Tuning diversity in bagged ensembles.

Autor: Carney JG; Department of Computer Science, University of Dublin, Trinity College, Ireland. John.Carney@cs.tcd.ie, Cunningham P
Jazyk: angličtina
Zdroj: International journal of neural systems [Int J Neural Syst] 2000 Aug; Vol. 10 (4), pp. 267-79.
DOI: 10.1142/S0129065700000272
Abstrakt: In this paper, we investigate how the level of diversity amongst individual neural networks in a bagged ensemble can significantly influence overall ensemble generalization performance. We propose a new technique that tunes this diversity so that ensemble generalization performance is optimized and evaluate its performance on benchmark regression data-sets.
Databáze: MEDLINE