CLASSIFICATION OF PUBLIC SENTIMENT TOWARDS STUNTING PREVENTION PROGRAM USING NAÏVE BAYES AND SUPPORT VECTOR MACHINE ON X APPLICATION

Indonesia has serious health problems, one of which is stunting among children. Stunting is caused by chronic malnutrition that affects a child's physical and cognitive growth. To address its impact, the government launched a prevention program that focuses on improving nutrition, improving sanitation, and health education. Public response to these programs has varied, with some supportive and others skeptical. In the digital age, public opinion is expressed through social media, making sentiment analysis important to understand public perception. This research aims to classify public sentiment towards the stunting prevention program using Naive Bayes and Support Vector Machine (SVM) methods. Data preprocessing includes cleaning, case folding, tokenizing, stopwords, and stemming, ensuring the text data is ready for analysis. The dataset consists of 5907 tweets divided by a ratio of 80:20, resulting in 4725 tweets for training data and 1182 tweets for testing data. The analysis results show that the Naive Bayes model achieved an accuracy of 95.34% for training data and 84.52% for testing data, while SVM achieved an accuracy of 95.43% for training data and 96.74% for testing data, indicating the performance of the SVM model is better than the Naïve Bayes model. The important impact of this research is to assist policymakers in understanding the public's perception of government programs so that they can design communication strategies.