Full text available in Adobe Acrobat format
Volume 51 Number 1 pp. 57-73
Karol Deręgowski 1
1President Stanislaw Wojciechowski Higher Vocational State School in Kalisz, Institute of Management, Nowy Świat 4, 62-800Kalisz, Poland
2Adam Mickiewicz University, Faculty of Mathematics and Computer Science, Umultowska 87, 61-614Poznań, Poland
2Adam Mickiewicz University, Faculty of Mathematics and Computer Science, Umultowska 87, 61-614Poznań, Poland
A kernel-based learning algorithm combining kernel discriminant coordinates and kernel principal components
SUMMARY
Kernel principal components (KPC) and kernel discriminant coordinates (KDC), which are the extensions of principal components and discriminant coordinates, respectively, from a linear domain to a nonlinear domain via the kernel trick, are two very popular nonlinear feature extraction methods. The kernel discriminant coordinates space has proven to be a very powerful space for pattern recognition. However, further study shows that there are still drawbacks in this method. To improve the performance of pattern recognition, we propose a new learning algorithm combining the advantages of KPC and KDC
Keywords: kernel principal components, kernel discriminant coordinates
DOI: 10.2478/bile-2014-0005
For citation:
| MLA | Deręgowski, Karol. "A kernel-based learning algorithm combining kernel discriminant coordinates and kernel principal components." Biometrical Letters 51.1 (2014): 57-73. DOI: 10.2478/bile-2014-0005 |
| APA | Deręgowski, K. (2014). A kernel-based learning algorithm combining kernel discriminant coordinates and kernel principal components. Biometrical Letters 51(1), 57-73 DOI: 10.2478/bile-2014-0005 |
| ISO 690 | DERęGOWSKI, Karol. A kernel-based learning algorithm combining kernel discriminant coordinates and kernel principal components. Biometrical Letters, 2014, 51.1: 57-73. DOI: 10.2478/bile-2014-0005 |