Not Logged In

Explicitly Representing Expected Cost: An Alternative to ROC Representation

This paper proposes an alternative to ROC representation, in which the expected cost of a classifier is represented ex­ plicitly. This expected cost representation maintains many of the advantages of ROC representation, but is easier to understand. It allows the experimenter to immediately see the range of costs and class frequencies where a particu­ lar classifier is the best and quantitatively how much better it is than other classifiers. This paper demonstrates there is a point/line duality between the two representations. A point in ROC space representing a classifier becomes a line segment spanning the full range of costs and class frequen­ cies. This duality produces equivalent operations in the two spaces, allowing most techniques used in ROC analysis to be readily reproduced in the cost space.

Citation

C. Drummond, R. Holte. "Explicitly Representing Expected Cost: An Alternative to ROC Representation". Knowledge Discovery and Datamining, -, pp 187-207, January 2000.

Keywords: ROC, representation, machine learning
Category: In Conference

BibTeX

@incollection{Drummond+Holte:KDD00,
  author = {Chris Drummond and Robert Holte},
  title = {Explicitly Representing Expected Cost:  An Alternative to ROC
    Representation},
  Pages = {187-207},
  booktitle = {Knowledge Discovery and Datamining},
  year = 2000,
}

Last Updated: June 18, 2007
Submitted by Staurt H. Johnson

University of Alberta Logo AICML Logo