U.S. flag

An official website of the United States government, Department of Justice.

Using Random Forest Risk Prediction in the Philadelphia Probation Department

Geoff Barnes, Ph.D.; Jordan Hyatt, Ph.D.

Watch two experts talk about developing a computerized system that successfully predicts — with a high degree of accuracy — which probationers are likely to violently reoffend within two years of returning to the community.

Drs. Barnes and Hyatt teamed up with the Philadelphia Adult Probation & Parole Department in an NIJ-funded project. Here they discuss:

  • Why was the Probation Department interested in exploring a new risk prediction tool?
  • How does the random forest prediction tool work?
  • Why did the researchers have to work hand-in-hand with the practitioners?
  • Why is random forest modeling such an effective tool?

Date Published: August 1, 2012