On the Sensitivity of the Probability of Error Rule for Feature Selection

Moshe Ben-Bassat*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

7 Scopus citations

Abstract

The low sensitivity of the probability of error rule (Pe rule) for feature selection is demonstrated and discussed. It is shown that under certain conditions features with significantly different discrimination power are considered as equivalent by the Pe rule. The main reason for this phenomenon lies in the fact that, directly, the Pe rule depends only on the most probable class and that, under the stated condition, the prior most probable class remains the posterior most probable class regardless of the result for the observed feature. A rule for breaking ties is suggested to refine the feature ordering induced by the Pe rule. By this tie-breaking rule, when two features have the same value for the expected probability of error, the feature with the higher variance for the probability of error is preferred.

Original languageEnglish
Pages (from-to)57-61
Number of pages5
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
VolumePAMI-2
Issue number1
DOIs
StatePublished - Jan 1980
Externally publishedYes

Keywords

  • Classification
  • feature selection
  • pattern recognition
  • probability of error
  • sensitivity of feature selection rules

Fingerprint

Dive into the research topics of 'On the Sensitivity of the Probability of Error Rule for Feature Selection'. Together they form a unique fingerprint.

Cite this