Mapping Criminal Activities Using Biclustering Methods based on Nationality Information of Offenders

Crimes such as terrorism, organized crime, cybercrime, smuggling, money laundering, and trafficking represent critical challenges to global security, often involving complex transnational networks. Traditional clustering techniques used to analyze offender data typically focus on either individuals or crime types in isolation. In this study, we propose a biclustering-based approach to simultaneously group offenders and crime categories, enabling the identification of hidden patterns and associations. Using the Bimax algorithm implemented in R, we analyzed a dataset consisting of offenders involved in similar criminal activities. The method facilitates the creation of crime maps that reveal clusters sharing common attributes. Results demonstrate that offenders often cluster according to nationality, leading to the discovery of distinct crime routes linked to specific national groups. These findings highlight the value of biclustering in uncovering latent structures in complex criminal networks and offer a novel methodological perspective for crime analysis. The approach provides actionable insights for law enforcement agencies by supporting targeted interventions based on shared behavioral patterns and demographic indicators. This work not only showcases the effectiveness of biclustering in criminological data analysis but also lays a foundation for future research integrating larger and more diverse datasets or exploring alternative algorithms to enhance predictive accuracy in applied security analytics.