Integrating Urban Morphology, Spatial Machine Learning, And Geodemographics: A Structured Analytical Review For Data Scarce Cities In Sub-Saharan Africa

This study examines the integration of urban morphology, spatial machine learning, and geodemographic classification in data-scarce cities of Sub-Saharan Africa (SSA) through a structured analytical review. Using a multi-criteria query applied through Publish or Perish (Google Scholar), an initial corpus of approximately 500 studies was identified, from which 14 studies were retained following multi-stage screening, intersection-based validation, and quality appraisal. A structured coding framework was applied to evaluate the presence and interaction of key analytical components, including morphology-derived data, spatial methods, machine learning techniques, and spatial adaptiveness. The findings reveal a fragmented methodological landscape in which morphology and machine learning are widely used, but rarely integrated with spatially explicit analytical frameworks. Only four studies demonstrate full methodological integration, highlighting a significant gap in the development of spatially coherent geodemographic systems. The study argues that this limitation is not due to a lack of data or analytical tools, but rather the absence of integrative frameworks that combine these components into unified analytical pipelines. The results have important implications for urban planning and environmental management, particularly in supporting evidence-based decisionmaking, infrastructure targeting, and risk assessment in rapidly urbanising SSA cities. The study concludes by emphasising the need for interdisciplinary and spatially adaptive approaches to advance geodemographic classification in data-scarce environments