Police departments, and predictive policing systems, have historically focused their efforts on reducing “street crimes”. However, the development of novel machine learning techniques presents law enforcement with an opportunity to expand their policing efforts into a new domain: high level financial crime.
We propose and develop a predictive policing algorithm, the White Collar Crime Early Warning System (WCCEWS), for identifying and assessing the risk of large-scale financial crime at the city block level. Our model achieves an impressive 90.12% accuracy at predicting the activity of white collar crime in a given area.
WCCEWS predicts the likelihood of a white collar crime occurring within a 76m2 square, which is a 197.37% improvement of precision when compared with other predictive policing algorithms. The model is augmented to predict the nature of the white collar crime, as well the severity of the crime (in terms
of expected fines).
In developing our model, we focused on the physical features of the under-lying landscape as described in the risk terrain modeling approach (explained further below), while deprioritizing temporal features. As such, we assert with a high degree of confidence that white collar crimes are occurring continuously in the predicted high-risk zones.
Our model is optimized for growth in the policing of high level financial criminals, with growth measured in higher arrest rates, improved quality of arrests, and higher recovery of funds. The system is designed to be further integrated into tools for citizen policing and awareness, such as the White Collar Crime Risk Zones iOS app, which alerts users when they enter high-risk areas for financial crime.
queerlybeloveds 