When Are Linear Classifiers Sufficient? A Re-evaluation and Conditional Boundary Exploration Based on OpenML
Frontiers in Computing and Intelligent Systems · 03.11.2026 · 18 (2) · 103-107
DOI: 10.54097/xssbv193 ↗Kurz gefasst
Fachartikel in der Zeitschrift Frontiers in Computing and Intelligent Systems, online erschienen am 03.11.2026. Verfasst von Lujuan Zhang. Die Arbeit nennt 17 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
The point at which a linear classifier stops being "sufficient" lacks a quantitative definition. To address this, we evaluated 1,687 records from 298 OpenML datasets across three model families. Using a practical significance framework, we defined linear sufficiency as an accuracy drop within a tolerance threshold δ. At δ=0.02, linear sufficiency was observed in 68.9% of SVM datasets, 62.4% of NN datasets, but only 46.1% of DT datasets. Shallow decision tree extraction highlighted low feature-to-instance ratios and small class counts as primary predictors. However, leave-one-out validation exposed a clear split. The extracted rules worked well for the DT family (0.686 accuracy vs. 0.539 baseline) but failed to surpass baselines for SVM and NN. This suggests that while simple meta-features can predict linear sufficiency for decision trees, they cannot fully capture the complex boundaries of SVMs and neural networks.
Zitieren
APA
Zhang, L. (2026). When Are Linear Classifiers Sufficient? A Re-evaluation and Conditional Boundary Exploration Based on OpenML. Frontiers in Computing and Intelligent Systems, 18(2), 103-107. https://doi.org/10.54097/xssbv193
Vancouver
Zhang L. When Are Linear Classifiers Sufficient? A Re-evaluation and Conditional Boundary Exploration Based on OpenML. Frontiers in Computing and Intelligent Systems. 2026;18(2):103-107. doi:10.54097/xssbv193
BibTeX
@article{Zhang2026,
author = {Zhang, Lujuan},
title = {When Are Linear Classifiers Sufficient? A Re-evaluation and Conditional Boundary Exploration Based on OpenML},
journal = {Frontiers in Computing and Intelligent Systems},
year = {2026},
volume = {18},
number = {2},
pages = {103--107},
doi = {10.54097/xssbv193}
}
EUVIMED