Spatio-temporal crime pattern analysis in Addis Ketema, Addis Ababa, Ethiopia: GIS and R based approaches

Abstract

Spatio-temporal crime analysis is a critical component of modern law enforcement and urban planning, aiming to understand the dynamic nature of criminal activities within a geographic context. This study presents a comprehensive approach to spatio-temporal crime analysis using R programming language. The study begins by collecting and preprocessing 5-year crime data, incorporating both spatial and temporal dimensions. It applies various exploratory data analyses to unveil underlying patterns and trends in the data, including hotspot identification, density estimation, and cluster detection to pinpoint areas of high criminal activity and potential crime clusters. The results reveal that fraud, accounting for 19.5 % of reported crimes, is the most prevalent crime, followed by attempted murder (11.1 %) and drug offenses (9 %). Temporal trends indicate that fraud peaked in 2018, while attempted murders showed the highest frequency in 2021. Crime incidents were most frequent in the summer months, with notable spikes in June and July for attempted murder, and August through October for fraud. The spatial analysis identified crime hotspots in areas such as Merkato and Autobis Tera, where commercial activity and transportation hubs correlate with high crime concentrations. The study recommends continuously collecting and analyzing crime data to identify emerging trends and adapt strategies accordingly.