In recent years, the development of automotive industry technology has made significant progress. Many vehicles are being produced using electric energy as an environmentally friendly alternative. The use of electric vehicles has become a hot topic in society, sparking various reactions and opinions on the social media platform Twitter. This research aims to analyze the sentiments on the public opinion of electric vehicles usage in Indonesia, using data collected from Twitter. Sentiment analysis was performed using machine learning approaches and Support Vector Machine classification methods to categorize each comment as positive or negative. To address imbalanced data, the Synthetic Minority Over-Sampling Technique (SMOTE) was used for oversampling, and the Random under-sampling (RUS) method was employed for undersampling. After the classification process and performance evaluation, the best model selected was the baseline of Support Vector Machine model with a data split ratio of 70:30 without applying imbalance handling techniques. This model achieved excellent results, with an accuracy of 94,8%, precision of 95,5%, recall of 99,1%, and F-1 Score of 97,2%.
Copyrights © 2023