An Optimal Global Nearest Neighbor Metric

A quadratic metric d AO (X, Y) =[(X - Y) T A O (X - Y)] 1/2 is proposed which minimizes the mean-squared error between the nearest neighbor asymptotic risk and the finite sample risk. Under linearity assumptions, a heuristic argument is given which indicates that this metric produces lower mean-squa...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	IEEE transactions on pattern analysis and machine intelligence 1984-05, Vol.PAMI-6 (3), p.314-318
Hauptverfasser:	Fukunaga, Keinosuke, Flick, Thomas E.
Format:	Artikel
Sprache:	eng
Schlagworte:	Applied sciences Artificial intelligence Asymptotic risk Bayes risk Computer science control theory systems distance measure Euclidean distance Exact sciences and technology Extraterrestrial measurements finite sample size Laboratories Linearity Measurement standards nearest neighbor rule Nearest neighbor searches Neural networks Pattern recognition. Digital image processing. Computational geometry Size measurement Statistics Time measurement
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Schreiben Sie den ersten Kommentar!