The mean subjective utility score, a novel metric for cost-sensitive classifier evaluation
The mean subjective utility score, a novel metric for cost-sensitive classifier evaluation
A simple solution is proposed to the problem of conveying a classifier's performance to a universal standard when misclassification costs are not equal. An application of this metric in the field of boosting is briefly described.
cost-sensitivity, cost matrix, utility, decision theory, boosting
1472-1477
McDonald, Ross A.
8aa626fc-1a94-4eb5-a644-849d51514136
1 October 2006
McDonald, Ross A.
8aa626fc-1a94-4eb5-a644-849d51514136
McDonald, Ross A.
(2006)
The mean subjective utility score, a novel metric for cost-sensitive classifier evaluation.
Pattern Recognition Letters, 27 (13), .
(doi:10.1016/j.patrec.2006.02.012).
Abstract
A simple solution is proposed to the problem of conveying a classifier's performance to a universal standard when misclassification costs are not equal. An application of this metric in the field of boosting is briefly described.
This record has no associated files available for download.
More information
Submitted date: 22 August 2005
Published date: 1 October 2006
Keywords:
cost-sensitivity, cost matrix, utility, decision theory, boosting
Identifiers
Local EPrints ID: 58109
URI: http://eprints.soton.ac.uk/id/eprint/58109
PURE UUID: 866ff4cd-9d0e-410c-9c02-199b9a2c6a4f
Catalogue record
Date deposited: 12 Aug 2008
Last modified: 15 Mar 2024 11:10
Export record
Altmetrics
Contributors
Author:
Ross A. McDonald
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics