Partition Tree Ensembles for Improving Multi-Class Classification

Miran Özdogan, Alan Jeffares, Sean B. Holden

2025

Abstract

We propose the Partition Tree Ensemble (PTE), a novel tree-based ensemble method for classification problems. This differs from previous approaches in that it combines ideas from reduction methods—that decompose multi-class classification problems into binary classification problems—with the creation of specialised base learners that are trained on a subset of the input space. By exploiting multi-class reduction, PTEs adapt concepts from the Trees of Predictors (ToP) method to successfully tackle multi-class classification problems. Each inner node of a PTE splits either the feature space or the label space into subproblems. For each node our method then selects the most appropriate base learner from the provided set of learning algorithms. One of its key advantages is the ability to optimise arbitrary loss functions. Through an extensive experimental evaluation, we demonstrate that our approach achieves significant performance gains over the baseline ToP and AdaBoost methods, across various datasets and loss functions, and outperforms the Random Forest method when the label space exhibits clusters where some classes are more similar to each other than to others.

Paper Citation

in Harvard Style

Özdogan M., Jeffares A. and Holden S. (2025). Partition Tree Ensembles for Improving Multi-Class Classification. In Proceedings of the 14th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM; ISBN 978-989-758-730-6, SciTePress, pages 55-69. DOI: 10.5220/0013163400003905

in Bibtex Style

@conference{icpram25,
author={Miran Özdogan and Alan Jeffares and Sean Holden},
title={Partition Tree Ensembles for Improving Multi-Class Classification},
booktitle={Proceedings of the 14th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM},
year={2025},
pages={55-69},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013163400003905},
isbn={978-989-758-730-6},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 14th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM
TI - Partition Tree Ensembles for Improving Multi-Class Classification
SN - 978-989-758-730-6
AU - Özdogan M.
AU - Jeffares A.
AU - Holden S.
PY - 2025
SP - 55
EP - 69
DO - 10.5220/0013163400003905
PB - SciTePress