Harnessing Deep Learning for EEG Emotion Recognition: A Hybrid Approach with Attention Mechanisms

Ali H. Abdulwahhab; Alaa Hussein Abdulaal; Ali M. Jasim; Riyam Ali Yassin; Morteza Valizadeh; Ahmed Nidham Qasim; A. F. M. Shahen Shah; Mehdi Chehel Amirani

doi:10.58564/IJSER.4.2.2025.323

Authors

Ali H. Abdulwahhab Department of Electrical and Computer Engineering, Altinbas University, Istanbul, Turkey
Alaa Hussein Abdulaal Department of Electrical Engineering, Urmia University, West Azerbaijan
Ali M. Jasim Department of Space Technology and Non-conventional Energy Sources, National Aerospace University, Kharkiv, Ukraine
Riyam Ali Yassin Department of Electrical Engineering, Urmia University, West Azerbaijan
Morteza Valizadeh Department of Electrical Engineering, Urmia University, West Azerbaijan
Ahmed Nidham Qasim Department of Electronic and Control Engineering, Northern Technical University, Mosul, Iraq
A. F. M. Shahen Shah Department of Electronics and Communication Engineering, Yildiz Technical University, Istanbul, Turkey
Mehdi Chehel Amirani Department of Electrical Engineering, Urmia University, West Azerbaijan

DOI:

https://doi.org/10.58564/IJSER.4.2.2025.323

Keywords:

Active Appearance Models (AAMs), Convolutional Neural Networks (CNNs), Face recognition, Noise, Viola-Jones algorithm

Abstract

Emotion recognition from EEG signals has emerged as a pivotal area of research, driven by its transformative potential in healthcare, brain-computer interfaces, and affective computing systems. However, the intrinsic complexity, non-linearity, and susceptibility to noise in EEG data present significant challenges to accurate emotional state classification. This study proposes a robust and interpretable hybrid deep learning model for EEG-based emotion recognition. The architecture integrates Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) networks, and attention mechanisms, together with advanced signal processing techniques such as Continuous Wavelet Transform (CWT) and Power Spectral Density (PSD). This integrated approach facilitates the extraction of comprehensive spatial, temporal, and spectral features from EEG signals, enhancing the model’s ability to capture intricate patterns associated with emotional states. Experimental evaluations on the SEED-IV dataset, encompassing four emotional categories—Neutral, Happy, Sad, and Fear—demonstrated the model’s exceptional performance, achieving a macro-average F1-score of 93% and an area under the ROC curve (AUC) of 0.94. These results validate the model’s effectiveness in accurately distinguishing complex emotional patterns, even under noisy conditions and inter-class ambiguities. Overall, this research advances the domain of EEG-based emotion recognition by introducing a high-performing, interpretable framework suitable for real-world applications while laying the foundation for future developments in adaptive neurofeedback systems and emotion-aware brain-computer interfaces.

References

[1] R. Ranjan, B. Chandra Sahana, and A. Kumar Bhandari, “Deep Learning Models for Diagnosis of Schizophrenia Using EEG Signals: Emerging Trends, Challenges, and Prospects,” Archives of Computational Methods in Engineering, vol. 31, no. 4, pp. 2345–2384, Jan. 2024,

doi: https://doi.org/10.1007/s11831-023-10047-6.

[2] A. H. Abdulwahhab, Indrit Myderrizi, and Muhammet Mustafa Yurdakul, “PAFWF-EEGC Net: parallel adaptive feature weight fusion based on EEG-dynamic characteristics using channels neural network for driver drowsiness detection,” Signal Image and Video Processing, vol. 19, no. 7, May 2025,

doi: https://doi.org/10.1007/s11760-025-04102-x.

[3] A. H. Abdulwahhab, Indrit Myderrizi, and M. K. Mahmood, “Drone Movement Control by Electroencephalography Signals Based on BCI System,” Advances in Electrical and Electronic Engineering, vol. 20, no. 2, Jun. 2022,

doi: https://doi.org/10.15598/aeee.v20i2.4413.

