<![CDATA[The development of digital technology has fundamentally transformed the way society communicates and consumes information, particularly through social media. Amidst the rapid and massive flow of information, sentiment analysis has become an essential tool for understanding public opinion. This study explores the use of machine learning as an analytical approach to identify and classify users' sentiments toward specific issues on social media. Through case studies on Twitter, Facebook, TikTok, and Instagram, machine learning algorithms such as Naive Bayes and Support Vector Machine were used to map public sentiment trends positive, negative, or neutral toward specific communication campaigns. The results indicate that machine learning can provide a faster, more accurate, and more dynamic sentiment analysis compared to manual methods. These findings serve as a strategic foundation for communication practitioners in designing more targeted, responsive, and data-driven messages. Thus, integrating machine learning into digital communication strategies not only enhances the effectiveness of message delivery but also strengthens the relationship between institutions and the public in an increasingly complex information age. ]]>
Copyrights © 2025
