Sentiment Analysis of Instagram Users Toward the Animated Film Merah Putih: One For All Using the Naïve Bayes Method

Abstract

Instagram is a widely used social media platform where users express opinions on various forms of entertainment, including animated films. The animated film Merah Putih One For All, as one of Indonesia’s local animation works, has received diverse responses from Instagram users that reflect positive, neutral, and negative sentiments. This study aims to analyze and classify the sentiment of Instagram user comments related to the film using the Naïve Bayes algorithm. This research utilized 200 Instagram comments, which were categorized into three sentiment classes. Text preprocessing was applied prior to classification. The dataset was evaluated using the Split Validation method with a 60:40 ratio, where 60% of the data were used for training and 40% for testing. Model performance was assessed using a confusion matrix along with accuracy, precision, and recall metrics. The experimental results show that the Naïve Bayes algorithm achieved an accuracy of 76,67%. The positive sentiment class obtained the highest recall value of 100%, followed by the neutral class with 83,33%, while the negative sentiment class recorded the lowest recall at 46,67%. These findings indicate that the model performs better in identifying positive and neutral sentiments than negative sentiment. Overall, the results demonstrate that the Naïve Bayes algorithm is sufficiently effective for sentiment analysis of Instagram comments, although further improvements are required to enhance the classification of negative sentiment.