Research on the influence of Citizen Sentiment on government response based on natural language processing

With the rise of digital governance, network politics has become an important way for the Chinese government to interact with citizens. What kind of sentiments do citizens express in their messages on environmental issues on network political platforms? What are the preferences and strategies of government responses? What are the differences in government responses between different cities? We selected text data from four Chinese municipalities (Beijing, Shanghai, Tianjin, and Chongqing) from 2023 to 2024 from the People's Daily Online Leaders' Message Board website. We analyzed the relationship between citizens' sentiments on the messages and the government's response strategies using Natural Language Processing (NLP) techniques. It was found that citizens' message emotions fluctuated greatly, while government responses showed stable and positive emotions, displaying standardized response strategies. In addition, citizens' demands and government responses do not match perfectly on certain key environmental topics, and there are differences in response attitudes and timing between different city governments. Based on these findings, this study proposes recommendations to strengthen public participation, optimize government response, and improve environmental infrastructure, with a view to providing new paths to optimize government-citizen interaction and targeted policy recommendations to promote the modernization of the environmental governance system and governance.