Home Browse by Subject Bestsellers New Titles Editor's Choice New Reviews Textbooks
Search Book Series Study Guides Rights Inspection Copy Contact Us Join Our Mailing List
For Authors How to Order E-Catalogues

Browse all Subjects
Search Bookshop
New Titles
Editor's Choice
Bestsellers
Book Series
Textbooks
Journals
Join Our Mailing List
 
Series in Machine Perception and Artificial Intelligence - Vol. 68

BRIDGING THE GAP BETWEEN GRAPH EDIT DISTANCE AND KERNEL MACHINES

by Michel Neuhaus & Horst Bunke (University of Bern, Switzerland)

Table of Contents (50k)
Preface (49k)
Chapter 1: Introduction (62k)

In graph-based structural pattern recognition, the idea is to transform patterns into graphs and perform the analysis and recognition of patterns in the graph domain — commonly referred to as graph matching. A large number of methods for graph matching have been proposed. Graph edit distance, for instance, defines the dissimilarity of two graphs by the amount of distortion that is needed to transform one graph into the other and is considered one of the most flexible methods for error-tolerant graph matching.

This book focuses on graph kernel functions that are highly tolerant towards structural errors. The basic idea is to incorporate concepts from graph edit distance into kernel functions, thus combining the flexibility of edit distance-based graph matching with the power of kernel machines for pattern recognition. The authors introduce a collection of novel graph kernels related to edit distance, including diffusion kernels, convolution kernels, and random walk kernels. From an experimental evaluation of a semi-artificial line drawing data set and four real-world data sets consisting of pictures, microscopic images, fingerprints, and molecules, the authors demonstrate that some of the kernel functions in conjunction with support vector machines significantly outperform traditional edit distance-based nearest-neighbor classifiers, both in terms of classification accuracy and running time.


Contents:

  • Introduction
  • Graph Matching
  • Graph Edit Distance
  • Kernel Machines
  • Graph Kernels
  • Experimental Results


Readership: Researchers in structural pattern recognition and machine learning, both from industry and with an academic background. It is also suitable for undergraduate or graduate students with a particular interest in applying kernel machines to structural pattern recognition problems.

244pp Pub. date: Sept 2007
ISBN 978-981-270-817-5
981-270-817-0
US$69 / £37


Copyright © 2008 World Scientific Publishing Co. All rights reserved.
Updated on 25 July 2008