Cognitive Load Classification Using Feature Masked Autoencoding and Electroencephalogram Signals

D. Eesha, D. Eesha, M. Nagaraju, M. Nagaraju, D. Divya, D. Divya, S. Kashyap Reddy, S. Kashyap Reddy

2025

Abstract

Electroencephalogram based Cognitive Load Classification has a wider range of applications that benefit different domains such as healthcare and adaptive systems. The paper explores the classification of cognitive load levels using EEG data through two different experiments: a standard machine learning model and an advanced Transformer-based autoencoding. The first experiment provides a moderate accuracy of 55%, indicating major differences in precision and recall, especially regarding positive cases. The second experiment uses a Masked Autoencoder pre-trained Transformer model, attaining a remarkable accuracy of 91% with balanced classification metrics across both classes. The paper showcases the effectiveness of deep learning in cognitive load classification, with significant potential for real-time applications across the medical field.

Paper Citation

in Harvard Style

Eesha D., Nagaraju M., Divya D. and Kashyap Reddy S. (2025). Cognitive Load Classification Using Feature Masked Autoencoding and Electroencephalogram Signals. In Proceedings of the 3rd International Conference on Futuristic Technology - Volume 2: INCOFT; ISBN 978-989-758-763-4, SciTePress, pages 624-634. DOI: 10.5220/0013599000004664

in Bibtex Style

@conference{incoft25,
author={D. Eesha and M. Nagaraju and D. Divya and S. Kashyap Reddy},
title={Cognitive Load Classification Using Feature Masked Autoencoding and Electroencephalogram Signals},
booktitle={Proceedings of the 3rd International Conference on Futuristic Technology - Volume 2: INCOFT},
year={2025},
pages={624-634},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013599000004664},
isbn={978-989-758-763-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 3rd International Conference on Futuristic Technology - Volume 2: INCOFT
TI - Cognitive Load Classification Using Feature Masked Autoencoding and Electroencephalogram Signals
SN - 978-989-758-763-4
AU - Eesha D.
AU - Nagaraju M.
AU - Divya D.
AU - Kashyap Reddy S.
PY - 2025
SP - 624
EP - 634
DO - 10.5220/0013599000004664
PB - SciTePress