Classification of Student Responses to the Free Nutritional Meal Program at SMAK Giovanni Kupang UsingDeep Learning

The Free Nutritious Meal Program (MBG) is a government initiative aimed at improving nutritional quality and supporting students’ learning processes in schools. The implementation of this program at SMAK Giovanni Kupang has generated various student responses, making it necessary to develop a system that can automatically classify these perceptions. This study aims to classify students’ responses to the MBG program using a deep learning approach based on Long Short-Term Memory (LSTM). The data used consists of 1,238 student comments collected through questionnaires and processed through several text preprocessing stages, including cleaning, tokenization, stopword removal, and padding. The LSTM model is employed to classify sentiment into three categories: positive, neutral, and negative. Evaluation is conducted using the K-Fold Cross Validation method to measure the model’s performance and consistency. The results show that the developed model achieves an average accuracy of 0.7036 (70.36%), with fold 4 being the best-performing fold, achieving an accuracy of 0.7611 (76.11%). The model performs well in classifying positive sentiment but still faces limitations in distinguishing between neutral and negative sentiments due to data imbalance and the ambiguous nature of the text.