An Investigation of Preliminary Feature Screening Using Signal-To-Noise Ratios

Abstract

A new saliency metric and a new saliency screening method are developed. This new metric, the SN saliency metric, is based upon signal-to-noise ratios, where the signal is provided by a sum of squared weights associated with a given feature, and the noise is based upon a sum of squared weights associated with a reference noise feature which is injected into the data. The resultant metric allows for a direct comparison of the feature of interest with a reference noise feature which is known to be nonsalient. The SN saliency screening method, which uses the SN saliency metric, offers the potential of identifying salient features in one saliency screening run and is envisioned as an economical rough screening tool to be used prior to more refined screening efforts or more exhaustive training efforts. During the screening run, features are removed individually based upon their rank as determined by the SN saliency metric. The classification error rate's reaction to a given feature's removal helps confirm that feature's saliency.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 1996
Accession Number
ADA324149

Entities

People

  • David B. Sumrell

Organizations

  • Air Force Institute of Technology

Tags

DTIC Thesaurus Topics

  • Algorithms
  • Bias
  • Computer Science
  • Data Science
  • Data Sets
  • Demography
  • Equations
  • Experimental Design
  • Information Processing
  • Information Science
  • Information Systems
  • Machine Learning
  • Neural Networks
  • Operations Research
  • Statistical Tests
  • Statistics
  • Test Sets

Readers

  • Computer Vision.
  • Phased Array Antenna Design.
  • Systems Analysis and Design