Further consideration of sample and feature size (Corresp.)

Author

Young, Ian A.

Volume

Issue

fYear

1978

fDate

11/1/1978 12:00:00 AM

Firstpage

773

Lastpage

775

Abstract

It is shown that in the context of a specific pattern classification decision metric the number of samples $M$ needed to characterize a cluster described by $N$ features is $M \\geq (1 + \\beta ^{-1})(N + 2)$ where $\\beta$ represents an interval width. The distance metric $d^{2}(X)=(X-\\hat{\\mu}_{x})^{t}S_{x}^{-l}(X-\\hat{\\mu}_{x})$ is shown to have an $F$ -distribution which leads to the result for $M$ . An additional application of the distribution of $d^{2}(X)$ is discussed in terms of a specific type of pattern classifier.

Keywords

Pattern classification; Adaptive arrays; Adaptive systems; Equations; Gaussian processes; Guidelines; Pattern recognition; Random processes; Signal to noise ratio; State estimation; Vectors;

fLanguage

English

Journal_Title

Information Theory, IEEE Transactions on

Publisher

ieee

ISSN

0018-9448

Type

jour

DOI

10.1109/TIT.1978.1055953

Filename

1055953

Link To Document

https://search.isc.ac/dl/search/defaultta.aspx?DTC=49&DC=929222