A study utilized the Long Short-Term Memory (LSTM) method to analyze public opinion on Indonesia's Free Nutritional Meal Program using 7,733 YouTube comments. The LSTM model achieved 89% accuracy in classifying sentiments, demonstrating strong performance for negative comments but facing challenges with positive sentiment due to data imbalance. This research highlights LSTM's effectiveness for Indonesian text sentiment analysis and its contribution to evaluating public policy through social media. AI
Summary written by gemini-2.5-flash-lite from 2 sources. How we write summaries →
IMPACT Demonstrates LSTM's utility for social media sentiment analysis in public policy evaluation.
RANK_REASON Academic paper on applying an LSTM model for sentiment analysis of social media comments.