突发公共卫生事件居民响应时空关联特征分析  

Analysis of Spatiotemporal Correlation Characteristics of Residents′ Response to Public Health Emergencies

作　　者：王艳东[1] 赖林玉 豆明宣 王东阳 吴晨琛[2] 林尚纬 WANG Yandong;LAI Linyu;DOU Mingxuan;WANG Dongyang;WU Chenchen;LIN Shangwei(State Key Laboratory of Information Engineering in Surveying,Mapping and Remote Sensing,Wuhan University,Wuhan 430079,China;National Geomatics Center of China,Beijing 100830,China)

机构地区：[1]武汉大学测绘遥感信息工程国家重点实验室,湖北武汉430079 [2]国家基础地理信息中心,北京100830

出　　处：《测绘与空间地理信息》2024年第11期8-12,共5页Geomatics & Spatial Information Technology

基　　金：国家自然科学基金(42371472);中央高校自主科研项目(2042022dx0001)资助。

摘　　要：客观认识城市居民在突发公共卫生事件不同防控阶段的响应模式,对提升应急管理能力、促进居民心理健康恢复至关重要。本文基于武汉市社交媒体、政策与病例数据,综合运用自然语言处理技术与空间分析等方法,在社区尺度上探究突发公共卫生事件不同防控阶段下居民响应的时空特征。结果表明:(1)居民对突发公共卫生事件的响应趋向于负面,事件相关的微博关注度随时间的变化与事件发展趋势同步;居民在封控时期情感状况较为平稳,防控政策放开后短期内负面情绪增加,而后逐渐恢复。(2)不同防控阶段下的居民响应与事件严重程度均具备时空关联特性,关注度与病例指数均呈正相关,而情感响应则呈负相关。典型区域如医院周边情感较为乐观,而产业园区则以负面情绪为主。An objective understanding of the public response patterns of urban residents to public health emergencies at different stages of prevention and control is crucial for improving emergency management capabilities and promoting the recovery of mental health.Based on the social media data,policy and case data in Wuhan,we comprehensively use NLP technology and spatial analysis to explore the spatiotemporal characteristics of the public response at different stages of prevention and control of public health emergency at the community scale.The results show that:(1) Residents' sentiment response to public health emergency tends to be negative,and the change of related microblog attention related over time is synchronized with the development trend of the emergency;the sentiment status during the period of lockdown is relatively stable,and the negative emotions after the release of lockdown increase in the short term,and then gradually recovered.(2) Residents' response and the severity of public health emergency at different stages of prevention and control had spatiotemporal correlation characteristics,and the attention and the case index were positively correlated,while the sentiment response was negatively correlated.Typical areas such as hospitals and surroundings are more optimistic while industrial parks are dominated by negative emotions.

关 键 词：突发公共卫生事件 微博数据 情感响应 时空特征 双变量空间自相关 

分 类 号：P208[天文地球—地图制图学与地理信息工程]