Spatio-temporal Analysis Method Based On The Characteristics Of Human Migration Network

The researches of infectious disease transmission are the focus of modern medical research.Infectious disease dynamics model and spatio-temporal characteristic analysis of infectious diseases are the main research methods at present.However,the traditional infectious dynamics model method has the problem that the parameters are difficult to obtain dynamically,while the spatiotemporal characteristic analysis method does not take into account the influence of human mobility factor.In this paper,a research method of spatiotemporal feature analysis based on the characteristics of human migration network is proposed.Firstly,constructing a human migration network and calculating the network Page Rank value based on the flow are studied.Secondly,time series comparative analysis(correlation coefficient method,timing diagram and thematic map),spatial autocorrelation analysis and hot spot analysis between the Page Rank value and infectious disease data are performed.Finally,an empirical analysis is conducted on the migration data and the new coronavirus(COVID-19:Corona Virus Disease 2019)data of 17 urban areas in China Hubei Province during the Spring Festival in 2020.The results prove that based on the method proposed in this paper,the correlation between the epidemic transmission law and human migration and movement pattern in 17 urban areas during the Spring Festival transportation in Hubei Province in 2020 can be found.In particular,the correlation between the Page Rank value and the infectious disease numbe include:(1)time series comparative analysis: 1)Page Rank value and epidemic data have a small positive correlation,that is,the intensity of human activities has a positive correlation with the number of epidemic situations.2)The Impact of Page Rank values on outbreaks had a delaying effect,with higher Page Rank values indicating the frequency of human activities and the increase in the number of outbreaks caused by them not manifesting until a few days apart.3)Page Rank value and epidemic data have the effect of alternating cycles at high and low intervals,indicating the alternating effects of human activities and epidemics;(2)Spatial autocorrelation analysis: the spatial form of Page Rank value and epidemic data mostly presents discrete state,and the Moran index of the two shows a positive correlation,that is,human activity space and epidemic distribution space have consistent forms;Perform hotspot analysis on epidemic data and Page Rank value that show the date of cluster status:Page Rank value is quite different from the distribution status of cold and hot spots in epidemic data,and there is a certain opposite relationship.When the epidemic presents is hot,the human Page Rank value at the corresponding time period is cold;(3)Medical data analysis: Basic medical facilities and epidemic data have a negative correlation,indicating that the spread of the epidemic is better controlled in areas with high medical level,while the control of epidemic transmission is slower in areas with relatively weak medical level.