Digital is always bringing a distinctive way to a communication and interaction between user through multiple platforms. The most popular social media platform for getting information and entertainment in Indonesia is YouTube. In Indonesia, news information topics related to government activity were also actively circulated on the YouTube, and it drew interest from each viewer. This research aims to analyze the dataset from YouTube and extract insightful information related to viewer perception of the topic and how the interaction between it. Data collected by crawling data from YouTube Application Interface and using the most upcoming topic keyword which is 2024 Indonesian presidential election candidate. Preprocessing data were handled by using multiple processes like tokenization, stop word and emoji removal, lemmatization, and pos tagging. Relevant data from the dataset like comments and replies were ingested by various machine learning algorithms to classify the user's perception and the interaction between users were visualized in the circular graph model. The findings revealed that there are more negative opinions related to the candidate and in the network also found that many users are connected when doing a counter-measure comments. Future research can adopt the present findings with another election topic on different social media especially that have more deep nested interactions to generate an even more comprehensive analysis.