Research Horizons


Pilot Study Validates AI to Help Predict School Violence

A pilot study indicates that artificial intelligence may be useful in predicting which students are at higher risk of perpetrating school violence.

The researchers found that machine learning—the science of getting computers to learn over time without human intervention—is as accurate as a team of child and adolescent psychiatrists, including a forensic psychiatrist, in determining this risk. The study is published online May 1, 2018, in Psychiatric Quarterly.

“Previous violent behavior, impulsivity, school problems and negative attitudes were correlated with risk to others,” says lead author Drew Barzman, MD, Director of the Child and Adolescent Forensic Psychiatry Service.

The team evaluated 103 at-risk teens in 74 traditional schools throughout the U.S. Their machine learning algorithm achieved 91.02 percent accuracy, considered excellent, when using interview content to predict risks. The next step: gathering outcome data to assess whether machine learning could actually help prevent school violence.

Publication Information
Original title: Automated Risk Assessment for School Violence: a Pilot Study
Published in: Psychiatric Quarterly
Publish date: May 1, 2018
Read the study

Research By

Drew Barzman, MD
Director, Child and Adolescent Forensic Psychiatry Service
Barzman's research focuses on child and adolescent forensic psychiatry; prevention of violence in youth; neuroimaging in children and adolescents; genetics; pediatric clinical trials