a Sentiment Analysis of Free Meal Plans on Social Media using Naïve Bayes Algorithms

Abstract

This study analyses public sentiment towards the "Free Meal Plan" initiative introduced by the political pair Prabowo-Gibran. This policy aims to assist underprivileged communities in Indonesia and is a significant issue in the social and political context. Data was collected from the social media platform X (formerly Twitter), gathering 501 relevant comments based on their connection to the topic and high levels of engagement (such as retweets and likes). The comments were then processed using Text Preprocessing and TF-IDF techniques and applied to a Naïve Bayes model. The model achieved an accuracy of 69.3%, a precision of 72%, a recall of 57.05%, and an F1 score of 54.5%. These results indicate that the model is capable of classifying public sentiment, though it has challenges in accurately detecting negative sentiment. These findings provide valuable insights for policymakers to design more effective communication and policy strategies, particularly in addressing criticism or public dissatisfaction. The study highlights the importance of using text processing and machine learning techniques to analyze social media data in a structured way.