<![CDATA[In this era of rapid development of information technology, the number of internet users is increasing, supported by the popularity of social media as a medium for sharing information and interacting. The X social media platform is one of the media that is often used to convey public opinion. One of the hot issues discussed on X social media is the Indonesian National Team naturalization program. This program has triggered various public responses, both pro and con. This study aims to analyze public sentiment regarding the program using the Bidirectional Encoder Representation from Transformer (BERT) algorithm with the Knowledge Discovery in Database method. Data was collected using scraping techniques on the X social media platform which were then selected and labeled positive, negative, and neutral. Testing the BERT algorithm using the pre-trained indoBERT model was tested by dividing the training and testing data 80:20, and evaluated with a confusion matrix. With a dropout of 0.3, the evaluation results showed the highest accuracy value of 90%, precision 81%, recall 74%, and f1-score 77%. The results of this study are expected to be useful for evaluation materials and to support decision making by related parties.]]>
Copyrights © 2025
