A Unified Strategy to Deal with Different Natures of Reject

Authors:
Harold Mouchere;Eric Anquetil
Affiliations:
IRISA / 1,2INSA / 1CNRS Campus Universitaire de Beaulieu, France;IRISA / 1,2INSA / 1CNRS Campus Universitaire de Beaulieu, France
Venue:
ICPR '06 Proceedings of the 18th International Conference on Pattern Recognition - Volume 02
Year:
2006

Citing 0
Cited 3

Eager interpretation of on-line hand-drawn structured documents: The DALI methodology

Pattern Recognition
Pattern Recognition Strategies for Interactive Sketch Composition

Proceedings of the 13th International Conference on Human-Computer Interaction. Part II: Novel Interaction Methods and Techniques
A family of measures for best top-n class-selective decision rules

Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

The interest of reject for classifier optimization has been shown many times. The diversity of the applications requiring this concept makes us to distinguish two main natures of reject with distinct goals: the confusion reject and the distance reject. After the description of this two kinds of reject, we present a unified formalism to define them using reliability functions and reject thresholds. Then we present a generic algorithm dedicated to the automatic learning of these thresholds. Finally, we compare various possibilities of reject to achieve application goals.