An early assessment of subjective well-being in an undergraduate by using heterogeneous stacked ensemble

Subjective Well-Being (SWB) is a significant indicator of undergraduate students’ life satisfaction and social welfare. Machine learning is gaining increasing prominence as a valuable tool in various fields. However, limited studies have employed machine learning techniques to explore SWB. To predic...

Full description

Saved in:
Bibliographic Details
Main Authors: Upadhyay, Devesh Kumar, Mohapatra, Subrajeet, Singh, Niraj Kumar
Format: Conference Proceeding
Language:English
Subjects:
Classifiers
Machine learning
Psychological factors
Students
Undergraduate study
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Subjective Well-Being (SWB) is a significant indicator of undergraduate students’ life satisfaction and social welfare. Machine learning is gaining increasing prominence as a valuable tool in various fields. However, limited studies have employed machine learning techniques to explore SWB. To predict Subjective Well-Being among undergraduate students, this paper introduces a Heterogeneous Stacked Ensemble classifier model, which combines XGBoost, SVM GridSearch, LR, LightGBM, Bayes Net, and CatBoost classifiers. Additionally, the feature importance index derived from tree models is utilized to uncover shifts in the key factors influencing SWB. The results indicate that the stacking model proposed in this study surpasses traditional models like LR and other standalone machine learning models regarding predictive accuracy. Furthermore, the findings shed light on common factors contributing to SWB over time. The methods employed in this research demonstrate effectiveness, and the outcomes endorse endeavours aimed at fostering a more harmonious society.
ISSN:0094-243X
1551-7616
DOI:10.1063/5.0214163