Share Email Print

Optical Engineering

Nearest unlike neighbor (NUN): an aid to decision confidence estimation
Author(s): Belur V. Dasarathy
Format Member Price Non-Member Price
PDF $20.00 $25.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

The concept of nearest unlike neighbor (NUN), proposed and explored previously in the design of nearest neighbor (NN) based decision systems, is further exploited in this study to develop a measure of confidence in the decisions made by NN-based decision systems. This measure of confidence, on the basis of comparison with a user-defined threshold, may be used to determine the acceptability of the decision provided by the NN-based decision system. The concepts, associated methodology, and some illustrative numerical examples using the now classical Iris data to bring out the ease of implementation and effectiveness of the proposed innovations are presented.

Paper Details

Date Published: 1 September 1995
PDF: 8 pages
Opt. Eng. 34(9) doi: 10.1117/12.210755
Published in: Optical Engineering Volume 34, Issue 9
Show Author Affiliations
Belur V. Dasarathy, Dynetics, Inc. (United States)

© SPIE. Terms of Use
Back to Top