By Rajan Tavathia and Kimberly Mann Bruch, SDSC Communications
The UC Berkeley Data Science Discovery Program (Discovery) is a leading data science and AI research hub that collaborates with students, faculty, government, non-profit, and industry partners. One of its current projects is to partner with the Department of Justice’s (DOJ) Federal Bureau of Prisons (FBOP) to examine criminal history data beyond re-arrest as a criterion for recidivism analysis.
This project, titled "Recidivism Coding Project: Investigating Potential Changes to Federal Recidivism Tracking", aims to address the need for better recidivism measures that was highlighted in the Attorney General's Annual FSA Report in April 2023. The report noted that stakeholder feedback suggested that re-arrest alone was not a sufficient indicator of recidivism. The project will also assess and improve the FBOP's recidivism tool called PATTERN (PATTERN - Prisoner Assessment Tool Targeting Estimated Risk and Needs).
The Data Science Discovery Program was established in 2015 with the vision of fostering and advancing data science and AI research to address real world challenges. Since then, it has hosted 1,000 research projects involving academic, government, non-profit, and industry researchers from various disciplines.