Classifier combinations: Implementation and theoretical issues

Suk Wah Louisa LAM

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Much work has been done in the past decade to combine decisions of multiple classifiers in order to obtain improved recognition results. Many methodologies have been designed and implemented for this purpose. This article considers some of the current developments according to the structure of the combination process, and discusses some issues involved in each structure. In addition, theoretical investigations that have been performed in this area are also examined, and some related issues are discussed. Copyright © 2000 Springer-Verlag Berlin Heidelberg.
Original languageEnglish
Title of host publicationMultiple Classifier Systems: The 1st International Workshop, MCS 2000
EditorsJosef KITTLER , Fabio ROLI
Place of PublicationBerlin
PublisherSpringer
Pages77-86
ISBN (Print)3540677046, 9783540677048, 9783540450146
DOIs
Publication statusPublished - 2000

Fingerprint

Classifiers

Citation

Lam, S. W. L. (2000). Classifier combinations: Implementation and theoretical issues. In J. Kittler & F. Roli (Eds.), Multiple Classifier Systems: The 1st International Workshop, MCS 2000 (pp. 77-86). Berlin: Springer.