CNN-Based Student Attentiveness Detection: A Hybrid Approach with Dimensional Reduction

DOI

10.2991/978-94-6463-884-4_81How to use a DOI?

Keywords

Attentiveness; Dimensional Reduction; Drowsiness; Lightweight; PCA

Abstract

Student attentiveness is widely known as essential in effective teaching, and is a task that is tough to sustain in traditional and online classrooms. In this study, a CNN-based solution is proposed to identify student attentiveness where the extent of the eye indicates their presence or absence. This solution uses Principal Component Analysis (PCA), Convolutional Neural Network (CNN) technique. CNN was chosen due to its ability to extract spatial and hierarchical features from eye images in its image representation. PCA was used to reduce the high-dimensional feature space created by CNN. The integration of this model reduced redundancy, increased computational efficiency, and improved the generalization capacity of the model by retaining only the most important features, which makes it a very lightweight model compared to others. The proposed model was trained on Drowsiness Detection Dataset which contains open and closed eye states. Using our fusion approach, we reached an accuracy of 99.81% with excellent precision, recall, and the F1-score.

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  - Nazmus Sakib Md Adil
AU  - Kowshik Das Ushna
AU  - Md. Tasnimur Rahman
AU  - Shuhena Salam Aonty
PY  - 2025
DA  - 2025/11/18
TI  - CNN-Based Student Attentiveness Detection: A Hybrid Approach with Dimensional Reduction
BT  - Proceedings of the 8th International Conference on Engineering Research, Innovation, and Education 2025 (ICERIE 2025)
PB  - Atlantis Press
SP  - 672
EP  - 679
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6463-884-4_81
DO  - 10.2991/978-94-6463-884-4_81
ID  - Adil2025
ER  -