| | |
Summary: Arbitrating Among Competing Classi ers Using Learned
Referees
Julio Ortega
julio@us.ibm.com, Phone: 1-972-406-5946
IBM Global Business Intelligence Solutions, 1503 LBJ Freeway
Dallas, TX 75234, USA
Moshe Koppel
koppel@cs.biu.ac.il, Phone: 972-3-531-8407
Dept. of Mathematics and Computer Science, Bar-Ilan University
Ramat Gan 52900, Israel
Shlomo Argamon (corresponding author)
argamon@mail.jct.ac.il, Phone: 972-2-675-1294
Dept. of Computer Science, Jerusalem College of Technology, Machon Lev
21 Havaad Haleumi St.
Jerusalem 91160,Israel
Abstract
The situation in which the results of several di erent classi ers and learning algorithms
are obtainable for a single classi cation problem is common. In this paper, we propose a
method that takes a collection of existing classi ers and learning algorithms, together with
a set of available data, and creates a combined classi er that takes advantage of all of these
|
