Spatial Analysis of Covid-19 Infection Patterns Using Unsupervised Classification with k-Means Clustering

Oluwafemi Bisiriyu, Obafemi Awolowo University, Ile-Ife, Nigeria
Temisola OYELAKIN, Early Childhood Development Project

COVID-19 pandemic has spread devastatingly throughout the world, causing immense loss of life and economic hardships. This study aims to analyze the characteristics and patterns within each cluster to understand variations in the spread of the disease and explore the spatial pattern across continent using spatial analysis. In this study, we utilized data sourced from online repositories, including a store dataset obtained from Kaggle for illustrative purposes. The data were analyzed using Exploratory Data Analysis (EDA) to delve deeper into the variables, identify patterns, and examine relationships among them. Clustering analysis and hotspot analysis provides valuable insights into distinct patterns and trends in the data, aiding in the understanding of the spread, severity, and other dynamics of the virus. Cluster 2 represents countries with substantial numbers of confirmed cases, deaths, and active cases, indicative of a severe impact from the Pandemic.

Keywords: Geographic Information Systems (GIS), Population projections, forecasts, and estimations, Health and Morbidity, Spatial Demography

See paper.