Density-Based Spatial Clustering to uncover hidden irregularities in Nairobi’s urban air pollution

Abstract

Urban air pollution poses significant health and environmental challenges, particularly in rapidly urbanizing cities like Nairobi. Traditional air quality monitoring methods often rely on costly infrastructure and lack the granularity required to detect localized anomalies. This study used Density-Based Spatial Clustering algorithms, such as DBSCAN and HDBSCAN, to uncover hidden irregularities in Nairobi’s urban air pollution. By integrating spatial and temporal data from low-cost sensors, the system identifies clusters of pollution anomalies and provides actionable insights for policymakers and urban planners. The framework combines real-time data processing, robust anomaly detection, and user-friendly visualization tools to bridge gaps in existing monitoring systems. The research also addresses challenges such as sparse sensor coverage and the dynamic nature of pollution sources in resource-constrained settings. By leveraging advanced clustering techniques and spatial analysis, this study aims to enhance air quality management and contribute to sustainable urban development in Nairobi.