SavedAdd your journal

Journals / Uludağ Üniversitesi Mühendislik Fakültesi Dergisi / 2025 / Cilt: 30 Sayı: 3

EEG-BASED FAMILIAR AND UNFAMILIAR FACE ANALYSIS: CLASSIFICATION USING RMS FEATURES WITH ARTIFICIAL NEURAL NETWORKS AND RANDOM FORESTS

EEG Tabanlı Tanıdık ve Tanıdık Olmayan Yüz Analizi: RMS Özellikleri ile Yapay Sinir Ağları ve Rastgele Orman Sınıflandırma

Fatma Kebire Bardak Özkul(BANDIRMA ONYEDI EYLUL UNIVERSITY)0000-0002-9380-2330
Feyzullah Temurtaş(BANDIRMA ONYEDI EYLUL UNIVERSITY)0000-0002-3158-4032
Pages
765–778
Publication Date
2025-12-19
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The results showed that the PNN model exhibited the highest performance with an accuracy rate of 95.05%.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:In this study, the features obtained using the Root Mean Square (RMS) method in EEG-based face recognition processes were analyzed with probabilistic neural networks (PNN), multilayer perceptrons (MLP), and random forest classifiers.
📊
Key Findings & Data:The results showed that the PNN model exhibited the highest performance with an accuracy rate of 95.05%.
💡
Conclusion & Impact:It suggests that more personalized models should be developed to balance these differences in the future.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Tanıdık ve Tanıdık Olmayan Yüz · EEG · Yapay Sinir Ağları

Abstract

In this study, the features obtained using the Root Mean Square (RMS) method in EEG-based face recognition processes were analyzed with probabilistic neural networks (PNN), multilayer perceptrons (MLP), and random forest classifiers. The results showed that the PNN model exhibited the highest performance with an accuracy rate of 95.05%. On the other hand, the MLP and Random Forest models showed lower performance with an accuracy rate of 73.34% and 78.01%, respectively. These differences may be due to the variability in EEG topographic responses among individuals and the inability of these models to generalize the differences in the data well enough. The study emphasizes the importance of considering individual neural differences in EEG-based classification systems. It suggests that more personalized models should be developed to balance these differences in the future.

Özet

Bu çalışmada, EEG tabanlı yüz tanıma işlemlerinde Kök Ortalama Kare (RMS) yöntemi kullanılarak elde edilen özellikler olasılıksal sinir ağları (PNN), çok katmanlı algılayıcılar (MLP) ve rastgele orman sınıflandırıcıları ile analiz edilmiştir. Sonuçlar, PNN modelinin %95.05 doğruluk oranıyla en yüksek performansı sergilediğini göstermiştir. Öte yandan, MLP ve Rastgele Orman modelleri sırasıyla %73.34 ve %78.01 doğruluk oranıyla daha düşük performans göstermiştir. Bu farklılıklar, bireyler arasındaki EEG topografik tepkilerinin değişkenliğinden ve bu modellerin verilerdeki farklılıkları yeterince iyi genelleştirememesinden kaynaklanıyor olabilir. Çalışma, EEG tabanlı sınıflandırma sistemlerinde bireysel sinirsel farklılıkların dikkate alınmasının önemini vurgulamaktadır. Gelecekte bu farklılıkları dengelemek için daha kişiselleştirilmiş modellerin geliştirilmesi gerektiğini önermektedir.

🔖

Cite This Article

Özkul, F. K. B., Temurtaş, F. (2025). EEG-BASED FAMILIAR AND UNFAMILIAR FACE ANALYSIS: CLASSIFICATION USING RMS FEATURES WITH ARTIFICIAL NEURAL NETWORKS AND RANDOM FORESTS. Uludağ Üniversitesi Mühendislik Fakültesi Dergisi, Cilt: 30 Sayı: 3, 765–778. https://doi.org/10.17482/uumfd.1573758

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (18)

