Vol 15, No 1 (2011) > Articles >

Two-Class Classification with Various Characteristics Based on Kernel Principal Component Analysis and Support Vector Machines

Ivanna Timotius 1 , Iwan Setyawan 1 , Andreas Febrianto 1


  1. Department of Electronic Engineering, Satya Wacana Christian University, Salatiga 50711, Indonesia


Abstract: Two class pattern classification problems appeared in many applications. In some applications, the characteristic of the members in a class is dissimilar. This paper proposed a classification system for this problem. The proposed system was developed based on the combination of kernel principal component analysis (KPCA) and support vector machines (SVMs). This system has been implemented in a two class face recognition problem. The average of the classification rate in this face image classification is 82.5%.
Keywords: characteristic, classification, face recognition, kernel principal component analysis, support vector machines
Published at: Vol 15, No 1 (2011) pages: 96-100

Access Counter: 1059 views, 572 PDF downloads, .

Full PDF Download


R.O. Duda, P.E. Hart, D.G. Stork, Pattern Classification, 2nd ed., Wiley, New York, 2001, p.635.

Timotius, I. K., Setyawan, I. & Febrianto, A. A., Int. J. Electr. Eng. Inf. 2/1 (2010) 53.

E. Makinen, R. Raisamo, IEEE Trans. Pattern Anal. Mach. Intell. 30/3 (2008) 541.

Z. Li, X. Tang, IEEE Trans. Inf. Forensics Secur. 2/2 (2007) 174.

I. Kotsia, I. Pitas, IEEE Trans. Image Process. 16/1 (2007) 172.

M. Gonen, A.G. Tanugur, E. Alpaydın, IEEE Trans. Neural Networks, 19/1 (2008) 130.

Z. Rustam, B. Widjaja, B. Kusumoputro, J. Makara Seri Sains 7/3 (2003) 15.

A.K. Jain, R.P. W. Duin, J. Mao, IEEE Trans. Pattern Anal. Mach. Intell. 22/1 (2000) 4.

K.R. Muller, S. Mika, G. Ratsch, K. Tsuda, B. Scholkopf, IEEE Trans. Neural Networks. 12/2 (2001) 181.

S. Haykin, Neural Network: A Comprehensive Foundation, Prentice-Hall, New Jersey, USA, 2008, p.936.

V.N. Vapnik, Statistical Learning Theory, Wiley, New York, 1998, p.736