Analysis of Community Sentiment Towards Free Nutrition Meal Programs on Twitter Using Naïve Bayes, Support Vector Machine, K-Nearest Neighbors, and Ensemble Methods

Journal: Journal of Information Systems and Informatics

Publication: 2025-07-01 · Vol. 7 No. 2 · pp. 1443–1460

DOI: 10.51519/journalisi.v7i2.1098

Abstract

Meal program free nutritious food that was planned government reap diverse response from society, especially on social media like Twitter. Research This aiming for analyze sentiment public to the program with utilize text mining and machine learning techniques. Data of 1500 tweets was collected through the scraping process using Python. The sentiment in the tweets is classified into three categories: positive, negative, and neutral. In this study, four classification algorithms were used: Naïve Bayes, Support Vector Machine (SVM), K-Nearest Neighbors (K-NN), and ensemble, to compare their performance in sentiment analysis. Additionally, a text weighting method, TF-IDF, was tested to examine its impact on classification accuracy. The analysis results show that the Support Vector Machine (SVM) algorithm, when combined with the TF-IDF weighting method, provides the highest accuracy of 95.05%. Other algorithms also showed varied performance, with Ensemble achieving 86.57%, K-Nearest Neighbors 77.03%, and Naïve Bayes 60.42% accuracy. It is expected from results study This can give description general to perception public about the meal program free nutritious an