Saved in:
Bibliographic Details
Main Authors: Essomba, Rose Yvette Bandolo, Fokoué, Ernest
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2601.04149
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918276706074624
author Essomba, Rose Yvette Bandolo
Fokoué, Ernest
author_facet Essomba, Rose Yvette Bandolo
Fokoué, Ernest
contents Class imbalance significantly degrades classification performance, yet its effects are rarely analyzed from a unified theoretical perspective. We propose a principled framework based on three fundamental scales: the imbalance coefficient $η$, the sample--dimension ratio $κ$, and the intrinsic separability $Δ$. Starting from the Gaussian Bayes classifier, we derive closed-form Bayes errors and show how imbalance shifts the discriminant boundary, yielding a deterioration slope that predicts four regimes: Normal, Mild, Extreme, and Catastrophic. Using a balanced high-dimensional genomic dataset, we vary only $η$ while keeping $κ$ and $Δ$ fixed. Across parametric and non-parametric models, empirical degradation closely follows theoretical predictions: minority Recall collapses once $\log(η)$ exceeds $Δ\sqrtκ$, Precision increases asymmetrically, and F1-score and PR-AUC decline in line with the predicted regimes. These results show that the triplet $(η,κ,Δ)$ provides a model-agnostic, geometrically grounded explanation of imbalance-induced deterioration.
format Preprint
id arxiv_https___arxiv_org_abs_2601_04149
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A Theoretical and Empirical Taxonomy of Imbalance in Binary Classification
Essomba, Rose Yvette Bandolo
Fokoué, Ernest
Machine Learning
62H10, 62H30
Class imbalance significantly degrades classification performance, yet its effects are rarely analyzed from a unified theoretical perspective. We propose a principled framework based on three fundamental scales: the imbalance coefficient $η$, the sample--dimension ratio $κ$, and the intrinsic separability $Δ$. Starting from the Gaussian Bayes classifier, we derive closed-form Bayes errors and show how imbalance shifts the discriminant boundary, yielding a deterioration slope that predicts four regimes: Normal, Mild, Extreme, and Catastrophic. Using a balanced high-dimensional genomic dataset, we vary only $η$ while keeping $κ$ and $Δ$ fixed. Across parametric and non-parametric models, empirical degradation closely follows theoretical predictions: minority Recall collapses once $\log(η)$ exceeds $Δ\sqrtκ$, Precision increases asymmetrically, and F1-score and PR-AUC decline in line with the predicted regimes. These results show that the triplet $(η,κ,Δ)$ provides a model-agnostic, geometrically grounded explanation of imbalance-induced deterioration.
title A Theoretical and Empirical Taxonomy of Imbalance in Binary Classification
topic Machine Learning
62H10, 62H30
url https://arxiv.org/abs/2601.04149