  1. Bardak, F. K., Seyman, M. N., and Temurtaş, F. (2022). EEG based emotion prediction with neural network models. Tehnički Glasnik, 16(4), 497-502. doi:10.31803/tg-20220330064309
  2. Bardak, F. K., Seyman, M. N., and Temurtaş, F. (2024). Adaptive neuro-fuzzy based hybrid classification model for emotion recognition from EEG signals. Neural Computing and Applications, 1-14. doi:10.1007/s00521-024-09573-6
  3. Bardak, F. K., and Temurtaş, F. (2024). A Review of EEG-Based Face Recognition: Methodologies, Feature Extraction Techniques, and Classification Methods. AIntelia Science Notes, 3(1), 1-11. doi: 10.5281/zenodo.14557085
  4. Chang, W., Wang, H., Yan, G., and Liu, C. (2020). An EEG based familiar and unfamiliar person identification and classification system using feature extraction and directed functional brain network. Expert Systems with Applications, 158, 113448. doi: 10.1016/j.eswa.2020.113448
  5. Farizal, A., Wibawa, A. D., Pamungkas, Y., Pratiwi, M., and Mas, A. (2022). Classifying known/unknown information in the brain using electroencephalography (EEG) signal analysis. In 2022 11th Electrical Power, Electronics, Communications, Controls. doi: 10.1109/EECCIS54468.2022.9902928
  6. Görür, K., Bozkurt, M. R., Başçıl, M. S., and Temurtaş, F. (2016). Literature Research: Bruxism. Electronic Letters on Science and Engineering, 12(3), 11-17.
  7. Jackson, M. C., and Raymond, J. E. (2006). The role of attention and familiarity in face identification. Perception and Psychophysics, 68, 543-557. doi:10.3758/BF03208757
  8. Karacan, S. Ş., and Saraoğlu, H. M. (2024). A simplified method for relapsing-remitting multiple sclerosis detection: Insights from resting EEG signals. Computers in Biology and Medicine, 108728. doi: 10.1016/j.compbiomed.2024.108728
  9. Kramer, R. S., Young, A. W., and Burton, A. M. (2018). Understanding face familiarity. Cognition, 172, 46-58. doi: 10.1016/j.cognition.2017.12.005
  10. Liu, G., Wen, Y., Hsiao, J. H., Zhang, D., Tian, L., and Zhou, W. (2023). EEG-Based Familiar and Unfamiliar Face Classification Using Filter-Bank Differential Entropy Features. IEEE Transactions on Human-Machine Systems, 54(1), 44-55. doi: 10.1109/THMS.2023.3332209
  11. Liu, G., Zhang, D., Tian, L., and Zhou, W. (2021). EEG-based familiar and unfamiliar face classification using differential entropy feature. In 2021 IEEE 2nd International Conference on Human-Machine Systems (ICHMS), 1-3. doi: 10.1109/ICHMS53169.2021.9582641
  12. Natu, V., and O’Toole, A. J. (2011). The neural processing of familiar and unfamiliar faces: A review and synopsis. British Journal of Psychology, 102(4), 726-747. doi: 10.1111/j.2044-8295.2011.02053.x
  13. Özbeyaz, A., and Arica, S. (2017). Classification of EEG signals of familiar and unfamiliar face stimuli exploiting most discriminative channels. Turkish Journal of Electrical Engineering and Computer Sciences, 25(4), 3342-3354. doi: 10.3906/elk-1608-13
  14. Podvigina, D. N., and Prokopenya, V. K. (2019). Role of familiarity in recognizing faces and words: an EEG study. Sovremennye Tehnologii v Medicine, 11(1), 76-82. doi: 10.17691/stm2019.11.1.09
  15. Tan, Z. H. E., Smitha, K. G., and Vinod, A. P. (2015). Detection of familiar and unfamiliar images using EEG-based brain-computer interface. In 2015 IEEE International Conference on Systems, Man, and Cybernetics, 3152-3157. doi: 10.1109/SMC.2015.547
  16. Temurtaş, H., and Temurtaş, F. (2016). A Study on Continuous Resilient Average for Sensor Transient Response. Electronic Letters on Science and Engineering, 12(1), 1-7.
  17. Vanzara, N., Shah, C. P., and Vithalani, A. (2024). Detection of Familiar and Unfamiliar faces from EEG. Journal of Integrated Science and Technology, 12(1), 715-715.
  18. Yan, X., Volfart, A., and Rossion, B. (2023). A neural marker of the human face identity familiarity effect. Scientific Reports, 13(1), 16294. doi:10.1038/s41598-023-40852-9