With the development of the internet and the increasing maturity of social media ecosystems, online collective actions within digital social networks have become more frequent and exhibit new patterns. Research on the mechanisms and governance of online collective action has become a crucial aspect of corporate crisis management. This study constructs an online social network with scale-free (BA) characteristics, defines individual properties and interaction rules based on the SIR model, and simulates the outbreak and intervention of online collective actions. The findings indicate that netizens with similar cognitions spontaneously form virtually clustered groups under the influence of triggering events, progressing through emotional arousal, emotional expression, and emotional mobilization, ultimately leading to large-scale online collective action. Collective actions initiated by opinion leaders tend to be more widespread, sudden, and persistent. To mitigate online collective actions, companies should enhance the transparency of core event information, promptly address the demands of action groups, soothe public sentiment, respond to societal concerns, and strengthen guidance over opinion leaders in public discourse.
Objective/Background: Social media has become deeply integrated into urban life, and digital collective actions by young people rooted in physical spaces are becoming increasingly common, posing new challenges to urban governance. There is an urgent need to understand the dynamic evolution of cross-platform public opinion in such events to provide a basis for precise governance. Research process/Methodology: Taking the “Night Riding to Kaifeng” incident as an example, this study integrated 27,216 data points from the Weibo (mass communication) and Zhihu (knowledge community) platforms. Using the life cycle theory to divide public opinion into stages, the study analyzed public emotions at each stage using the emotion dictionary and employed the LDA topic model to explore the evolution of themes. Research conclusion/Results: The study found Weibo exhibited “emotional resonance” with dominant positive emotions, effectively mobilizing offline action, while Zhihu featured diverse emotional profiles with rational debate emphasis. Grounded in collective action theory and urban social movement theory within hybrid space, this research uncovered the organizational logic and cross-platform expression patterns of emergent youth collective action in social media contexts. Research value/Significance: This study deepens understanding of public opinion complexity in collective emergency incidents within social media contexts, offering empirical and theoretical foundations for multi-tier early warning systems, agile collaborative governance, and youth-inclusive resilient urban development.