Araştırma Makalesi
BibTex RIS Kaynak Göster

mRMR Tabanlı Açıklanabilir EEG Sınıflandırması ile Tüketici Tercihi Tahmini

Yıl 2025, Cilt: 5 Sayı: 2, 125 - 131, 23.12.2025

Öz

Bu çalışma, EEG sinyalleri ile tüketici beğenisini sınıflandırmayı amaçlamaktadır. Çalışmada açık erişimli bir EEG veri seti kullanılmış ve 18-38 yaşları arasındaki 25 katılımcıdan toplam 1045 EEG kaydı elde edilmiştir. Veriler, 128 Hz örnekleme frekansında 14 kanallı bir Emotiv Epoc+ cihazı ile kaydedilmiştir. Ön işleme sonrasında, her kanaldan zaman, entropi, istatistik ve spektral verilere dayalı olarak toplam 1190 özellik çıkarılmıştır. Özellik seçimi için Minimum Yedeklilik Maksimum Alaka Düzeyi (mYMA) algoritması kullanılmış ve en bilgilendirici altı özellik belirlenmiştir. Sınıflandırma aşamasında Destek Vektör Makineleri (DVM), K-En Yakın Komşu (KEYK), Naive Bayes (NB) ve Rastgele Orman (RO) algoritmaları uygulanmış ve model performansı 10 katlı çapraz doğrulama kullanılarak değerlendirilmiştir. Tüm özellik setiyle yapılan sınıflandırmada, RO algoritması %99 ile en yüksek doğruluk oranına ulaşmıştır. mYMA yöntemi kullanılarak seçilen yalnızca altı özellik kullanılarak bile, RO modeli %95 doğruluk, %95 F1 puanı ve %94 duyarlılık elde etmiştir. Çalışmanın önemli bir katkısı, yüksek doğruluk elde etmesinin yanı sıra, her bir özelliğin hangi EEG kanalına ve frekans bandına karşılık geldiğini açıklayarak modelin açıklanabilirliğini de artırmasıdır. Bu bağlamda, çalışma EEG tabanlı nöropazarlama çalışmalarına açıklanabilir bir yapay zeka yaklaşımı sunmaktadır. Sonuç olarak, mYMA yöntemi kullanılarak seçilen az sayıda özellik kullanılarak yüksek doğruluk ve yorumlanabilirliğe ulaşmak, hem hesaplama verimliliği hem de fizyolojik yorumlama açısından EEG tabanlı tüketici zevki tahmininde önemli bir ilerlemeyi temsil etmektedir.

