U.S. flag

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

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

Predicting Police Misconduct With a Neural Network Program

NCJ Number
155750
Journal
Law Enforcement Technology Volume: 22 Issue: 6 Dated: (June 1995) Pages: 56-58,68
Author(s)
T W Burke
Date Published
1995
Length
4 pages
Annotation
Using an artificial intelligence neural network program called Brainmaker, the Chicago Police Department has developed an automated early warning system designed to identify police officers at risk of misconduct.
Abstract
Brainmaker is based on a software program that uses neural networking to analyze data. The Chicago Police Department used this program to collect and store data electronically in order to identify police officers at risk of misconduct and minimize future disciplinary actions. Using Brainmaker software, the Chicago Police Department designed an automated early warning system in an effort to profile potential police misconduct. The system analyzes data on all sworn police officers, including former police officers who have been fired or dismissed, and compares data between those who have been discharged and those who have not. The system actually teaches itself to recognize patterns in the discharged group and generalizes those patterns to incumbent personnel. Thus, the system identifies police officers who are likely to engage in future misconduct. Criticisms of and legal concerns associated with the Brainmaker system are discussed, and recommendations are offered for police agencies who want to use artificial intelligence techniques to identify police misconduct.