<![CDATA[Sentiment analysis determines positive, negative, or neutral sentiments or opinions contained in the text, including social media analysis. The role of social media, such as the TikTok and Instagram applications in sentiment analysis is important because these applications are places where people talk openly about various things, including facilities on the FT UMJ campus. In this study, the author took data from feed uploads and for your page from previous research, with 4,758 Instagram followers, and 2,616 TikTok followers on the FT UMJ account. The goal is that the results can be used as evaluation material for the campus regarding campus and academic facilities, as well as social media for publication. The author uses the following four steps in conducting the research methodology; data collection, data preprocessing, sentiment analysis, and data classification using the Naïve Bayes method. The author also designed the system and appearance of the application. The results are that F-1 Score, Recall, and Precision on Instagram are higher for negative labels than on TikTok. The accuracy value on Instagram is 54%, which has a better performance than TikTok which has an accuracy value of 33%. Thus, Instagram can be recommended as a priority for development.]]>
Copyrights © 2025
