In defense of one-vs-all classification

DMCA

In defense of one-vs-all classification (2004)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Ryan Rifkin , Aldebaro Klautau

Venue:	Journal of Machine Learning Research
Citations:	307 - 0 self

BibTeX

@ARTICLE{Rifkin04indefense,
    author = {Ryan Rifkin and Aldebaro Klautau},
    title = {In defense of one-vs-all classification},
    journal = {Journal of Machine Learning Research},
    year = {2004},
    volume = {5},
    pages = {101--141}
}

OpenURL

Abstract

Editor: John Shawe-Taylor We consider the problem of multiclass classification. Our main thesis is that a simple “one-vs-all ” scheme is as accurate as any other approach, assuming that the underlying binary classifiers are well-tuned regularized classifiers such as support vector machines. This thesis is interesting in that it disagrees with a large body of recent published work on multiclass classification. We support our position by means of a critical review of the existing literature, a substantial collection of carefully controlled experimental work, and theoretical arguments.

Keyphrases

one-vs-all classification multiclass classification critical review main thesis underlying binary classifier large body john shawe-taylor support vector machine simple one-vs-all scheme substantial collection well-tuned regularized classifier experimental work theoretical argument

DMCA