Lightweight Ensemble Framework for Predicting Student Engagement Levels Using Synthetic Time-Series Data

DOI

10.2991/978-94-6463-718-2_135How to use a DOI?

Keywords

student engagement; lightweight ensemble framework; synthetic data; time-series analysis; predictive modeling; privacy-preserving

Abstract

In education, Academic participation is crucial to achievement and retention. Predicting the level of engagement accurately is still challenging because of privacy, computational inefficiencies, and the ever-changing nature of interaction. To overcome these challenges, we propose a Lightweight Ensemble Framework for predicting student engagement levels in synthetic time-series data. Interesting synthetic data can be used at scale without privacy risks which also allows this framework to generate diverse and representative datasets for model training. The proposed framework incorporates temporal dynamics and applies an ensemble of lightweight machine learning algorithms with high predictive accuracy and computational efficiency. The experimental results show that the framework outperforms its predecessors by achieving a 91.5% accuracy and a low MSE score. One of the important predictors was temporal features, such as “Engagement Change Rate” and “Cumulative Interaction Score”. Furthermore, the validation of synthetic data established its accurate representation of real-world datasets and policy as a privacy-preserving technique. The proposed approach provides a scalable, ethical, and interpretable solution for engagement prediction, thus enabling emerging data-informed and personalized learning interventions in a wide spectrum of educational settings.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - M. K. Nivodhini
AU  - P. Vasuki
AU  - R. Banupriya
AU  - V. Ananthabarani
AU  - K. M. Dilip Kumar
AU  - K. Jayadev
PY  - 2025
DA  - 2025/05/23
TI  - Lightweight Ensemble Framework for Predicting Student Engagement Levels Using Synthetic Time-Series Data
BT  - Proceedings of the International Conference on Sustainability Innovation in Computing and Engineering (ICSICE 2024)
PB  - Atlantis Press
SP  - 1618
EP  - 1629
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-718-2_135
DO  - 10.2991/978-94-6463-718-2_135
ID  - Nivodhini2025
ER  -