<![CDATA[The Tapera program, a long-term savings initiative for housing financing, particularly for low-income communities, has faced predominantly negative sentiment from the public on social media, especially Twitter. This study employs the Naïve Bayes Classifier (NBC) method to analyze the sentiment of Indonesian citizens' tweets regarding the Tapera program. Of the 2600 tweets analyzed, 57.35% expressed negative sentiment, while 42.65% showed positive sentiment. The model evaluation revealed an average accuracy of 68.4% for the NBC, with the highest accuracy reaching 73.3% at an 80:20 training-testing data split ratio and the lowest at 61.9% at a 10:90 ratio. This study provides an in-depth overview of the public's response to the Tapera program on social media and confirms the effectiveness of the Naïve Bayes Classifier in classifying sentiment from Twitter tweets. The findings suggest the need for more effective communication strategies and policy improvements that are more responsive to public expectations and needs.]]>
Copyrights © 2025
