Ishan Mysore

Predicting Depression Risk for Adolescents through Social Media Post Data: A Machine Learning Approach

Started June 13, 2024

Abstract or project description

In this digital age, the prevalence of social media platforms in adolescents’ lives is linked with a rise in adolescent depression rates. According to a survey from the Kaiser Family Foundation (KFF), over 17% of adolescents report experiencing symptoms of depression; however, 20% of at-risk adolescents don’t receive the necessary support. If left unaddressed, these symptoms can invoke attempts at suicide, which is the fourth leading cause of death among those aged 15-29. This paper proposes a predictive model that leverages social media post data to identify early signs of depression in adolescents and recommend proactive treatment and care. Users can enter their Twitter or Instagram handle, and the application will scrape their 100 most recent posts, assign a risk score to each post, and from there predict an overall risk level for depression. The website will then provide links to some reputable mental health sources and, if necessary, recommend the adolescent to seek a formal diagnosis. This project aims to enhance early intervention efforts for adolescent depression to proactively treat anyone at risk. If successfully integrated with online support platforms, the application could provide a seamless connection between at-risk teens and mental health professionals, thereby reducing the time to treatment and potentially saving lives.