An almost surely optimal combined classification rule

  • Authors:
  • Majid Mojirsheibani

  • Affiliations:
  • Carleton University, Ottawa, Ontario, Canada

  • Venue:
  • Journal of Multivariate Analysis
  • Year:
  • 2002

Quantified Score

Hi-index 0.00

Visualization

Abstract

We propose a data-based procedure for combining a number of individual classifiers in order to construct more effective classification rules. Under some regularity conditions, the resulting combined classifier turns out to be almost surely superior to each of the individual classifiers. Here, superiority means lower misclassification error rate.