Paper Details

Name: Priyanka Gupta & Anil Pandit
Country: India

Year: 2026
Volume: Volume No: 13, January, Year: 2026 (Special Issue)
Page Number: 622-628
DOI: https://doi.org/10.5281/zenodo.19090443

ABSTRACT:
Student stress is a significant concern affecting academic performance, mental well-being, and overall health. This study examines the use of ensemble learning methods, particularly hard voting and soft voting classifiers, on a dataset that includes several academic, social, physiological, psychological, and environmental elements that affect students' stress levels. A range of machine learning models, including Decision Trees, Support Vector Machines, and Logistic Regression, are used in ensemble frameworks. Performance indicators including accuracy, precision, recall, and F1-score are used to
assess the efficacy of hard versus soft voting techniques. The results demonstrate the advantages of soft voting in handling
complex, multi-dimensional data in student stress prediction.Furthermore, we analyse the feature importance and the impact of individual predictors on stress levels, which provides deeper insights into the contributing factors. This research highlights the importance of ensemble learning in psychological and academic research, offering improved stress prediction and early intervention possibilities

KEYWORDS: Ensemble Learning, Hard Voting, Soft Voting, Student Stress Prediction, Machine Learning, Psychological Factors, Physiological Factors , Academic Performance