Government risk communication and response networks: Insights from typhoon news in China

Risk communication plays a critical role in disaster warning and management, and social media is emerging as a powerful tool for information dissemination. While substantial research exists on risk communication, studies focusing on governments as the primary subject are limited, and there is a noticeable gap in examining inter-city and city-organization emergency responses. This study aims to address these gaps by exploring and evaluating China’s governmental communication strategies during typhoon disasters. We collect typhoon-related news from the WeChat Official Accounts of 237 city governments over an entire year, alongside corresponding typhoon data for the same period. Based on this dataset, we propose an analytical framework for government risk communication. First, we design a three-stage analysis method (pre-disaster, mid-disaster, and post-disaster) based on the Disaster Management Cycle to analyze the dynamics of public engagement. Second, we develop city and organization extractors to construct city and organization response networks, offering fresh approaches to studying city collaboration and organizational coordination. Finally, leveraging large language models, we develop a title feature extractor and sentiment classification model, providing valuable insights for optimizing government media strategies. Overall, this study offers novel methods and a fresh perspective on government risk communication.