Artificial intelligence (AI) is changing our lives in many ways. Can it be employed to help stop school violence? That’s the topic of a study published in the International Journal of Medical Informatics by researchers at Cincinnati Children’s Hospital Medical Center. The study developed an AI system to analyze linguistic patterns to predict a student’s risk for violence and aggression at school and online.If the idea sounds like science fiction, consider that linguistic AI is already being used to predict risk for suicide and other mental health issues.
Consider also that the best known way to reduce school violence is timely, direct, individualized intervention by psychiatrists. That is an unwieldy and expensive solution, especially given the heartbreaking statistic that Cincinnati Children’s alone receives about 30,000 psychiatric admissions each year. So an AI intervention could be invaluable and save lives.The three-year study assessed 103 middle- and high-school students between the ages of 10 and 18. All of the students had a history of minor or major behavioral change or aggression toward themselves or others. Of the participants, 55 were grouped as moderate to high-risk and 48 were found to be low risk.
Researchers used recorded interview content to develop the machine-learning algorithm for the system.
The study concluded that AI was successful in identifying 94% of cases identified by physicians. The researchers plan to widen the study to include other hospitals and school systems.
Through this study, the team “will be able to build artificial intelligence to augment human clinical judgment,” said Drew Barzman, a child forensic psychiatrist at CCHMC and lead author of the study.