극우 유튜브 채널에서의 비시민성 및 불관용 표현 기반 네트워크 분석: 전염효과와 선택효과를 중심으로

초록

This study compares and analyzes the network structures and temporal dynamics of incivility and intolerance expressed in YouTube comments during the politically contentious period surrounding the impeachment of former President Yoon Suk-yeol. By tracking overlapping user activity across far-fight YouTube channels, the research constructs inter-channel networks and examines weekly changes in density, centrality, and modularity. Based on approximately 8.3 million comments posted on videos from 102 far-right channels, the findings show that incivility expressions spread in a contagion-like manner, appearing repeatedly across different channels and gradually clustering around key nodes. In contrast, expressions of intolerance were more confined to ideologically like-minded communities, forming tightly knit groups that consistently displayed high modularity. These results suggest that the diffusion patterns of political discourse on YouTube differ structurally depending on the type of expression. This study highlights the need for more nuanced criteria in AI-driven hate speech detection and platform regulation strategies.

키워드