Integrating Spatial Analytics and Explainable Machine Learning to Analyse Multi-Pollutant Mortality Burdens in Africa

Abstract Air pollution in Africa is a critical public health issue that results in deaths of people, especially in places where urbanisation and vehicle emissions are increasing. To minimize risks, the present study used spatial analysis and a hybrid Genetic Algorithm–Machine Learning (GA–ML) approach to predict mortality rates. The study employs varied datasets, in which Support Vector Regression (SVR), Support Vector Regression-Generic Algorithm(SVR-GA), Light Gradient Boosting Machine (LightGBM), and Light Gradient Boosting Machine- Generic Algorithm (LightGBM-GA) were examined, focusing on Egypt, Nigeria, Kenya and South Africa. In the spatial analysis, Nigeria and Egypt were identified as hotspots of mortality, whereas South Africa had low mortality rates. The SVR-GA algorithm had superior results, producing an R 2 of 86.4%, MAE(0.034) and MSE(0.01), while the SVR had the second-best. SHARP analysis revealed PM2.5 and CO as the dominant factors in building the models, highlighting the need for close monitoring of these pollutants. Moreover, the results were validated by the Taylor diagram and the Empirical Cumulative Distribution. The results demonstrate the effectiveness of integrating optimization with machine learning in this field. The study demonstrates the potential for improved public health awareness of adaptive behaviors in mortality rates in hotspots. The study contributes novel insights to environmental modeling through the integration of spatial statistics, advanced machine learning, metaheuristic optimization, and a thorough evaluation. The hybrid GA-machine learning framework is a powerful and adaptable tool that can assist with data-driven environmental management and achieve Sustainable Development Goals of health, climate action, and sustainable urban development.