An Upper Bound on the Asymptotic Error Probability of the k-Nearest Neighbor Rule for Multiple Classes

László Györfi, Zoltán Györfi

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

If Rk denotes the asymptotic error probability of the k-nearest neighbor rule for m classes and R* denotes the Bayes probability of error, then conditions are given that yield Rk − R* ≼ √MR1/k.

Original languageEnglish
Pages (from-to)512-514
Number of pages3
JournalIEEE Transactions on Information Theory
Volume24
Issue number4
DOIs
Publication statusPublished - Jul 1978

ASJC Scopus subject areas

  • Information Systems
  • Computer Science Applications
  • Library and Information Sciences

Cite this