Determining the probability of juvenile delinquency by using support vector machines and designing a clinical decision support system.

It is a known fact that individuals who engaged in delinquent behavior in childhood are more probable to carry on similar behavior in adulthood. If the factors that lead children to involve in delinquency are defined, the risk of dragging children into crime can be detected before they are involved in crime and delinquency can be prevented with appropriate preventive rehabilitation programs, in the early period. However, given that delinquent behavior occurs under the influence of multiple conditions and factors rather than a single risk factor; the need for diagnostic tools to evaluate multiple factors together is obvious. Artificial intelligence-based clinical decision support systems have already been used in the field of psychiatry as well as many other fields of medicine. In this study, we assume that thanks to artificial intelligence-based clinical decision support systems, children and adolescents at risk can be detected before the criminal behavior occurs by addressing certain factors. In this way, we anticipate that it can provide psychiatrists and other experts in the field.