Sentiment Analysis and Hate Speech in Online News About IKN Using the K-NN Algorithm

Abstract

Online news about Ibu Kota Nusantara (IKN) has sparked diverse public reactions, particularly regarding the capital relocation, a highly sensitive topic. The spread of information shapes public perception, especially when news contains hate speech, which can damage IKN’s reputation. This study applies sentiment analysis to online news about IKN using the K-Nearest Neighbor (KNN) algorithm. Data were gathered from Google News (595 articles) and YouTube (398 videos) and classified into four categories: positive, negative, neutral, and hate speech. The results show that Google News achieved 100% accuracy, while YouTube data reached 88.19% at K=3. These findings suggest that Google News articles are easier to classify with KNN compared to YouTube content, highlighting differences in text structure and characteristics between platforms.