Spatial-Temporal Analysis and Characterization of Crime in the Constitutional Province of Callao, 2024: An Approach Based on Spatial Data Mining (#1004)

Authors

Cutid-Aguero, Jasmin

Alarcon-Ventura, Katicsa

Canales-Escalante, Carlos

Huaman-Yrigoin, Dennis

Zarate-Bocanegra, Jhony-Alex

Solis-Tipian, Martin

Zevallos-Vera, Erika

Abstract

Effective management of public safety requires a deep understanding of crime dynamics. This study presents a spatio-temporal analysis of crime in the Constitutional Province of Callao, Peru, using data from official reports from February 2024. Using a quantitative approach, descriptive analysis and spatial data mining techniques were applied to characterize criminal activity. The methodology included Kernel Density Estimation (KDE) for the visual identification of hotspots and the DBSCAN clustering algorithm for the detection of criminal clusters. The results showed a high concentration of crime, with the districts of Callao Cercado and Ventanilla accounting for almost 80% of incidents. Distinctive crime profiles were identified: the prevalence of property crimes in Callao Cercado and a high incidence of gender-based violence in Ventanilla. Spatial analysis identified statistically significant clusters, highlighting one with 317 incidents in Callao Cercado. It was concluded that crime is not random but follows defined geographical patterns. These findings provide actionable intelligence for the design of targeted patrolling strategies and prevention policies adapted to each territorial reality.