This study seeks to mitigate the analytical limitations of traditional Geographic Information Systems (GIS) in urban governance by proposing an integrated, cloud-based Spatial Data Mining Decision Support System (SDM-DSS) framework. Through a systematic literature review methodology, recent peer-reviewed studies (2021–2026) from major scientific databases were extracted, comparatively analyzed, and thematically synthesized to identify architectural vulnerabilities within current smart city models. The findings indicate that although advanced SDM algorithms exhibit high theoretical accuracy in modeling urban phenomena, their practical deployment is frequently impeded by fragmented architectures, localized computational bottlenecks, and the absence of real-time Internet of Things (IoT) integration. To rectify these operational deficiencies, this research formulates a three-layered conceptual architecture comprising a Data Management Layer, a Spatial Analytics Engine, and a Presentation Dashboard. By offloading computationally intensive workloads to a scalable cloud environment, the proposed framework seamlessly translates complex algorithmic outputs into intuitive, actionable policy directives, as demonstrated through scenarios involving dynamic public facility allocation and predictive disaster mitigation. Ultimately, the integrated SDM-DSS architecture fundamentally transitions urban planning from a reactive approach to a proactive, predictive paradigm. Future research should prioritize the empirical prototyping of this framework utilizing real-time municipal data streams, alongside the incorporation of privacy-preserving machine learning techniques to safeguard data sovereignty.
Copyrights © 2025