Sentiment Analysis of 'Free Lunch for Children' Policy on Social Media X Using Random Forest Algorithm

Journal: Journal of Information Systems and Informatics

Publication: 2025-03-22 · Vol. 7 No. 1 · pp. 649–662

DOI: 10.51519/journalisi.v7i1.1039

Abstract

The concept of a welfare state emphasizes the main role of the government in providing protection and improving welfare such as health and education to its people. The free lunch program proposed by Prabowo Subianto and Gibran Rakabuming Raka aims to improve the nutritional quality of school children while driving the national economy. The public's reaction to Prabowo Subianto's work program plan, on free school lunch program and nutritional support for Indonesian students, is very diverse in perspective on X. The Random Forest algorithm proved to be quite effective in classifying public sentiment regarding the policy of “Free Lunch for Children.” With an overall accuracy of 73%, the model was able to categorize public opinion into positive, negative, and neutral categories. To improve the performance of the model, upsampling was performed to balance the classes in the dataset as well as hyperparameter tuning. After applying these techniques, the accuracy of the model increased significantly to 80%.