Sentiment Analysis of Twitter towards the 2024 Indonesian Presidential Candidates Using the Naïve Bayes Algorithms

Abstract

The increasing use of social media (Twitter) has made it a platform for the public to express their views on the Indonesian presidential candidate in the 2024 elections. The sentiment expressed through comments on Twitter provides important insights into the public perception of the candidates. However, given the volume and speed at which information is disseminated on social media, manual analysis of this sentiment becomes impractical. Therefore, the use of the Naïve Bayes algorithm for automatic sentiment analysis is considered essential to understanding voter support and preferences. The study aims to analyze Twitter users' sentiments towards three Indonesian presidential candidates in 2024, Anies, Ganjar, and Prabowo, using the Naïve Bayes algorithm. We categorize the results of this analysis into three sentiment categories: positive, negative, and neutral. The methods used in the study involved collecting Twitter comment data related to the three candidates, pre-processing data, labeling data, applying the Naïve Bayes algorithm for the classification of sentiment, and evaluation of the performance of the algorithm performed by calculating the level of accuracy. The results of the research showed that the Naïve Bayes algorithm was able to classify sentiments with fairly high precision, namely 75.54% for Anies, 82.74% for Ganjar, and 75.24% for Prabowo. The conclusion of this study is that sentimental analysis using the Naïve Bayes algorithm can provide significant insights into voter preferences and support. The sentimental data generated can serve as a strong foundation for decision-makers to design campaign strategies that are more effective and responsive to public perception. This research also opens up opportunities for further development in the use of sentimental analysis techniques in politics and campaigns.