Kaynakça

  • N. Lee, A. J. Broderick, and L. Chamberlain, “What is ‘neuromarketing’? A discussion and agenda for future research,” Int. J. Psychophysiol., vol. 63, no. 2, pp. 199–204, 2007.
  • S. Kumar, M. Yadava, and P. P. Roy, “Fusion of EEG response and sentiment analysis of products review to predict customer satisfaction,” Inf. Fusion, vol. 52, pp. 41–52, 2019.
  • M. F. K. Khondakar et al., “A systematic review on EEG-based neuromarketing: recent trends and analyzing techniques,” Brain Informatics, vol. 11, no. 1, p. 17, 2024.
  • M. Singh, M. Singh, and S. Gangwar, “Emotion recognition using electroencephalography (EEG): a review,” Int. J. Inf. Technol. Knowl. Manag., vol. 7, no. 1, 2013.
  • J. Zamani and A. B. Naieni, “Best feature extraction and classification algorithms for EEG signals in neuromarketing,” Front. Biomed. Technol., 2020.
  • M. Ouzir, H. C. Lamrani, R. L. Bradley, and I. El Moudden, “Neuromarketing and decision-making: Classification of consumer preferences based on changes analysis in the EEG signal of brain regions,” Biomed. Signal Process. Control, vol. 87, p. 105469, 2024.
  • R. N. Khushaba, C. Wise, S. Kodagoda, J. Louviere, B. E. Kahn, and C. Townsend, “Consumer neuroscience: Assessing the brain response to marketing stimuli using electroencephalogram (EEG) and eye tracking,” Expert Syst. Appl., vol. 40, no. 9, pp. 3803–3812, 2013.
  • G. Vecchiato et al., “On the use of EEG or MEG brain imaging tools in neuromarketing research,” Comput. Intell. Neurosci., vol. 2011, no. 1, p. 643489, 2011.
  • H. Peng, F. Long, and C. Ding, “Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 27, no. 8, pp. 1226–1238, 2005.
  • M. Yadava, P. Kumar, R. Saini, P. P. Roy, and D. P. Dogra, “Analysis of EEG signals and its application to neuromarketing,” Multimedia Tools Appl., vol. 76, no. 18, pp. 19087–19111, 2017.
  • J. G. Proakis and D. G. Manolakis, Digital Signal Processing: Principles, Algorithms, and Applications, 3rd ed. Upper Saddle River, NJ, USA: Prentice Hall, 1996.
  • L. Sörnmo and P. Laguna, Bioelectrical Signal Processing in Cardiac and Neurological Applications. Amsterdam, Netherlands: Academic Press, 2005.
  • S. Hargittai, “Savitzky–Golay least-squares polynomial filters in ECG signal processing,” 2005.
  • E. Dağdevir, D. Prof, and M. Tokmakçi, “Motor imge tabanlı beyin bilgisayar arayüzü sistemlerinde ön işleme yönteminin sınıflandırma performansına etkisinin araştırılması,” Erciyes Univ. Inst. Sci. Technol., Kayseri, Türkiye.
  • N. Sakib, M. K. Islam, and T. Faruk, “Machine learning model for computer‐aided depression screening among young adults using wireless EEG headset,” Comput. Intell. Neurosci., vol. 2023, no. 1, Jan. 2023, doi: 10.1155/2023/1701429.
  • H. Peng, “Python binding to mRMR feature selection algorithm.” [Online]. Available: https://github.com/fbrundu/pymrmr
  • C. Ding and H. Peng, “Minimum redundancy feature selection from microarray gene expression data,” J. Bioinform. Comput. Biol., vol. 3, pp. 185–205, 2005, doi: 10.1142/S0219720005001004.
  • G. G. Knyazev, “Motivation, emotion, and their inhibitory control mirrored in brain oscillations,” Neurosci. Biobehav. Rev., vol. 31, no. 3, pp. 377–395, 2007.
  • E. Başar and B. Güntekin, “A review of brain oscillations in cognitive processes and in the electroencephalogram: What we learned and what we need to know,” Int. J. Psychophysiol., vol. 86, no. 2, pp. 129–143, 2012.
  • Y. Zhang, Y. Chen, S. L. Bressler, and M. Ding, “Response preparation and inhibition: The role of the cortical sensorimotor beta rhythm,” Neuroscience, vol. 156, no. 1, pp. 238–246, 2008.

Consumer Preference Prediction with mRMR-Based Explainable EEG Classification

Yıl 2025, Cilt: 5 Sayı: 2, 125 - 131, 23.12.2025

Öz

This study aimed to classify consumer taste using EEG signals. An open-access EEG dataset was used in the study, and a total of 1045 EEG recordings were obtained from 25 participants aged 18–38. Data were recorded with a 14-channel Emotiv Epoc+ device at a sampling frequency of 128 Hz. After preprocessing, a total of 1190 features were extracted from each channel based on time, entropy, statistics, and spectral data. The Minimum Redundancy Maximum Relevance (mRMR) algorithm was used for feature selection, and the six most informative features were identified. Support Vector Machines (SVM), K-Nearest Neighbor (KNN), Naive Bayes (NB), and Random Forest (RF) algorithms were applied during the classification phase, and model performance was evaluated using 10-fold cross-validation. In the classification performed with the full feature set, the RF algorithm achieved the highest accuracy rate, with 99%. Even using only six features selected using the mRMR method, the RF model achieved 95% accuracy, an F1 score of 95%, and a sensitivity of 94%. A significant contribution of the study is that, in addition to achieving high accuracy, it also increases the model's explainability by clarifying which EEG channel and frequency band each feature corresponds to. In this respect, the study provides an explainable artificial intelligence approach to EEG-based neuromarketing studies. In conclusion, achieving high accuracy and interpretability using a small number of features selected using the mRMR method represents a significant advance in EEG-based consumer taste prediction in terms of both computational efficiency and physiological interpretation.

