ON THE POSTERIOR-PROBABILITY ESTIMATE OF THE ERROR RATE OF NONPARAMETRIC CLASSIFICATION RULES

Authors
LUGOSI G PAWLAK M
Citation
G. Lugosi et M. Pawlak, ON THE POSTERIOR-PROBABILITY ESTIMATE OF THE ERROR RATE OF NONPARAMETRIC CLASSIFICATION RULES, IEEE transactions on information theory, 40(2), 1994, pp. 475-481
Citations number
33
Categorie Soggetti
Information Science & Library Science","Engineering, Eletrical & Electronic
Journal title
IEEE transactions on information theory → ACNP
ISSN journal
00189448
Volume
40
Issue
2
Year of publication
1994
Pages
475 - 481
Database
ISI
SICI code
0018-9448(1994)40:2<475:OTPEOT>2.0.ZU;2-E
Abstract
The posterior-probability estimate of the classification error rate of some nonparametric classification rules is studied. The variance of t he estimator is shown to have some remarkable distribution-free proper ties for the k-nearest neighbor, kernel, and histogram rules. We also investigate the bias of the estimate and establish its consistency and upper bounds. The version of the estimate calculated from an independ ent set of unclassified patterns is also considered.