Early Intervention System

About the Program

Category: Criminal Justice and Public Safety (Best in Category)

Year: 2017

In 2016, as part of Johnson County’s ongoing Stepping Up and Data-Driven Justice Initiatives, the county partnered with the University of Chicago’s Data Science for Social Good on a project that used machine learning to prioritize outreach to individuals most at risk of being booked into jail within the next year. For the first time, Johnson County brought together jail, emergency medical, and mental health data, identified individuals who touch multiple systems, and built a model to predict individual bookings. Our model significantly outperformed both a random baseline and several simple heuristics that domain experts are likely to use and implement. By focusing on 200 individuals who had interacted with both mental health services and the criminal justice system, and only using data from three disparate systems, we were able to predict jail bookings in the following year with 52% accuracy. This work has provided Johnson County with a framework and prototype system to develop intervention models to proactively connect social and mental health workers to individuals in need of care.