FaradayOtieno/Health_Care_Kenya

Identify patterns in healthcare facility distribution and ownership type in Kenya # Geospatial Clustering of Health Facilities in Kenya Based on Location and Ownership Type This project uses unsupervised machine learning to analyze health facilities in Kenya based on their **ownership** and ** regional patterns in healthcare facility distribution**. The goal is to uncover patterns that can inform health planning, investment, or policy. ## 🗂️ Dataset - Health Facility Master List (Kenya) - Includes variables like: - Ownership (public, private, NGO, religious) - Facility type and services - Region (optional) ## 🧠 Techniques Used - Label Encoding for categorical fields - KMeans clustering (4 clusters) - PCA for dimensionality reduction and visualization - Heatmaps and group-wise summaries for interpretation ## 📊 Key Insights - Cluster 0: Predominantly government-owned facilities - Cluster 1: Dominated by private practice (clinics, specialists) - Cluster 2: Mixed community/NGO ownership - Cluster 3: Specialized private practices ## 🖼️ Visuals - Ownership distribution by cluster - PCA scatter plot of clusters - Heatmaps showing ownership dominance ## 🛠️ Tools - Python (pandas, scikit-learn, matplotlib, seaborn) ## 📌 Conclusion This project highlights how clustering can expose hidden structure in healthcare infrastructure. The results may guide resource allocation, policy targeting, or further analysis in health systems research. ## 📄 License MIT License