In this research, an analysis of public sentiment towards presidential candidate pairs in Indonesia in 2024 was carried out via the social media Instagram. Indonesia itself is one of the countries with the highest number of Instagram users. One of the Instagram posts that is currently in the public spotlight found on the kompascom, najwashihab, and detikcom accounts which post news about the 2024 presidential candidate pairs. In these comments there are various kinds of comments ranging from positive things such as support to negative things such as commenting on the shortcomings of each candidate pair, for this reason text classification is carried out to find out public opinion about the presidential candidate pair. The comment classification process in this research uses the Naïve Bayes algorithm to determine positive, neutral and negative values from thousands of comments. The method that will be applied uses the Python programming language with confusion matrix testing to determine the level of accuracy in the model. Based on the test results, it can be concluded that the use of the Naïve Bayes algorithm used as a classification method in comment-based sentiment analysis on Instagram has a relatively good accuracy rate with an average accuracy of more than 60%.
Copyrights © 2024