Kernel-Based Methods Kernel@Kernel-based method

Inspired by the success of support vector machines, to improve generalization and classification abilities, conventional pattern classification techniques have been extended to incorporate maximizing margins and mapping to a feature space. For example, perceptron algorithms [1–4], neural networks (C...

Full description

Saved in:
Bibliographic Details
Published inSupport Vector Machines for Pattern Classification pp. 305 - 329
Main Author Abe, Shigeo
Format Book Chapter
LanguageEnglish
Published London Springer London 22.01.2010
SeriesAdvances in Pattern Recognition
Subjects
Feature Space
Linear Discriminant Analysis
Mahalanobis Distance
Radial Basis Function Neural Network
Recognition Rate
Online AccessGet full text

Cover

More Information
Summary:Inspired by the success of support vector machines, to improve generalization and classification abilities, conventional pattern classification techniques have been extended to incorporate maximizing margins and mapping to a feature space. For example, perceptron algorithms [1–4], neural networks (Chapter 9), and fuzzy systems (Chapter 10) have incorporated maximizing margins and/or mapping to a feature space.
ISBN:9781849960977
1849960976
ISSN:2191-6586
DOI:10.1007/978-1-84996-098-4_6