[4] A. H. Abdulaal, M. Valizadeh, M. C. Amirani, and A. F. M. Shahen Shah, “A self-learning Deep Neural Network for Classification of Breast Histopathological Images,” Biomedical Signal Processing and Control, vol. 87, no. B, p. 105418, Jan. 2024,

doi: https://doi.org/10.1016/j.bspc.2023.105418.

[5] A. H. Abdulaal et al., “Unsupervised Histopathological Sub-Image Analysis for Breast Cancer Diagnosis Using Variational Autoencoders, Clustering, and Supervised Learning,” Journal of Engineering and Sustainable Development, vol. 28, no. 6, pp. 729–744, Nov. 2024,

doi: https://doi.org/10.31272/jeasd.28.6.6.

[6] H. S. Mansour, M. Valizadeh, A. H. Abdulaal, and M. C. Amirani, “Enhanced Electrocardiogram Arrhythmia Diagnosis with Deep Learning and Selective Attention Mechanism,” Advances in Technology Innovation, 2025,

doi: https://doi.org/10.46604/aiti.2024.14034.

[7] C. J. O. Echeverri, S. Salazar-Colores, and G. Hernández-Nava, “A Study of the Relationship of Wavelet Transform Parameters and Their Impact on EEG Classification Performance,” Brain-Computer Interfaces, pp. 115–130, Apr. 2025,

doi: https://doi.org/10.1016/B978-0-323-95439-6.00012-0.

[8] A. H. Abdulaal et al., “Deep Learning-based Signal Identification in Wireless Communication Systems: a Comparative Analysis on 3G, LTE, and 5G Standards,” Al-Iraqia Journal for Scientific Engineering Research, vol. 3, no. 3, pp. 60–70, 2024,

doi: https://doi.org/10.58564/IJSER.3.3.2024.224.

[9] A. H. Abdulaal et al., “Hybrid CNN and RNN Model for Histopathological Sub-Image Classification in Breast Cancer Analysis Using Self-Learning,” Journal of Engineering and Sustainable Development, vol. 29, no. 3, pp. 310–320, May 2025,

doi: https://doi.org/10.31272/jeasd.2746.

[10] N. Bajaj, “Wavelets for EEG Analysis,” Wavelet Theory, Feb. 2021, doi: https://doi.org/10.5772/intechopen.94398.

[11] M. Liu, Y. Lu, S. Long, J. Bai, and W. Lian, “An attention-based CNN-BiLSTM Hybrid Neural Network Enhanced with Features of Discrete Wavelet Transformation for Fetal Acidosis Classification,” Expert Systems with Applications, vol. 186, pp. 115714–115714, Dec. 2021,

doi: https://doi.org/10.1016/j.eswa.2021.115714.

[12] M. M. Farag, “Design and Analysis of Convolutional Neural Layers: a Signal Processing Perspective,” IEEE Access, vol. 11, pp. 27641–27661, 2023,

doi: https://doi.org/10.1109/access.2023.3258399.

[13] A. H. Abdulwahhab, A. H. Abdulaal, Assad, A. A. Mohammed, and M. Valizadeh, “Detection of Epileptic Seizure Using EEG Signals Analysis Based on Deep Learning Techniques,” Chaos, Solitons & fractals/Chaos, Solitons and Fractals, vol. 181, pp. 114700–114700, Apr. 2024,

doi: https://doi.org/10.1016/j.chaos.2024.114700.

[14] S. Du, W. Han, Z. Kang, X. Lu, Y. Liao, and Z. Li, “Continuous Wavelet Transform Peak-Seeking Attention Mechanism Conventional Neural Network: a Lightweight Feature Extraction Network with Attention Mechanism Based on the Continuous Wave Transform Peak-Seeking Method for Aero-Engine Hot Jet Fourier Transform Infrared Classification,” Remote Sensing, vol. 16, no. 16, p. 3097, Aug. 2024,

doi: https://doi.org/10.3390/rs16163097.

