A Unified Strategy to Deal with Different Natures of Reject

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 reje...

Full description

Saved in:
Bibliographic Details
Main Authors: Mouchere, H., Anquetil, E.
Format: Conference Proceeding
Language:English
Subjects:
Constraint optimization
Fuzzy neural networks
Fuzzy systems
Histograms
Particle swarm optimization
Pattern recognition
Prototypes
Radial basis function networks
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary: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
ISSN:1051-4651
2831-7475
DOI:10.1109/ICPR.2006.193