Academic Press Library in Signal Processing

Chapter 16

Kernel Methods and Support Vector Machines

John Shawe-Taylor* and Shiliang Sun^†,    ^*Department of Computer Science, University College London, Gower Street, London, United Kingdom,    ^†Department of Computer Science and Technology, East China Normal University, 500 Dongchuan Road, Shanghai, China, [e-mail]

Abstract

As the new generation of data analysis methods, kernels methods of which support vector machines are the most influential are extensively studied both in theory and in practice. This article provides a tutorial introduction to the foundations and implementations of kernel methods, well-established kernel methods, computational issues of kernel methods, and recent developments in this field. The aim of this article ...