Scribd
Upload
22 views

Applications of Kernel Methods: Gustavo Camps-Valls Manel Martínez-Ramón José Luis Rojo-Álvarez

Kernel Methods

Uploaded by

Srinivas
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
22 views

Applications of Kernel Methods: Gustavo Camps-Valls Manel Martínez-Ramón José Luis Rojo-Álvarez

Kernel Methods

Uploaded by

Srinivas
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Section: Kernal Methods 

Applications of Kernel Methods A


Gustavo Camps-Valls
Universitat de València, Spain

Manel Martínez-Ramón
Universidad Carlos III de Madrid, Spain

José Luis Rojo-Álvarez


Universidad Rey Juan Carlos, Spain

INTRODUCTION ics (Mukherjee et al., 1998), the applications of these


methods have grown exponentially, and many novel
In this chapter, we give a survey of applications of the and powerful methods have been developed (only in
kernel methods introduced in the previous chapter. We 2004, more than 1000 papers have been devoted to this
focus on different application domains that are particu- topic). The use of kernel methods in computational
larly active in both direct application of well-known biology has been accompanied by new developments
kernel methods, and in new algorithmic developments to match the specificities and the needs of the field,
suited to a particular problem. In particular, we consider such as methods for feature selection in combination
the following application fields: biomedical engineer- with the classification of high-dimensional data, the
ing (comprising both biological signal processing and introduction of string kernels to process biological
bioinformatics), communications, signal, speech and sequences, or the development of methods to learn
image processing. from several kernels simultaneously (‘composite ker-
nels’). The interested reader can find a comprehensive
introduction in (Vert, 2006).
KERNEL METHODS IN BIOMEDICINE
AND BIOINFORMATICS
KERNEL METHODS IN
Kernel methods have been extensively used to solve COMMUNICATIONS
biomedical problems. For instance, a study of predic-
tion of cyclosporine dosage in patients after kidney There are four situations that make kernel methods
transplantation using neural networks and kernel-based good candidates for use in electromagnetics (Martínez-
methods was carried out in (Camps-Valls et al., 2002). Ramon, 2006): 1) No close solutions exist, and the
Recently, (Osowski, Hoai, & Markiewicz, 2004) pro- only approaches are trial and error methods. In these
posed a committee of experts formed by several sup- cases, kernel algorithms can be employed to solve the
port vector machines (SVM) for the recognition of 13 problem. 2) The application requires operating in real
heart rhythm types. time, and the computation time is limited. In these cases,
The most impressive results of kernels have been a kernel algorithm can be trained off-line, and used in
obtained in genomics and computational biology, due test mode in real time. The algorithms can be embedded
to both the special characteristics of data and the great in any hardware device. 3) Faster convergence rates
interest in solving biological problems since the Human and smaller errors are required. Kernel algorithms have
genome sequencing. Their ability to work with high shown superior performance in generalization ability
dimensional data, to process and efficiently integrate in many problems. Also, the block optimization and
non-vectorial string data, make them very suitable to the uniqueness of solutions make kernelized versions
solve various problems arising in computational biol- of linear algorithms (as SVM) faster than many other
ogy. Since the early papers using SVM in bioinformat- methods. 4) Enough measured data exist to train a

Copyright © 2009, IGI Global, distributing in print or electronic forms without written permission of IGI Global is prohibited.

You might also like