Title: Comparison of Dimension Reduction Algorithms on EEG Signals

Abstract: Like in all classification applications, the most important process which increases classification success of electroencephalography (EEG) applications is to choose the proper features for signals. Since there is not certain feature extraction method for data classification applications, used feature matrix size can be redundantly large and this state effect the system's speed and success negatively. In this study Data Set III of BCI competition 2003 was used. We extract features using this data set and then dimension of feature matrix size reduced by using Principal Component Analysis, Kernel Principal Component Analysis and Locality Preserving Projections method which is alternative to Principal Component Analysis. As a result, the best success rate is obtained as 83.28% when Linearity Preserving Projections algorithm with Chebycev distance measuring method is used.

Keywords: EEG; principal component analysis; kernel principal component analysis; lcality preserving projection

Başlık: EEG İşaretlerinde Boyut İndirgeme Algoritmalarının Karşılaştırılması

Özet: Bütün sınıflandırma algoritmalarında olduğu gibi, elektroensefalografi (EEG) uygulamalarının sınıflandırma başarısını arttıran en önemli işlem, işarete ait uygun özniteliklerin elde edilmesi aşamasıdır. Sınıflandırma uygulamalarında belirli bir öznitelik çıkarma metodu bulunmadığından dolayı, seçilen özniteliklerin oluşturduğu öznitelik matrisi büyük boyutlarda olabilmekte ve bu durum sistemin hızını ve başarısını olumsuz olarak etkilemektedir. Bu çalışmada, 'BCI Competition III' yarışmasında kullanılan III. veri seti kullanılmıştır. Bu veri seti için, Temel Bileşen Analizi, Çekirdek Temel Bileşen Analizi ve veri setindeki komşuluk yapısını koruyarak verinin projeksiyonunu sağlayan, Temel Bileşen Analizine alternatif yöntem olan Yerellik Koruyucu İzdüşüm algoritmaları kullanılarak öznitelik matrisi üzerinde boyut azaltma işlemi yapılmıştır. Sonuç olarak, Chebyshev mesafe ölçümü yöntemini kullanan Yerellik Koruyucu İzdüşüm algoritması ile en yüksek başarı % 83.28 olarak elde edilmiştir.

Anahtar kelimeler: EEG; Temel Bileşen Analizi; Çekirdek Temel Bileşen Analizi; Yerellik Koruyucu İzdüşüm algoritmasıEEG; temel bileşen analizi; çekirdek temel bileşen analizi; yerellik koruyucu izdüşüm algoritması

• Teplan M. Fundamentals of EEG measurement. Measurement Science Review 2002; 2(2).
• Saka K, Aydemir O, Ozturk M. Classification of EEG signals recorded during right/left hand movement imagery using Fast Walsh Hadamard Transform based features. In 39th International Conference on Telecommunications and Signal Processing (TSP), June 27-29, 2016, Vienna, Austria, pp. 413-416.
• Varela F, Lachaux J, Rodriguez E, Martinerie J. The BrainWeb: Phase synchronization and large-scale integration. Nature Reviews Neuroscience 2001; 2: 229-239.
• Mohamed AK. Towards improved EEG interpretation in a sensorimotor BCI for the control of a prosthetic or orthotic hand. MSc Thesis, University of Witwatersrand, Johannesburg, South Africa, 2011.
• Wang Y, Hong B, Gao X, Gao S. Implementation of a brain-computer interface based on three states of motor imagery. In 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, August 22-26, 2007, Lyon, France, pp. 5059-5062.
• Cheong LC, Sudirman R, Hussin SS. Feature extraction of EEG signal using wavelet transform for autism classification. ARPN Journal of Engineering and Applied Sciences 2015; 10(19): 8533-8540.
• Lin SC, Chang YCI, Yang WN. Meta-learning for imbalanced data and classification ensemble in binary classification. Neurocomputing 2009; 73(1-3): 484-494.
• Guerra EJ, Aquino VA, Gil PG. Epilepsy seizure detection in EEG signals using wavelet transforms and neural networks. In International Joint Conferences on Computer, Information Systems Sciences, & Engineering (CISSE), December 12-14, 2013.
• Xu R, Jiang N, Lin C, Mrachacz-Kersting N, Dremstrup K, Farina D. Enhanced low-latency detection of motor intention from EEG for closed-loop brain-computer interface applications. IEEE Transactions on Biomedical Engineering 2014; 61(2): 288-296.
• Zacharaki EI, Garganis K, Mporas I, Megalooikonomou V. Spike detection in EEG by LPP and SVM. In IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI), June 1-4, 2014, Valencia, Spain, pp. 668-671.
• Zheng WL, Zhang YQ, Zhu JY, Lu BL. Transfer components between subjects for EEG-based emotion recognition. In 2015 International Conference on Affective Computing and Intelligent Interaction (ACII), September 21-24, 2015, Xian, China, pp. 917-922.
• Kaczorowska M, Plechawska-Wojcik M, Tokovarov M, Dmytruk R. Comparison of the ICA and PCA methods in correction of EEG signal artefacts. In 2017 10th International Symposium on Advanced Topics in Electrical Engineering (ATEE), March 23-25, 2017, Bucharest, Romania, pp. 262-267.
• Lin SC, Chang YCI, Yang WN. Meta-learning for imbalanced data and classification ensemble in binary classification. Neurocomputing 2009; 73(1-3): 484-494.
• Smith LI. A Tutorial on Principal Components Analysis. 2002. Retrieved from http://www.cs.otago.ac.nz/cosc453/student_tutorials/principal_components.pdf
• He X, Niyogi P. Locality preserving projections. Advances in Neural Information Processing Systems 16 (NIPS 2003), 2003.
• Kiang MY. A comparative assessment of classification methods. Decision Support Systems 2003; 35: 441-454.