ANALISIS SENTIMEN TERHADAP PASANGAN CAPRES-CAWAPRES PEMILU 2024 DI MEDIA SOSIAL TWITTER MENGGUNAKAN METODE NAIVE BAYES

In 2024, Indonesia will hold a general election that includes the election of the President and Vice President for the period 2024-2029. The purpose of this research is to analyze public sentiment towards presidential and vice presidential candidate pairs on the Twitter social media platform in the context of the 2024 General Election and evaluate the performance of the PSO-based Naive Bayes classification method in identifying sentiment towards presidential and vice presidential candidate pairs in the 2024 General Election on Twitter. The results of sentiment analysis show that there are variations in people's responses to each candidate pair. While candidate pairs 1 and 2 experienced an almost equal ratio of positive and negative sentiment, candidate pair 3 showed a more positive response with a significant dominance of positive sentiment. This reflects the complexity of the issues discussed and the diverse responses from the public towards each candidate pair. The Naïve Bayes method without PSO produces an accuracy of 61.71%, while Naïve Bayes with PSO gives an accuracy of 64.50%. The use of PSO selection features provides a slight improvement in the performance of the Naïve Bayes model in predicting sentiment from twitter data.