An upper bound on the asymptotic error probability on the k-nearest neighbor rule for multiple classes (Corresp.)

Author

Gyorfi, Laszlo ; Gyorfi, Zoltan

Volume

Issue

fYear

1978

fDate

7/1/1978 12:00:00 AM

Firstpage

512

Lastpage

514

Abstract

If $R_{k}$ , denotes the asymptotic error probability of the $k$ - nearest neighbor rule for $M$ classes and $R\\ast$ denotes the Bayes probability of error, then conditions are given that yield $R_{k} - R\\ast \\leq \\sqrt {MR{1}/k}$ .

Keywords

Pattern classification; Bayesian methods; Convergence; Error probability; Information theory; Nearest neighbor searches; Pattern classification; Random variables; Upper bound; Virtual manufacturing;

fLanguage

English

Journal_Title

Information Theory, IEEE Transactions on

Publisher

ieee

ISSN

0018-9448

Type

jour

DOI

10.1109/TIT.1978.1055900

Filename

1055900

Link To Document

https://search.isc.ac/dl/search/defaultta.aspx?DTC=49&DC=928684