U.S. flag

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

Using Machine Learning to Identify High Risk Domestic Violence Offenders in NYC, Final Summary Overview

NCJ Number
305443
Author(s)
Date Published
2022
Length
11 pages
Annotation

This report details a project designed to develop and test a novel machine-learning based statistical model to predict the risk of domestic-violence victimization to improve intervention in cases at high risk for violence.

Abstract

The purpose of this project, a collaboration between the University of Chicago Crime Lab and the New York City Police Department, was to develop and test a novel machine-learning based statistical model to predict the risk of domestic-violence victimization to improve intervention in cases at high risk for violence. The field intervention with the NYPD was launched in July 2017. NYPD command maintains a list of high-priority individuals who are thought to be at risk for serious domestic assault. Individuals on this list receive regular home visits from one of the local NYPD’s Domestic Violence Officers (DVOs) to reduce the risk of future victimization. The researchers believe that upon the completion and dissemination of this work, the results will be relevant to researchers and policymakers who are assessing ways to reduce domestic violence.

 

Date Published: January 1, 2022