Sciweavers

ROCAI
2004
Springer
13 years 11 months ago
What ROC Curves Can't Do (and Cost Curves Can)
Abstract. This paper shows that ROC curves, as a method of visualizing classifier performance, are inadequate for the needs of Artificial Intelligence researchers in several sign...
Chris Drummond, Robert C. Holte
ICPR
2008
IEEE
14 years 5 days ago
A class-selective rejection scheme based on blockwise similarity of typicality degrees
Overlapping classes and outliers can significantly decrease a classifier performance. We adress here the problem of giving a classifier the ability to reject some patterns eith...
Hoel Le Capitaine, Carl Frélicot