Deciphering Political Discourse on X: A Deep Dive into SVM vs. Naïve Bayes Approaches

social media plays a significant role in society, influencing people's opinions and views. In this study, in this research focus on X as a platform for sentiment analysis and topic modelling. The objective is to analyse and categorize the opinions of society on the issues that occurred when Joko Widodo became president of Indonesia. The study utilizes the Support Vector Machine (SVM) method for sentiment classification and the Latent Dirichlet Allocation (LDA) method for topic modelling. A total of 1200 tweets were collected, classified into 600 negative and 600 positive tweets. The results demonstrate the effectiveness of the sentiment classification system and topic modelling. Accuracy was 86.39%, precision was 85.75%, recall was 85.86%, and the F1 score was 85.79% with a total of 3236 features. The LDA method produced perplexity values of 7.0693 for positive sentiment and 7.2897 for negative sentiment. This study highlights the potential of sentiment analysis and topic modelling for understanding public opinions and fostering social change. The system has the potential to enhance the decision-making process of policymakers and contribute to the overall well-being of society.