Department of Engineering, Federal Rural University of the Semi-Arid, Angicos, RN.
Department of Production Engineering, Federal University of Pernambuco, Recife, PE.
Geospat Health. 2022 Jan 14;17(s1). doi: 10.4081/gh.2022.1000.
The paper presents an innovative application to identify areas vulnerable to coronavirus disease 2019 (COVID-19) considering a combination of spatial analysis and a multi-criteria learning approach. We applied this methodology in the state of Pernambuco, Brazil identifying vulnerable areas by considering a set of determinants and risk factors for COVID-19, including demographic, economic and spatial characteristics and the number of human COVID-19 infections. Examining possible patterns over a set number of days taking the number of cases recorded, we arrived at a set of compatible decision rules to explain the relation between risk factors and COVID-19 cases. The results reveal why certain municipalities are critically vulnerable to COVID-19 highlighting locations for which knowledge can be gained about environmental factors.
本文提出了一种创新的应用方法，通过空间分析和多准则学习方法的结合，来识别易感染 2019 年冠状病毒病（COVID-19）的区域。我们在巴西伯南布哥州应用了这种方法，通过考虑一组 COVID-19 的决定因素和风险因素，包括人口、经济和空间特征以及人类 COVID-19 感染的数量，来确定易感染区域。通过检查一定天数内记录的病例数量，我们得出了一组兼容的决策规则，以解释风险因素和 COVID-19 病例之间的关系。结果揭示了为什么某些城市易受 COVID-19 的严重影响，突出了需要了解环境因素的位置。
