In response to the escalating incidence of burglary incidents in rapidly urbanizing metropolitan regions, this study innovatively integrates Emerging Hot Spot Analysis (EHSA) with Space-Time Pattern Mining (STPM) to examine the spatio-temporal dynamics of burglary across Selangor, Kuala Lumpur Federal Territory (KLFT) and Putrajaya Federal Territory (PFT) between 2015 and 2020. This paper aims to delineate the intricate interplay between urban land use configurations and the evolving patterns of burglary, thereby addressing critical research gaps in crime mapping and predictive resource allocation. The research employed robust methodological framework within the ArcGIS Pro 3.1 environment, the research stratifies crime data into four distinct temporal intervals to construct space-time netCDF cubes, applies the Getis-Ord Gi* statistic with False Discovery Rate (FDR) correction to identify statistically significant clusters, and utilizes the Mann-Kendall trend test to classify hotspots into eight categories (new, consecutive, intensifying, persistent, diminishing, sporadic, oscillating, and historical). The results reveal a nuanced spatial clustering of burglary incidents that is significantly influenced by varied land use types—ranging from residential and industrial zones to open spaces—thereby enhancing the granularity of hotspot detection and offering empirical insights into the temporal evolution of crime patterns. The study dinds that the integration of advanced geospatial analyses not only clarifies the complex dynamics between urban morphology and burglary occurrences but also provides a solid empirical basis for informed law enforcement and urban planning strategies. Moreover, these findings underscore the need for ongoing longitudinal investigations and the development of adaptive, data-driven models to refine predictive capabilities further and foster sustainable urban safety initiatives.
Ahmad et al. (Thu,) studied this question.