Early Intervention System for Adverse Police Interactions

Fellows: Samuel Carton, Kenneth Joseph, Ayesha Mahmud, Youngsoo Park
Data Science Mentor(s): Joe Walsh
Project Partner: The White House and Charlotte-Mecklenburg Police Department
[Github Repository]

This project is being continued at the Center for Data Science & Public Policy at the University of Chicago. For recent updates, results, and news, please visit the project web page.

Many police departments in the United States have developed and deployed “early warning systems” to identify officers who may benefit from additional training, resources, or counseling. These systems attempt to determine behavioral patterns that predict a higher risk of future adverse incidents, such as excessive use of force or citizen complaints. Detecting these actions opens new opportunities to develop targeted interventions for officers to protect their safety and improve police/community interactions.

As part of the 2015 White House Police Data Initiative, DSSG partnered with multiple police departments, including the Charlotte-Mecklenburg Police Department, to apply data analysis to identify which factors should be used in early warning systems to flag at-risk officers before an adverse interaction occurs. Using anonymized police data, as well as contextual data about local crime and demographics, this model detects the factors most indicative of future problems, so that departments can provide additional support to their officers. In 2016, we partnered with additional police departments, including the Metro Nashville Police Department, to test and expand this work in new municipalities, improving both the overall model and local performance. Like our work in 2015, we used anonymized police data and contextual data about local crime and demographics to detect the factors most indicative of future issues, so that departments can provide additional support to their officers.

Both departments continue to work with the Center for Data Science and Public Policy on implementing the new EIS. The Nashville team gave their department a list of the highest-risk officers according to our model, which MNPD subsequently used to send letters to the officers and their supervisors informing them of the results and specific risk factors that led to their score. We’re now helping them integrate the EIS into their existing IT system, so that it will continuously update with new data.

Similarly, CMPD awarded us a contract to help implement our EIS on their system. We’re building a web interface to help them and other partners evaluate and understand the performance of the models. The interface will also allow for feedback from supervisors in the department on the quality of the predictions, providing valuable new data to further refine the model. CMPD hopes to bring the system live in the coming months. In addition, the Pittsburgh Bureau of Police will be involved in the expansion of this EIS in early 2017.

You can find updated information about this project here.