Campus Units
Industrial and Manufacturing Systems Engineering, Statistics
Document Type
Article
Publication Version
Accepted Manuscript
Publication Date
2013
Journal or Book Title
The American Statistician
Volume
67
Issue
2
First Page
94
Last Page
96
DOI
10.1080/00031305.2013.778788
Abstract
The technique of “majority voting” of classifiers is used in machine learning with the aim of constructing a new combined classification rule that has better characteristics than any of a given set of rules. The “Condorcet Jury Theorem” is often cited, incorrectly, as support for a claim that this practice leads to an improved classifier (i.e., one with smaller error probabilities) when the given classifiers are sufficiently good and are uncorrelated. We specifically address the case of two-category classification, and argue that a correct claim can be made for independent (not just uncorrelated) classification errors (not the classifiers themselves), and offer an example demonstrating that the common claim is false. Supplementary materials for this article are available online.
Copyright Owner
American Statistical Association
Copyright Date
2013
Language
en
File Format
application/pdf
Recommended Citation
Vardeman, Stephen B. and Morris, Max, "Majority Voting by Independent Classifiers Can Increase Error Rates" (2013). Industrial and Manufacturing Systems Engineering Publications. 140.
https://lib.dr.iastate.edu/imse_pubs/140
Included in
Industrial Engineering Commons, Statistics and Probability Commons, Systems Engineering Commons
Comments
This is an Accepted Manuscript of an article published by Taylor & Francis in The American Statistician on March 25, 2013 available online: http://www.tandfonline.com/10.1080/00031305.2013.778788