 100 n-estimators produces the highest accuracy of 90.72% and F1-Score of 87.54%, while SVM with RBF kernel produces 89.35% accuracy and F1-Score of 85.15%. The performance difference of 1.37% shows that Random Forest is more effective in this sentiment analysis