Temporal Crime Pattern Analysis Using Seasonal Decomposition and k-Means Clustering

Authors

  • Agung Darmawan Buchdadi Faculty of Economics, State University of Jakarta, Indonesia
  • Ammar Salamh Mujali Al-Rawahna Department of Business Administration, Amman Arab University, Jordan

DOI:

https://doi.org/10.63913/jcl.v1i1.4

Keywords:

crime analysis, seasonal decomposition, k-means clustering, geospatial visualization, cybercrime analysis

Abstract

This study explores the temporal and spatial patterns of crime through the application of seasonal decomposition and clustering techniques, providing actionable insights for law enforcement and policymakers. Using a dataset of reported crimes, the research dissects the data into trend, seasonal, and residual components using the Seasonal and Trend decomposition using Loess (STL) methodology. The analysis reveals long-term crime trends, recurring seasonal fluctuations, and anomalies that warrant targeted interventions. Furthermore, k-means clustering is employed to identify high-crime and low-crime periods, offering a granular understanding of crime dynamics across time. Geospatial visualization complements the analysis, illustrating crime clusters within urban hotspots and highlighting areas of concentrated criminal activity. The findings underscore the critical role of temporal and spatial analytics in crime prevention, demonstrating how patterns in reported crimes can guide resource allocation and strategic planning. Moreover, the study bridges the gap between traditional crime analysis and the emerging field of cybercrime prevention. By extending these methodologies to the digital domain, the research highlights their potential application in analyzing cyber threats, such as ransomware campaigns or phishing attacks, which often exhibit temporal regularities and geographic dispersion. This interdisciplinary approach advocates for the integration of data-driven methods into law enforcement strategies and legislative frameworks, emphasizing the importance of adaptive policies in addressing both physical and digital crime landscapes. Future work aims to incorporate advanced predictive models and external data sources to deepen insights into crime causation and prevention. The research contributes to the evolving discourse on smart policing and adaptive cybercrime legislation, paving the way for safer and more resilient societies in both urban and virtual environments.

Downloads

Published

2025-03-15

How to Cite

Buchdadi, A. D., & Al-Rawahna, A. S. M. (2025). Temporal Crime Pattern Analysis Using Seasonal Decomposition and k-Means Clustering . Journal of Cyber Law, 1(1), 65–87. https://doi.org/10.63913/jcl.v1i1.4