CRIME CLUSTERING AND SPATIOTEMPORAL ANALYSIS: A NARRATIVE LITERATURE REVIEW ON METHODS, PATTERNS, AND URBAN SECURITY APPLICATIONS

The study of criminality from a spatiotemporal perspective has gained prominence due to the increasing complexity and variability of crime patterns in urban environments. This narrative literature review aimed to identify the main techniques used for crime clustering, with a focus on spatial, temporal, and algorithmic approaches. To this end, searches were conducted in the Scielo and Scopus databases using specific keyword combinations related to crime clustering, spatial and temporal analysis, and clustering algorithms. The findings reveal a growing use of analytical tools such as complex networks, machine learning algorithms, clustering methods, geographic information systems (GIS), and statistical models. These approaches have significantly contributed to a deeper understanding of criminal dynamics in urban areas. The reviewed studies indicate that factors such as the built environment, sociodemographic variables, and seasonal characteristics directly influence the distribution and frequency of crimes. The literature reinforces that integrating these diverse techniques not only enhances the ability to map and predict crime occurrences but also provides concrete support for the development of more effective preventive public policies. When data-driven and tailored to the specificities of each locality, such policies hold greater potential to generate meaningful responses to the challenges of urban security.