Sentiment Analysis of Public Response to the Free Nutritious Meal Program on Instagram Using IndoBERT

Abstract

The Free Nutritious Meal (MBG) Program, as one of the government’s social initiatives, has attracted diverse discussions on social media, particularly Instagram. This study aims to analyze public sentiment toward the program by applying a machine learning approach based on IndoBERT. The research data were collected from user comments on Instagram posts related to the MBG program. The methodological framework includes data collection, text preprocessing, sentiment labeling, as well as the training and evaluation of the IndoBERT model. The results indicate that IndoBERT successfully classifies public sentiment into three categories: positive, negative, and neutral, with a high level of accuracy. Furthermore, the analysis shows that most public responses reflect positive sentiment, although neutral opinions and criticism are also present. In conclusion, IndoBERT is effective in capturing patterns of public opinion on social media and provides valuable insights for policymakers in evaluating and improving the implementation of the MBG program.