Chapter 3. Adaptive Kernel Classifiers Via Matrix Decomposition Updating for Biological Data Analysis

  1. Yan-Qing Zhang2 and
  2. Jagath C. Rajapakse3
  1. Hyunsoo Kim and
  2. Haesun Park

Published Online: 21 APR 2008

DOI: 10.1002/9780470397428.ch3

Machine Learning in Bioinformatics

Machine Learning in Bioinformatics

How to Cite

Kim, H. and Park, H. (2008) Adaptive Kernel Classifiers Via Matrix Decomposition Updating for Biological Data Analysis, in Machine Learning in Bioinformatics (eds Y.-Q. Zhang and J. C. Rajapakse), John Wiley & Sons, Inc., Hoboken, NJ, USA. doi: 10.1002/9780470397428.ch3

Editor Information

  1. 2

    Georgia State University, Atlanta, Georgia

  2. 3

    School of Computer Engineering, and The Bioinformatics Research Center, Nanyang Technological University, Nanyang, Singapore

Author Information

  1. Georgia Institute of Technology, Atlanta, Georgia

Publication History

  1. Published Online: 21 APR 2008
  2. Published Print: 12 NOV 2008

Book Series:

  1. Bioinformatics: Computational Techniques and Engineering

Book Series Editors:

  1. Professor Yi Pan and
  2. Professor Albert Y. Zomaya

ISBN Information

Print ISBN: 9780470116623

Online ISBN: 9780470397428

SEARCH

Keywords:

  • adaptive kernel classifiers via matrix decomposition;
  • adaptive KDA/MSE based on complete orthogonal decompositions;
  • adaptive KDA and eigenvalue decomposition update or SVD update

Summary

This chapter contains sections titled:

  • Introduction

  • Kernel Discriminant Analysis Based on the Minimum Squared Error Formulation

  • Adaptive KDA/MSE Based on Complete Orthogonal Decompositions

  • Adaptive KDA/RMSE Based on Regularization and QR Decomposition

  • Efficient Leave-One-Out Cross-Validation for Kernel Discriminant Analysis by Downdating

  • Experimental Results

  • Summary

  • Acknowledgments

  • References