Data mining and socio-spatial patterns of COVID-19: geo-prevention keys for tackling the pandemic

Abstract

A geographic perspective is essential in tackling COVID-19. This research study is framed in the collaboration project set up by the University of Cantabria, the Valdecilla Hospital Research Institute (IDIVAL) and the Regional Government of Cantabria. The case study is the Santander functional urban area (FUA), which is considered from a multi-scale perspective. The main source is the daily records of micro-data on COVID-19 cases and the methodology is based on ESRI geo-technologies, and more specifically on a tool called SITAR (a Spanish acronym which stands for Fast-Action Territorial Information System). The main goal is to analyse and contribute to knowledge of the spatial patterns of COVID-19 at neighbourhood level from a space-time perspective. To that end the research is based on data mining methods (3D bins and emerging hot-spots) and exploratory geo-statistical analysis (Global Moran’s Index, Nearest Neighbourhood and Ordinary Least Square analyses, among others). The study identifies space-time patterns that show significant hot-spots and demonstrates a high presence of the virus at building level in neighbourhoods where residential and economic uses are mixed. Knowing the spatial behaviour of the virus is strategically important for proposing geo-prevention keys, reducing spread and balancing trade-offs between potential health gains and economic burdens resulting from interventions to deal with the pandemic.

Keywords

• emerging hot-spots
• geo-technologies
• micro-data
• social space
• multi-scale