Saved in:
Bibliographic Details
Main Authors: Zhiyanov, Anton, Shklyaev, Alexander, Galatenko, Alexey, Galatenko, Vladimir, Tonevitsky, Alexander
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2501.14430
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913664523567104
author Zhiyanov, Anton
Shklyaev, Alexander
Galatenko, Alexey
Galatenko, Vladimir
Tonevitsky, Alexander
author_facet Zhiyanov, Anton
Shklyaev, Alexander
Galatenko, Alexey
Galatenko, Vladimir
Tonevitsky, Alexander
contents We propose a homogeneity test closely related to the concept of linear separability between two samples. Using the test one can answer the question whether a linear classifier is merely ``random'' or effectively captures differences between two classes. We focus on establishing upper bounds for the test's \emph{p}-value when applied to two-dimensional samples. Specifically, for normally distributed samples we experimentally demonstrate that the upper bound is highly accurate. Using this bound, we evaluate classifiers designed to detect ER-positive breast cancer recurrence based on gene pair expression. Our findings confirm significance of IGFBP6 and ELOVL5 genes in this process.
format Preprint
id arxiv_https___arxiv_org_abs_2501_14430
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Statistical Verification of Linear Classifiers
Zhiyanov, Anton
Shklyaev, Alexander
Galatenko, Alexey
Galatenko, Vladimir
Tonevitsky, Alexander
Machine Learning
Probability
Statistics Theory
Applications
62P10
G.3
We propose a homogeneity test closely related to the concept of linear separability between two samples. Using the test one can answer the question whether a linear classifier is merely ``random'' or effectively captures differences between two classes. We focus on establishing upper bounds for the test's \emph{p}-value when applied to two-dimensional samples. Specifically, for normally distributed samples we experimentally demonstrate that the upper bound is highly accurate. Using this bound, we evaluate classifiers designed to detect ER-positive breast cancer recurrence based on gene pair expression. Our findings confirm significance of IGFBP6 and ELOVL5 genes in this process.
title Statistical Verification of Linear Classifiers
topic Machine Learning
Probability
Statistics Theory
Applications
62P10
G.3
url https://arxiv.org/abs/2501.14430