Beyond global and local multi-target learning

Marcio Basgalupp, Ricardo Cerri, Leander Schietgat, Isaac Triguero, Celine Vens

Onderzoeksoutput: Articlepeer review

4 Citaten (Scopus)
38 Downloads (Pure)


In multi-target prediction, an instance has to be classified along multiple target variables at the same time, where each target represents a category or numerical value. There are several strategies to tackle multi-target prediction problems: the local strategy learns a separate model for each target variable independently, while the global strategy learns a single model for all target variables together. Previous studies suggested that the global strategy should be preferred because (1) learning is more efficient, (2) the learned models are more compact, and (3) it overfits much less than the local strategy, as it is harder to overfit on several targets at the same time than on one target. However, it is not clear whether the global strategy exploits correlations between the targets optimally. In this paper, we investigate whether better results can be obtained by learning multiple multi-target models on several partitions of the targets. To answer this question, we first determined alternative partitions using an exhaustive search strategy and a strategy based on a genetic algorithm, and then compared the results of the global and local strategies against these. We used decision trees and random forests as base models. The results show that it is possible to outperform global and local approaches, but finding a good partition without incurring in overfitting remains a challenging task.
Originele taal-2English
Pagina's (van-tot)508-524
Aantal pagina's17
TijdschriftInformation Sciences
StatusPublished - nov 2021

Bibliografische nota

Funding Information:
The authors would like to thank CNPq, CAPES and Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), grant 2016/02870–0, for funding this research. They are also grateful for the support from KU Leuven through its Latin America Fund and from the Flemish Government through its AI Research Program.

Publisher Copyright:
© 2021

Copyright 2021 Elsevier B.V., All rights reserved.


Duik in de onderzoeksthema's van 'Beyond global and local multi-target learning'. Samen vormen ze een unieke vingerafdruk.

Citeer dit