A Study of Public Opinion Reversal Recognition of Emergency Based on Hypernetwork.

With the rapid development of social media and online platforms, the speed and influence of emergency dissemination in cyberspace have significantly increased. The swift changes in public opinion, especially the phenomenon of opinion reversals, exert profound impacts on social stability and government credibility. The hypernetwork structure, characterized by its multilayered and multidimensional complexity, offers a new theoretical framework for analyzing multiagents and their interactions in the evolution of public opinion. Based on hypernetwork theory, this study constructs a four-layer subnet model encompassing user interaction network, event evolution network, semantic association network, and emotional conduction network. By extracting network structural features and conducting cross-layer linkage analysis, an identification system for public opinion reversals in emergencies is established. Taking the donation incident involving Hongxing Erke during the Henan rainstorm in 2021 as a case study, an empirical analysis of the public opinion reversal process is conducted. The research results indicate that the proposed hypernetwork model can effectively identify key nodes in public opinion reversals. The multi-indicator collaborative identification system for public opinion reversals aids in rapidly and effectively detecting signals of such reversals. This study not only provides new methodological support for the dynamic identification of public opinion reversals but also offers theoretical references and practical guidance for public opinion monitoring and emergency response decision-making in emergencies.