The abstract is read at the publisher; the summary is JClub's.
XGBoost classified binary psychological wellbeing status of 2,913 University of Peradeniya undergraduates with a 70.5% macro-F1 score and 72.5% test accuracy.
- Why it matters: Early detection of students with lower psychological wellbeing is needed, but actionable, interpretable signals from survey data are limited.
- What they did: Cross-sectional questionnaire data were preprocessed and used to train SVM, Random Forest, XGBoost, and MLP with repeated stratified 10-fold cross-validation; SHAP interpreted feature contributions.
- The result: XGBoost correctly classified 63.6% of lower-wellbeing and 80% of higher-wellbeing participants, and SHAP highlighted academic coping, health and wellbeing, financial circumstances, and interpersonal relationships as key predictors.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.