Public Opinion Analysis on Free Nutritious Food through Random Forest-Based Sentiment Classification

Abstract

The Free Nutritious Meal Program is a social policy aimed at improving public access to nutritious food, particularly for vulnerable groups. The implementation of this program has generated various public responses that are widely expressed on social media. These opinions can serve as valuable information to understand public acceptance of the policy. This study aims to analyze public opinion toward the Free Nutritious Meal Program through sentiment classification using the Random Forest algorithm. The dataset consists of 1,000 Indonesian-language comments collected from social media and categorized into positive, negative, and neutral sentiments. The research stages include data collection, labeling, text preprocessing, feature extraction using TF-IDF, data splitting, Random Forest classification, and model evaluation. The results show that Random Forest can effectively classify public opinion by combining multiple decision trees to produce more stable predictions than a single-tree model. The model achieved an accuracy of 94%, precision of 0.93, recall of 0.92, and F1-score of 0.92. These findings indicate that Random Forest is a relevant method for sentiment analysis of public policy based on social media data.