A Study of the Evolution of Haze Microblog Concerns Based on a Co-Word Network Analysis
Haiyue Lu,Xiaoping Rui,Runkui Li,Guangyuan Zhang,Ziqian Zhang,Mingguang Wu
DOI: https://doi.org/10.3390/ijgi13100352
IF: 3.4
2024-10-04
ISPRS International Journal of Geo-Information
Abstract:Haze is a phenomenon caused by excessive PM2.5 (air-borne particulate matter having a diameter of fewer than 2.5 μm) and other pollutants and results from the interaction between specific climatic conditions and human activities. It significantly impacts human health, transportation, and the natural environment and has aroused widespread concern. However, the influence of haze on human mental health, being hidden and indirect, is often overlooked. When haze pollution occurs, people express their feelings and concerns about haze events on media such as Weibo. At present, few studies focus on haze public opinion, as well as the changing trends in people's discussion of haze since its emergence, which is of great significance for haze response and resource management. Based on the perspective of topic analysis, this study explores the psychological impact of haze on people by exploring the feelings of netizens in haze public opinion and investigates the evolution of people's concerns based on long-term public opinion data. In this study, seven typical provinces and cities in China with severe haze pollution were selected as the research area. Based on data on the "haze" theme from Weibo from 2013 to 2019, first, the microblog posts were preprocessed, and the keyword co-word network was constructed. Second, the Louvain algorithm was used to detect the topic community. Based on this, the cosine similarity was calculated to realize the temporal evolution analysis of topics. The results show that with the development and change in haze pollution, the content and intensity of the topics netizens pay attention to have changed, including five types: merger, split, survival, transformation, and rebirth/extinction. People's attention to haze shows obvious spatial differences, and it is related to the degree of haze pollution, which is bipolar. Areas with severe haze tend to pay more attention to haze itself and its influence, while areas with light haze pay more attention to haze control. The research results can provide valuable insights for governments and relevant departments in guiding public opinion and resource allocation.
geography, physical,remote sensing,computer science, information systems