Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Search
 
loader loading subjects...
cover
Post-processing a classifier's predictions: Strategies and empirical evaluation
Salem Benferhat, Karim Tabia, Mouaad Kezih, Mahmoud Taibi
965 - 966
10.3233/978-1-61499-419-0-965
Frontiers in Artificial Intelligence and Applications
Volume 263: ECAI 2014
Abstract

In this paper, we propose an approach allowing to revise the outputs of a classifier in order to take into account the available domain knowledge. This approach can be applied for any classifier be it probabilistic or not. We propose post-processing criteria and methods to encode and exploit different kinds of domain knowledge. Finally, we provide experimental studies on a set of benchmarks.

Download
Creative Commons License