Kaynakça

  • N. Lee, A. J. Broderick, and L. Chamberlain, “What is ‘neuromarketing’? A discussion and agenda for future research,” Int. J. Psychophysiol., vol. 63, no. 2, pp. 199–204, 2007.
  • S. Kumar, M. Yadava, and P. P. Roy, “Fusion of EEG response and sentiment analysis of products review to predict customer satisfaction,” Inf. Fusion, vol. 52, pp. 41–52, 2019.
  • M. F. K. Khondakar et al., “A systematic review on EEG-based neuromarketing: recent trends and analyzing techniques,” Brain Informatics, vol. 11, no. 1, p. 17, 2024.
  • M. Singh, M. Singh, and S. Gangwar, “Emotion recognition using electroencephalography (EEG): a review,” Int. J. Inf. Technol. Knowl. Manag., vol. 7, no. 1, 2013.
  • J. Zamani and A. B. Naieni, “Best feature extraction and classification algorithms for EEG signals in neuromarketing,” Front. Biomed. Technol., 2020.
  • M. Ouzir, H. C. Lamrani, R. L. Bradley, and I. El Moudden, “Neuromarketing and decision-making: Classification of consumer preferences based on changes analysis in the EEG signal of brain regions,” Biomed. Signal Process. Control, vol. 87, p. 105469, 2024.
  • R. N. Khushaba, C. Wise, S. Kodagoda, J. Louviere, B. E. Kahn, and C. Townsend, “Consumer neuroscience: Assessing the brain response to marketing stimuli using electroencephalogram (EEG) and eye tracking,” Expert Syst. Appl., vol. 40, no. 9, pp. 3803–3812, 2013.
  • G. Vecchiato et al., “On the use of EEG or MEG brain imaging tools in neuromarketing research,” Comput. Intell. Neurosci., vol. 2011, no. 1, p. 643489, 2011.
  • H. Peng, F. Long, and C. Ding, “Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 27, no. 8, pp. 1226–1238, 2005.
  • M. Yadava, P. Kumar, R. Saini, P. P. Roy, and D. P. Dogra, “Analysis of EEG signals and its application to neuromarketing,” Multimedia Tools Appl., vol. 76, no. 18, pp. 19087–19111, 2017.
  • J. G. Proakis and D. G. Manolakis, Digital Signal Processing: Principles, Algorithms, and Applications, 3rd ed. Upper Saddle River, NJ, USA: Prentice Hall, 1996.
  • L. Sörnmo and P. Laguna, Bioelectrical Signal Processing in Cardiac and Neurological Applications. Amsterdam, Netherlands: Academic Press, 2005.
  • S. Hargittai, “Savitzky–Golay least-squares polynomial filters in ECG signal processing,” 2005.
  • E. Dağdevir, D. Prof, and M. Tokmakçi, “Motor imge tabanlı beyin bilgisayar arayüzü sistemlerinde ön işleme yönteminin sınıflandırma performansına etkisinin araştırılması,” Erciyes Univ. Inst. Sci. Technol., Kayseri, Türkiye.
  • N. Sakib, M. K. Islam, and T. Faruk, “Machine learning model for computer‐aided depression screening among young adults using wireless EEG headset,” Comput. Intell. Neurosci., vol. 2023, no. 1, Jan. 2023, doi: 10.1155/2023/1701429.
  • H. Peng, “Python binding to mRMR feature selection algorithm.” [Online]. Available: https://github.com/fbrundu/pymrmr
  • C. Ding and H. Peng, “Minimum redundancy feature selection from microarray gene expression data,” J. Bioinform. Comput. Biol., vol. 3, pp. 185–205, 2005, doi: 10.1142/S0219720005001004.
  • G. G. Knyazev, “Motivation, emotion, and their inhibitory control mirrored in brain oscillations,” Neurosci. Biobehav. Rev., vol. 31, no. 3, pp. 377–395, 2007.
  • E. Başar and B. Güntekin, “A review of brain oscillations in cognitive processes and in the electroencephalogram: What we learned and what we need to know,” Int. J. Psychophysiol., vol. 86, no. 2, pp. 129–143, 2012.
  • Y. Zhang, Y. Chen, S. L. Bressler, and M. Ding, “Response preparation and inhibition: The role of the cortical sensorimotor beta rhythm,” Neuroscience, vol. 156, no. 1, pp. 238–246, 2008.
Toplam 20 adet kaynakça vardır.

Ayrıntılar

Birincil Dil İngilizce
Konular Nöral Ağlar, Makine Öğrenme (Diğer), Veri Mühendisliği ve Veri Bilimi, Yapay Zeka (Diğer)
Bölüm Araştırma Makalesi
Yazarlar

Suzan Saban 0009-0009-8235-9108

Eda Dağdevir 0000-0001-7065-9829

Gönderilme Tarihi 11 Kasım 2025
Kabul Tarihi 9 Aralık 2025
Yayımlanma Tarihi 23 Aralık 2025
Yayımlandığı Sayı Yıl 2025 Cilt: 5 Sayı: 2

Kaynak Göster

IEEE S. Saban ve E. Dağdevir, “Consumer Preference Prediction with mRMR-Based Explainable EEG Classification”, Journal of Artificial Intelligence and Data Science, c. 5, sy. 2, ss. 125–131, 2025.