An empirical investigation on the use of diversity for creation of classifier ensembles

AHMED, MUHAMMAD ATTA OTHMAN;DIDACI, LUCA;FUMERA, GIORGIO;ROLI, FABIO
2015-01-01

Abstract

We address one of the main open issues about the use of diversity in multiple classifier systems: the effectiveness of the explicit use of diversity measures for creation of classifier ensembles. So far, diver- sity measures have been mostly used for ensemble pruning, namely, for selecting a subset of classifiers out of an original, larger ensemble. Here we focus on pruning techniques based on forward/backward selection, since they allow a direct comparison with the simple estimation of accu- racy of classifier ensemble. We empirically carry out this comparison for several diversity measures and benchmark data sets, using bagging as the ensemble construction technique, and majority voting as the fusion rule. Our results provide further and more direct evidence to previous observations against the effectiveness of the use of diversity measures for ensemble pruning, but also show that, combined with ensemble accuracy estimated on a validation set, diversity can have a regularization effect when the validation set size is small.
2015
Inglese
Multiple Classifier Systems 12th International Workshop, MCS 2015, Günzburg, Germany, June 29 - July 1, 2015, Proceedings
978-3-319-20247-1
978-3-319-20248-8
Springer International
Friedhelm Schwenker, Fabio Roli, Josef Kittler
9132
206
219
14
12th International Workshop on Multiple Classifier Systems, MCS 2015
Contributo
Esperti anonimi
Jun 29, 2015 - Jul 1, 2015
Günzburg, Germany
internazionale
scientifica
Diversity; Ensemble pruning; Forward/backward selection; Ensemble construction
no
4 Contributo in Atti di Convegno (Proceeding)::4.1 Contributo in Atti di convegno
Ahmed, MUHAMMAD ATTA OTHMAN; Didaci, Luca; Fumera, Giorgio; Roli, Fabio
273
4
4.1 Contributo in Atti di convegno
reserved
info:eu-repo/semantics/conferencePaper
Files in This Item:
File Size Format  
An Empirical Investigation on the Use of Diversity for Creation of Classifier Ensembles.pdf

Solo gestori archivio

Type: versione editoriale
Size 216.76 kB
Format Adobe PDF
216.76 kB Adobe PDF & nbsp; View / Open   Request a copy

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Questionnaire and social

Share on:
Impostazioni cookie