Rank and response combination from confusion matrix data

The use of prior behavior of a classifier, as measured by the confusion matrix, can yield useful information for merging multiple classifiers. In particular, response vectors can be estimated and a ranking of possible classes can be produced which can allow Borda type reconciliation methods to be ap...

Full description

Saved in:
Bibliographic Details
Published inInformation fusion Vol. 2; no. 2; pp. 113 - 120
Main Author Parker, J.R.
Format Journal Article
LanguageEnglish
Published Elsevier B.V 01.06.2001
Subjects
Classifier combination
Data fusion
Data fusion
Classifier combination
Online AccessGet full text

Cover

More Information
Summary:The use of prior behavior of a classifier, as measured by the confusion matrix, can yield useful information for merging multiple classifiers. In particular, response vectors can be estimated and a ranking of possible classes can be produced which can allow Borda type reconciliation methods to be applied. A combination of real data and the simulation of multiple classifiers is used to evaluate this idea, and to compare with 11 other classifier combination techniques. Millions of classifications were used in the evaluation.
ISSN:1566-2535
1872-6305
DOI:10.1016/S1566-2535(01)00030-6