Journal of Geo-information Science >
Analysis of Time Series Features of COVID-19 in Various Countries based on Pedigree Clustering
Received date: 2020-08-17
Revised date: 2020-11-21
Online published: 2021-04-25
Supported by
Chinese Academy of Sciences strategic leading science and technology project (Class A)(XDA23100000)
National Science and technology basic resources survey project(2017FY101004)
National Natural Science Foundation of China(42041001)
Copyright
Since the outbreak of COVID-19, countries around the world have shown different time-series characteristics. Studying the characteristics of the development patterns of different countries and revealing the dominant factors behind them can provide references for future prevention and control strategies. In order to reveal the similarities and differences between the epidemic time series in different countries, this article extracts the standard deviation, Hurst index, cure rate, growth time, average growth rate, and prevention and control efficiency of the daily time series of new cases in the main epidemic countries for pedigree clustering. We also analyzes the causes of clustering results from the aspects of economics, medical treatment, and humanistic conflicts. The results show that the global epidemic development model can be divided into three categories: C-type, S-type, and I-type. The time series of C-type countries are characterized by continuous fluctuations and rising, and the cure rate is low. The reason is that humanistic conflicts are not conducive to epidemic prevention and control. Economic and medical resources have become scarce after a long period of large consumption. It is recommended to strengthen publicity and guidance in prevention and control, change concepts, and coordinate the allocation of economic and medical resources. The time series of S-type countries is characterized by a rapid rise and then an immediate decline, and eventually maintains a stable trend. The overall cure rate is relatively high. The reason is that these countries have domestic stability, high economic and medical standards, and timely prevention and control measures. It is recommended to strengthen international cooperation and scientific research, and prepare for the possible second epidemic. The time series of I-shaped countries is characterized by a slow rise, the overall development trend is unstable, and the cure rate is low. The reason is that its outbreak is relatively late and less severe. Most of the economic and medical levels and humanistic conflicts are not conducive to epidemic prevention and control. It is recommended to learn better prevention and control experience, implement strict isolation measures, try to meet the material needs during the epidemic, and optimize treatment methods.
XIE Conghui , WU Shixin , ZHANG Chen , SUN Wentao , HE Haifang , PEI Tao , LUO Geping . Analysis of Time Series Features of COVID-19 in Various Countries based on Pedigree Clustering[J]. Journal of Geo-information Science, 2021 , 23(2) : 236 -245 . DOI: 10.12082/dqxxkx.2021.200470
表1 COVID-19综合防控严峻指数构成Tab.1 Composition of COVID-19 comprehensive prevention and control severity index |
评价指标 | 评价内容 | 意义 | 来源 |
---|---|---|---|
经济因子 | 经济实力、增长、发展 | 经济基础决定了国家是否有足够的经济实力支撑抗疫进度,但经济发达也意味着人群活动性强,从而加大疫情传播,影响时间序列发展趋势 | 兰德公司传染病脆弱性指数中的经济指标[26] |
交通运输、技术、通讯等基础设施 | |||
医疗因子 | 个人医疗服务的获取和质量 | 国家的医疗体系是否完善决定了治愈率和死亡率的大小 | 《柳叶刀》发布的2019全球医疗质量和可及性榜单[27] |
人文冲突因子 | 暴力内部冲突概率 | 国家的人文冲突程度影响着疫情的传播,国内冲突较多会增强疫情的传播,国民配合程度也会较低,从而降低防控效率,新增确诊病例时间序列也会难以下降 | 欧盟委员会(JRC)联合研究中心开发的多危害风险评估信息全球风险指数增强版(GRI)中的人文指标[28] |
高暴力内部冲突概率 | |||
国家权力冲突强度 | |||
国家以下各级冲突强度 |
图2 各国疫情时间序列特征谱系聚类结果注:纵坐标的数字代表国家序号。 Fig. 2 The pedigree clustering results of the time series characteristics of epidemics in various countries |
[1] |
|
[2] |
|
[3] |
|
[4] |
|
[5] |
|
[6] |
|
[7] |
|
[8] |
|
[9] |
|
[10] |
|
[11] |
|
[12] |
|
[13] |
|
[14] |
|
[15] |
|
[16] |
|
[17] |
|
[18] |
|
[19] |
|
[20] |
|
[21] |
|
[22] |
|
[23] |
|
[24] |
宋辞, 裴韬. 基于特征的时间序列聚类方法研究进展[J]. 地理科学进展, 2012,31(10):1307-1317.
[
|
[25] |
|
[26] |
Moore, Melinda,
|
[27] |
|
[28] |
|
[29] |
袁勇. 非洲疫情形势值得密切关注[N]. 经济日报, 2020-04-11(002).
[
|
[30] |
刁海洋. 全球疫情复杂演化关键问题依然待解[J]. 健康中国观察, 2020(4):90-91.
[
|
[31] |
|
[32] |
|
/
〈 | 〉 |