Predictive modeling, gatekeeper training, and decision trees
We are currently working on 3 suicide-related projects, each with a unique population–active duty military members, college students, and homeless youth. Techniques (e.g., predictive modeling, coverage algorithms for gatekeeper training) from the homeless youth focused project are the foundation for the work with college students and active duty military members.
USC CAIS is using AI and in-person social networks to predict suicidal thinking among active duty military members.
College students experience elevated risk for suicide. We want to use AI to determine who should be trained for gatekeeper training to allow the greatest coverage.
Using machine learning (a type of AI) to predict suicidal thinking through mining social network data of homeless youth.