Title
Protecting Student Mental Health with a Context-Aware Machine Learning Framework for Stress Monitoring
Authors
Md Sultanul Islam Ovi, Jamal Hossain, Md Raihan Alam Rahi, Fatema Akter
Abstract
Student mental health is an increasing concern in academic institutions, where stress can severely impact well-being and academic performance. Traditional assessment methods rely on subjective surveys and periodic evaluations, offering limited value for timely intervention. This paper introduces a context-aware machine learning framework for classifying student stress using two complementary survey-based datasets covering psychological, academic, environmental, and social factors. The framework follows a six-stage pipeline involving preprocessing, feature selection (SelectKBest, RFECV), dimensionality reduction (PCA), and training with six base classifiers: SVM, Random Forest, Gradient Boosting, XGBoost, AdaBoost, and Bagging. To enhance performance, we implement ensemble strategies, including hard voting, soft voting, weighted voting, and stacking. Our best models achieve 93.09% accuracy with weighted hard voting on the Student Stress Factors dataset and 99.53% with stacking on the Stress and Well-being dataset, surpassing previous benchmarks. These results highlight the potential of context-integrated, data-driven systems for early stress detection and underscore their applicability in real-world academic settings to support student well-being.
Conference
Submitted to a Conference
PDF
ArXiv
BibTeX
@article{ovi2025protecting,
title={Protecting Student Mental Health with a Context-Aware Machine Learning Framework for Stress Monitoring},
author={Ovi, Md Sultanul Islam and Hossain, Jamal and Rahi, Md Raihan Alam and Akter, Fatema},
journal={arXiv preprint arXiv:2508.01105},
year={2025}
}
- File: stressLevelDataset.csv
- Rows: 1100
- Columns: 21
- Missing Values: No
- Duplicates: None
- Target: stress_level (3 classes) This dataset includes psychological, physiological, academic, social, and environmental stress factors reported by students in a nationwide survey.
- File: Stress Dataset.csv
- Rows: 843
- Columns: 26
- Missing Values: No
- Duplicates: 27
- Target: Type of stress (eustress or distress or no stress)
Collected via Google Forms, this dataset captures emotional, academic, and health-related stress indicators from college students aged 18–21.
| ML Model | Configuration | Accuracy | F1 Score | Recall | Precision |
|---|---|---|---|---|---|
| Voting Classifier (hard) | Mixed Preprocessing | 93.091% | 93.099% | 93.086% | 93.126% |
| Voting Classifier (weighted_hard) | Mixed Preprocessing | 93.091% | 93.091% | 93.086% | 93.099% |
| Random Forest | SelectKBest | 92.364% | 92.365% | 92.356% | 92.378% |
| Voting Classifier (soft) | Mixed Preprocessing | 92.364% | 92.356% | 92.356% | 92.356% |
| Voting Classifier (weighted_soft) | Mixed Preprocessing | 92.364% | 92.365% | 92.356% | 92.378% |
| AdaBoost | SelectKBest | 92.000% | 92.005% | 92.083% | 92.121% |
| XGBoost | SelectKBest | 91.636% | 91.636% | 91.776% | 92.014% |
| Gradient Boosting | SelectKBest | 91.273% | 91.279% | 91.404% | 91.498% |
| Stacking Classifier | Mixed Preprocessing | 91.273% | 91.252% | 91.220% | 91.317% |
| Support Vector Machine | PCA | 90.546% | 90.557% | 90.551% | 90.567% |
| Bagging | SelectKBest | 89.455% | 89.493% | 89.508% | 89.716% |
| ML Model | Configuration | Accuracy | F1 Score | Recall | Precision |
|---|---|---|---|---|---|
| Stacking Classifier | Mixed Preprocessing | 99.530% | 97.950% | 99.830% | 96.300% |
| Support Vector Machine | PCA | 99.052% | 96.494% | 93.939% | 99.656% |
| Voting Classifier (weighted_hard) | Mixed Preprocessing | 99.052% | 96.494% | 93.939% | 99.656% |
| Voting Classifier (weighted_soft) | Mixed Preprocessing | 99.052% | 96.494% | 93.939% | 99.656% |
| Voting Classifier (soft) | Mixed Preprocessing | 97.630% | 89.942% | 83.712% | 99.154% |
| Voting Classifier (hard) | Mixed Preprocessing | 97.156% | 87.318% | 79.546% | 98.990% |
| AdaBoost | Original Data | 96.683% | 85.866% | 79.372% | 94.818% |
| XGBoost | SelectKBest | 96.209% | 82.058% | 73.485% | 98.667% |
| Gradient Boosting | Normalized Data | 95.735% | 79.044% | 69.318% | 98.508% |
| Random Forest | SelectKBest | 94.787% | 72.332% | 63.258% | 98.194% |
| Bagging | SelectKBest | 94.313% | 68.831% | 59.091% | 98.039% |
