基于多源地理数据的城市突发公共卫生事件 风险评估研究

摘要：

风险评估是突发公共卫生事件应急管理的关键环节,基于地理数据开展风险评估可有效提高精度.通过采集多源地理数据,结合随机森林算法和地理探测器、核密度分析、空间自相关等空间统计分析提出城市突发公共卫生事件风险评估方法,并通过实证分析验证模型可行性.结果表明:采用随机森林算法构建的风险评估模型表现良好;餐饮美食、 公司企业和交通设施等场所是影响疫情的主要因素,疫情流行具有因子交互性,其中餐饮美食与其他因子的交互作用最强;空间传播上位于城市中心区域的社区风险等级较高并呈现向外围逐渐减弱的趋势,同时有明显的高值或低值聚集 .

Risk assessmentis a crucialcomponentin the emergency managementofsudden public health incidents, and geospatialdata-based risk assessmentcan effectively enhanceaccuracy. By collecting multi-sourcegeospatialdataand employing spatialstatisticalanalyses such as the random forestalgorithm , geographic detector, kerneldensity analysis, and spatialauto- correlation, a methodology for risk assessmentofsudden public health incidents in urban areas is proposed. The feasibility of the modelis validated through empirical analysis. The results indicate that the risk assessment model is constructed wellby using the random forest algorithm performs. Places such as dining establishments, corporate enterprises, and transportation facilities are identified as the major factors influencing spatial differentiation of the epidemic.  The epidemic spread exhibits factorinteractions, with dining establishments having the strongestinteraction with otherfactors. In terms of spatial dissemi- nation, communities located in the centralareas ofthe city exhibithigherrisk levels, showing a gradualreduction towards the periphery, accompanied by distinctpatterns ofhigh orlow-value aggregation.

作者：

熊励 ,王思媛

Xiong Li, Wang Siyuan

机构地区：

上海大学管理学院

引用本文：

熊励,王思媛.基于多源地理数据的城市突发公共卫生事件风险评估研究[J].betway官方app 学报(自然科学版),2024,52(5):91-100. 

Xiong Li,Wang Siyuan. Research on risk assessmentof urban public health emergen- cies based on multi-source geographic data[J] . Journal of Henan Normal University(Natural Science Edition) ,  2024,52(5) :91-100. DOI:10. 16366/j. cnki.1000-2367. 2023. 07. 10. 0005.

基金：

国家社科基金重大项目

关键词：

多源数据 ;机器学习 ;空间统计分析 ;公共卫生 ;风险评估

multi-source data; machine learning; spatialdata analysis; public health; risk assessment

分类号：

TP311. 13;R181. 8

基于多源地理数据的城市突发公共卫生事件 风险评估研究.pdf