B.C. Firm Puts Odds On Who Will Stay Clean
Triant Technologies has developed a software program that predicts the odds of whether criminals released from prison will re-offend after they re-enter society. The Corrections Risk Analysis System (C-RAS) uses advanced pattern-recognition technology to compare an offender with a database of already profiled criminals who have been released.
An offender’s risk to re-offend and capacity for violence is modeled on 80 factors, including alcohol use, past crimes, family/marital relations, and employability.
The program is also used to determine which areas should be targeted for treatment.
According to Kevin Lorette of Triant Psychometrics, a unit of Triant Technologies, the system is based on four years’ worth of data gathered from Wisconsin’s prison system, but the results are expected to be universal across North America.
However, Neil Boyd, a criminologist at Simon Fraser University, is skeptical of the program’s universality because penalties for crimes vary from state to state and between the United States and Canada.
At the same time, a study conducted last year in Wisconsin revealed that the C-RAS had an 85 percent to 90 percent accuracy rate of identifying which offenders would be re-arrested within a release window of two years.
In addition, Lorette says the C-RAS is not meant to be the only tool used by officials to decide who is released and when. Triant Technologies plans to design applications of the C-RAS for homeland security and the industries of finance, insurance, and law enforcement.
Abstracted by the National Law Enforcement and Corrections Technology Center(NLECTC) from the Vancouver Sun (08/25/03) P. D4; Alexander, Doug.