[15] C. Sun, C. Xu, H. Li, H. Bo, L. Ma, and H. Li, “A Novel multi-feature Fusion Attention Neural Network for the Recognition of Epileptic EEG Signals,” Frontiers in Computational Neuroscience, vol. 18, Jun. 2024,

doi: https://doi.org/10.3389/fncom.2024.1393122.

[16] I. S. Masad, A. Alqudah, and S. Qazan, “Automatic Classification of Sleep Stages Using EEG Signals and Convolutional Neural Networks,” PLOS ONE, vol. 19, no. 1, pp. e0297582–e0297582, Jan. 2024,

doi: https://doi.org/10.1371/journal.pone.0297582.

[17] N. Berrahou, A. E. Alami, R. E. Alami, and H. Qjidaa, “Synergistic Approaches for Accurate Arrhythmia Prediction: a Hybrid AI Model Integrating Higuchi Dimensional Fractal, RR-intervals and Attention-based Convolutional Neural Network in ECG Signal Analysis,” Statistics Optimization & Information Computing, vol. 13, no. 2, pp. 547–567, Mar. 2024,

doi: https://doi.org/10.19139/soic-2310-5070-2091.

[18] W.-L. Zheng, W. Liu, Y. Lu, B.-L. Lu, and A. Cichocki, “EmotionMeter: a Multimodal Framework for Recognizing Human Emotions,” IEEE Transactions on Cybernetics, vol. 49, no. 3, pp. 1110–1122, Mar. 2019,

doi: https://doi.org/10.1109/tcyb.2018.2797176.

[19] Z. Huang, Y. Ma, R. Wang, W. Li, and Y. Dai, “A Model for EEG-Based Emotion Recognition: CNN-Bi-LSTM with Attention Mechanism,” Electronics, vol. 12, no. 14, pp. 3188–3188, Jul. 2023,

doi: https://doi.org/10.3390/electronics12143188.

[20] M. R. Khan, A. A. Tania, and M. Ahmad, “A Comparative Study of Time-frequency Features Based spatio-temporal Analysis with Varying Multiscale Kernels for Emotion Recognition from EEG,” Biomedical Signal Processing and Control, vol. 107, p. 107826, Sep. 2025,

doi: https://doi.org/10.1016/j.bspc.2025.107826.

[21] K. F. Walters et al., “Resting-State EEG Power Spectral Density Analysis between Healthy and Cognitively Impaired Subjects,” Brain Sciences, vol. 15, no. 2, p. 173, Feb. 2025,

doi: https://doi.org/10.3390/brainsci15020173.

[22] A. H. Abdulaal et al., “Cutting-edge CNN Approaches for Breast Histopathological Classification: the Impact of Spatial Attention Mechanisms,” ShodhAI: Journal of Artificial Intelligence, vol. 1, no. 1, Oct. 2024,

doi: https://doi.org/10.29121/shodhai.v1.i1.2024.14.

[23] A. H. Abdulwahhab, O. Bayat, and A. A. Ibrahim, “HAFMAB-Net: hierarchical adaptive fusion based on multilevel attention-enhanced bottleneck neural network for breast histopathological cancer classification,” Signal Image and Video Processing, vol. 19, no. 5, Mar. 2025,

doi: https://doi.org/10.1007/s11760-025-04001-1.

[24] F. Rivas, J. Enrique Sierra-Garcia, and J. M. Camara, “Comparison of LSTM- and GRU-Type RNN Networks for Attention and Meditation Prediction on Raw EEG Data from Low-Cost Headsets,” Electronics, vol. 14, no. 4, pp. 707–707, Feb. 2025,

doi: https://doi.org/10.3390/electronics14040707.

[25] M. A. Mulkey, H. Huang, T. Albanese, S. Kim, and B. Yang, “Supervised Deep Learning with Vision Transformer Predicts Delirium Using Limited Lead EEG,” Scientific Reports, vol. 13, no. 1, p. 7890, May 2023,

doi: https://doi.org/10.1038/s41598-023-35004-y.