Share Email Print

Proceedings Paper

Effects of finite sample size and correlated/noisy input features on neural network pattern classification
Author(s): David G. Brown; Alexander C. Schneider; Mary S. Pastel; Robert F. Wagner
Format Member Price Non-Member Price
PDF $14.40 $18.00
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

In many areas of practical interest, for example medical decision making problems, input data for training and testing neural networks are severely limited in number, are corrupted by noise, and may be highly correlated. In this study we examine these factors by investigating network performance on a simulated Gaussian data set with known first and second order statistics. Following the work of Wagner et al. for statistical (likelihood- ratio) classifiers, we study how the addition of noisy/correlated features affects the performance of neural network classifiers. Results are similar to that of the previous study, demonstrating that for small data sets, additional noisy/correlated features in fact degrade network performance. In addition, the use of sophisticated statistical techniques including the jackknife, Fukunaga-Hayes group jackknife, and bootstrap to estimate performance variation and remove small-sample bias are examined and found to offer significant advantages.

Paper Details

Date Published: 11 May 1994
PDF: 10 pages
Proc. SPIE 2167, Medical Imaging 1994: Image Processing, (11 May 1994); doi: 10.1117/12.175100
Show Author Affiliations
David G. Brown, Ctr. for Devices and Radiological Health/FDA (United States)
Alexander C. Schneider, Stanford Univ. (France)
Mary S. Pastel, Ctr. for Devices and Radiological Health/FDA (United States)
Robert F. Wagner, Ctr. for Devices and Radiological Health/FDA (United States)

Published in SPIE Proceedings Vol. 2167:
Medical Imaging 1994: Image Processing
Murray H. Loew, Editor(s)

© SPIE. Terms of Use
Back to Top