Police in Charlotte, NC, and Nashville, TN, have turned to the Center for Data Science and Public Policy (DSaPP) at the Computation Institute and Harris School of Public Policy at the University of Chicago (UChicago) to predict and prevent adverse incidents.
The initiative applies machine-learning methods to department data to identify officers and police calls at a higher risk of producing adverse events. The predictive model can be used to guide personalized interventions for at-risk officers or adjust dispatch procedures to reduce high-stress situations.
"The goal is to take historical data about these police officers...and use that data to assign each officer a risk score," says UChicago researcher Rayid Ghani.
In Charlotte, four fellows worked with representatives from the local police department to build an early intervention system that identifies at-risk officers in need of additional training. Using anonymized data on arrests, citizen complaints, disciplinary actions, and other relevant information, the team applied machine-learning methods to find combinations of factors that predict a future adverse incident.
The system found more true positives in which officers flagged by the system were later involved in an adverse event, and fewer false positives.
"We want to produce interpretable results that can help departments create specific interventions and safer operations for their officers," says DSaPP's Joe Walsh.
From UChicago News (IL)
View Full Article
Abstracts Copyright © 2016 Information Inc., Bethesda, Maryland, USA