Using Boosted Machine Learning to Predict Suicidal Ideation by Socioeconomic Status among Adolescents

Abstract

(1) Background: This study aimed to use machine learning techniques to identify risk factors for suicidal ideation among adolescents and understand the association between these risk factors and socioeconomic status (SES); (2) Methods: Data from 54,948 participants were analyzed. Risk factors were identified by dividing groups by suicidal ideation and 3 SES levels. The influence of risk factors was confirmed using the synthetic minority over-sampling technique and XGBoost; (3) Results: Adolescents with suicidal thoughts experienced more sadness, higher stress levels, less happiness, and higher anxiety than those without. In the high SES group, academic achievement was a major risk factor for suicidal ideation; in the low SES group, only emotional factors such as stress and anxiety significantly contributed to suicidal ideation; (4) Conclusions: SES plays an important role in the mental health of adolescents. Improvements in SES in adolescence may resolve their negative emotions and reduce the risk of suicide.