Examining the Effect of Dimension Reduction on EEG Signals by K-Nearest Neighbors Algorithm

Öz

Machine learning which a paradigm of methods that makes inferences from existing data using mathematical and statistical methods and is inferred to be unknown. The proposed method in this paper, supervised learning algorithm is applied to EEG (electroencephalography) data and classification algorithm performance is analyzed and results are examined in MATLAB. K-Nearest Neighbors Algorithm (k-NN) is used in this paper as algorithm. This classification was evaluated in two stages, with and without Principal Component Analysis (PCA). Dimension reduction is the process of reducing the size of dimension of the data. By reducing the size of the data set with PCA, it is expected to protect important data features. KNN has given results that can be regarded as prudent in terms of classification accuracy. The results of the present work showed that appropriate features combined with classifier can be done significant classification for different bioelectrical signal

Anahtar Kelimeler

• Denetimli öğrenme algoritması; En yakın komşu algoritması (kNN); Temel Bileşen Analizi; Boyut Azaltma; EEG

En Yakın Komşu Algoritması Kullanılarak EEG Sinyallerine Boyut Azaltmanın Etkilerinin İncelenmesi

Öz

Makine öğrenmesi, var olan verilerin çıkarımlarını matematiksel ve istatistiksel yöntemlerle yapan ve bilinmeyen bir yöntem paradigmasıdır. Bu çalışmada, denetimli öğrenme algoritması, EEG (elektroensefalografi) verilerine uygulanmış, sınıflandırma algoritması performans analiz sonuçları MATLAB ile incelenmiştir. Bu çalışmada, algoritma olarak en yakın komşu algoritması (k-NN) kullanılmıştır. Bu sınıflandırma, Temel Bileşen Analizinin (TBA) kullanıldığı ve kullanılmadığı durumlar için iki aşamada değerlendirilmiştir. Boyut azaltma, verilerin boyut boyutunu küçültme işlemidir. TBA ile veri kümesinin boyutunun azaltılarak, önemli veri özelliklerini korunması beklenir. KNN, sınıflandırma doğruluğu açısından önemli sayılabilecek sonuçlar vermiştir. Mevcut çalışma, farklı biyoelektriksel sinyaller için uygun özelliklerin uygun bir sınıflandırıcı ile kombine edildiğinde anlamlı bir sınıflandırma yapılabileceğini göstermiştir.

Anahtar Kelimeler

• Supervised learning algorithms; k-nearest neighbors algorithm (kNN); Principal Component Analysis (PCA); Dimension reduction; EEG

Kaynakça

1. [1] Shaker M. M., EEG Waves Classifier using Wavelet Transformand Fourier Transform, Int. Journal of Biological and Life Sciences, 1(3), (2005), 85-90.
2. [2] Bhattacharya J. and Petsche H. Universality in the brain while listening to music, Proc. Royal Society Lond. B., 268(1484), (2001), 2423-2433.
3. [3] C. J. Stam, J.P.M. Pijn, P. Suffczynski and F.H. Lopes da Silva. Dynamics of the human alpha rhythm: evidence for non-linearity?, Clinical Neurophysiology, 110(10), . (1999), 18011813.
4. [4] G. Buzsaki. Rhythms of the Brain Oxford University Press, (2006). Oxford.
5. [5] S. P. Kumar, N. Sriraam, P.G. Benakop and B.C.Jinaga. Entropies based detection of epileptic seizures with artificial neural network classifiers, Expert Syst. Appl., 37(4), (2010), 3284–3291
6. [6] L.D. Iasemidis, Epileptic seizure prediction and control, IEEE Trans Biomed Eng. 50(5), (2003), 549-558.
7. [7] Z. Haydari, Y. Zhang, and H.S. Zadeh. SemiAutomatic Epilepsy Spike Detection from EEG Signal Using Genetic Algorithm and Wavelet Transform, IEEE International Conference on Bioinformatics and Biomedicine Workshops, (2011). Atlanta, GA, USA.
8. [8] H. S. Park, Y. H. Lee, N. G. Kim, D.S. Lee and S. I. Kim. Detection of epileptic form activities in the EEG using neural network and expert system, Studies in health technology and informatics, 9(2), (1998), 1255–1259.

9. [9] Kaya D., Türk M., Biyoelektriksel Kökenli İşaretlerde Rahatsızlık Teşhisinin Yorumlanması, Fırat Üniversitesi Mühendislik Bilimleri Dergisi, Cilt 29, Sayı 1(2017).
10. [10] Kaya T. and Ince M.C., The Obtaining of Window Function Having Useful Spectral Parameters by Helping of Genetic Algorithm, Procedia-Social and Behavioral Journal, 83, (2012), 563-568, Elsevier.
11. [11] Makinac, M., Support Vector Machine Approach for Classification of Cancerous Prostate Regions. 2005. World Academy of Science, Engineering and Technology [12] Christopher J. C. Burges, A Tutorial on Support Vector Machines for Pattern Recognition, Data Mining and Knowledge Discovery, 2(2), June 1998, 121-167,
12. [13] Chandaka S., Chatterjee A., Munshi S., Cross-correlation aided support vector machine classifier for classification of EEG signals, Expert Systems with Applications: An International Journal, v.36 n.2, (March, 2009), 1329-1336.
13. [14] Silver, A.E., Lungren, M.P., Johnson, M.E., O'Driscoll, S.W., An, K.N. and Hughes, R.E., Using support vector machines to optimally classify rotator cuff strength data and quantify post-operative strength in rotator cuff tear patients. J. Biomech. v39. 973-979.
14. [15] J. Fan, C. Shao, Y. Ouyang, J. Wang , S. Li, Z. Wang, Automatic seizure detection based on support vector machines with genetic algorithms, Proceedings of the 6th international conference on Simulated Evolution And Learning, October 15-18, 2006, Hefei, China.
15. [16] Siuly, Y. Li, P. Wen, EEG signal classification based on simple random sampling technique with least square support vector machines, Int. J. Biomed. Eng. Technol (2010), in press.
16. [17] G. Chen and R. Hou, “A New Machine Double-Layer Learning Method and Its Application in Non-Linear Time Series Forecasting,” in International Conference on Mechatronics and Automation, ICMA, (2007), 795 –799.
17. [18] Labview machine learning toolkit user manual.
18. [19] C.M. Bishop Pattern Recognition and Machine Learning, Springer, New York (2006)
19. [20] Hotelling, H. Analysis of a complex of statistical variables into principal components. Journal of Educational Psychology, 24, (1933), 417–441, and 498–520.
20. [21] Hotelling, H. Relations between two sets of variates. Biometrika, 28, (1936), 321–377.
21. [22] Ö. Türk, M. S, Özerdem, “EEG İşaretlerinin k-NN ile Sınıflandırılmasında Dalgacıklara İlişkin Performansların Karşılaştırılması, TIPTEKNO-14, KAPADOKYA.
22. [23] Lee, S., Kang, P., Cho S., “Probabilistic local reconstruction for k-NN regression and its application to virtual metrology in semiductor manufacturing”, Neurocomputing-131, (2014), 427– 439,
23. [24] L.Tomak and Y Bek. İşlem Karakteristik Eğrisi Analizi ve Eğri Altında Kalan Alanların Karşılaştırılması, Journal of Experimental and Clinical Medicine, (2009), 27(2), 58-65.
24. [25]. EEG Data http://www.meb.uni- bonn.de/ epileptologie/science/ physik/eegdata.html