Classification of Public Opinion on the Free Nutritional Meal Program on YouTube Media Using the LSTM Method

TL;DR

This study uses LSTM to classify public sentiment on YouTube comments about Indonesia's Free Nutritious Meal Program, achieving high accuracy but facing challenges with data imbalance.

Contribution

It demonstrates the effectiveness of LSTM for sentiment analysis in Indonesian social media data and highlights issues with class imbalance.

Findings

LSTM achieved 89% overall accuracy.

Negative sentiment classification had an F1-score of 0.94.

Positive sentiment classification had an F1-score of 0.55.

Abstract

Public opinion towards the Free Nutritious Meal Program (MBG) on YouTube social media reflects diverse community responses. This study applies the Long Short-Term Memory (LSTM) method to classify sentiments from 7,733 YouTube comments. The results show that the LSTM model achieves 89% accuracy, with strong performance on negative sentiment (F1-score 0.94) but weaker performance on positive sentiment (F1-score 0.55) due to class imbalance, as negative data account for 87.7% of the dataset. These findings confirm the effectiveness of LSTM for sentiment analysis of Indonesian text while highlighting the challenge of imbalanced data. This research contributes to social media-based public policy evaluation