codermiki/clustering-african-countries-based-on-COVID-19-spread-patterns

# Clustering African Countries Based on COVID-19 Spread Patterns ## 1. Project Overview This project implements **Unsupervised Machine Learning** techniques to cluster African countries based on their COVID-19 spread patterns. The primary objective is to group countries with similar pandemic trajectories to facilitate fair comparison and provide actionable insights for regional public health planning. The core metric used for clustering is the **14-day cumulative COVID-19 cases per 100,000 population**. This normalized metric allows for an equitable comparison between countries with vastly different population sizes. ## 2. Objectives - **Identify Patterns**: Detect distinct patterns of COVID-19 transmission across the African continent. - **Comparative Analysis**: Group countries into clusters to enable comparative analysis of mitigation strategies and outcomes. - **Public Health Insights**: Provide data-driven insights to support policy formulation and resource allocation. ## 3. Dataset The project utilizes data reported by the **European Centre for Disease Prevention and Control (ECDC)**. - **Content**: Daily records of COVID-19 cases, deaths, and population data for African countries. - **Preprocessing**: The raw data is aggregated and normalized to calculate the 14-day cumulative cases per 100,000 population for each country. ## 4. Methodology ### 4.1. Preprocessing Data cleaning and feature scaling were performed using `StandardScaler` to ensure that all features contribute equally to the distance computations in clustering algorithms. ### 4.2. Algorithms Three unsupervised machine learning algorithms were implemented and compared: 1. **K-Means Clustering**: Partitioning n observations into k clusters in which each observation belongs to the cluster with the nearest mean. 2. **Agglomerative Hierarchical Clustering**: Building a hierarchy of clusters using a bottom-up approach. 3. **BIRCH (Balanced Iterative Reducing and Clustering using Hi …