Binary Split Categorical Feature with Mean Absolute Error Criteria in CART
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (33) · 27961-27968
DOI: 10.1609/aaai.v40i33.40020 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Peng Yu und 4 weiteren Autor:innen.
Abstract (Originalsprache)
In the context of the Classification and Regression Trees (CART) algorithm, the efficient splitting of categorical features using standard criteria like GINI and Entropy is well-established. However, using the Mean Absolute Error (MAE) criterion for categorical features has traditionally relied on various numerical encoding methods. This paper demonstrates that unsupervised numerical encoding methods are not viable for MAE criteria. Furthermore, we present a novel and efficient splitting algorithm that addresses the challenges of handling categorical features with the MAE criterion. Our findings underscore the limitations of existing approaches and offer a promising solution to enhance the handling of categorical data in CART algorithms.
Zitieren
APA
Yu, P., Chen, Y., Xu, C., Bifet, A., & Read, J. (2026). Binary Split Categorical Feature with Mean Absolute Error Criteria in CART. Proceedings of the AAAI Conference on Artificial Intelligence, 40(33), 27961-27968. https://doi.org/10.1609/aaai.v40i33.40020
Vancouver
Yu P, Chen Y, Xu C, Bifet A, Read J. Binary Split Categorical Feature with Mean Absolute Error Criteria in CART. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(33):27961-27968. doi:10.1609/aaai.v40i33.40020
BibTeX
@article{Yu2026,
author = {Yu, Peng and Chen, Yike and Xu, Chao and Bifet, Albert and Read, Jesse},
title = {Binary Split Categorical Feature with Mean Absolute Error Criteria in CART},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {33},
pages = {27961--27968},
doi = {10.1609/aaai.v40i33.40020}
}
EUVIMED