|
Abstract:
|
Social distancing was associated with a decrease in the transmission COVID-19. By the mobility data visualization, we realized that social distancing varied corresponding with different income levels. Moreover, we find that wealthier areas decreased mobility significantly more than poorer areas. Then, we decided to build up a supervised learning model in order to predict the social distancing by the mobility data. However, the social distancing data are time series data, we need to calculate a social distancing score for each state by weighted normalizing the mobility data based on trends across the USA. By collecting the census data, we consider those following aspects, such as income, health, education, geographical mobility, transportation, insurance and employment. For those factors, we also conduct the spatial analysis in order to identify the autocorrelation. The model could give credence to the effectiveness of social distancing while demonstrating that social distancing is not randomly spread within the country. For those areas with severe COVID-19, we also do the cluster analysis for helping the government on the appropriate decision-making for the inequality issues.
|
