BingLi2026/AI-Student-Mental-Health-Detection

AI Student Mental Health Detection

Research code for The Application of AI-Based Early Detection of Mental Health Issues Among University Students in Academic Settings by Bing Li and Wenkai Li.

The project implements a modular HybridDL-ML-Ensemble workflow combining Random Forest, gradient boosting, MLP, CNN-1D, and logistic-regression stacking for student mental health risk detection.

Reported Results

Repository Structure

• —data/: dataset loading and preprocessing
• —models/: model interfaces, encoding, alignment, and retrieval
• —strategies/: relevance, intent, and sequence analysis
• —experiments/: training, evaluation, and visualization
• —utils/: shared reproducibility and artifact helpers
• —main.py: command-line entry point

Local Usage

pip install -r requirements.txt
python main.py --input data/student_mental_health.csv --output_dir results --mode train

Dataset reference: Student Mental Health on Kaggle.

Authors

• —Bing Li, School of Foreign Languages and Literature, Shan Dong University
• —Wenkai Li, Department of Mathematics and Statistics, Manchester University

Responsible Use

This project is intended for research and educational use. Predictions must not be treated as clinical diagnoses or used as the sole basis for interventions affecting students.