Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Yang, Jing, Wang, Ruibo, Song, Yijun, Li, Jihong
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2304.03990
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916551272169472
author Yang, Jing
Wang, Ruibo
Song, Yijun
Li, Jihong
author_facet Yang, Jing
Wang, Ruibo
Song, Yijun
Li, Jihong
contents In the task of comparing two classification algorithms, the widely-used McNemar's test aims to infer the presence of a significant difference between the error rates of the two classification algorithms. However, the power of the conventional McNemar's test is usually unpromising because the hold-out (HO) method in the test merely uses a single train-validation split that usually produces a highly varied estimation of the error rates. In contrast, a cross-validation (CV) method repeats the HO method in multiple times and produces a stable estimation. Therefore, a CV method has a great advantage to improve the power of McNemar's test. Among all types of CV methods, a block-regularized 5$\times$2 CV (BCV) has been shown in many previous studies to be superior to the other CV methods in the comparison task of algorithms because the 5$\times$2 BCV can produce a high-quality estimator of the error rate by regularizing the numbers of overlapping records between all training sets. In this study, we compress the 10 correlated contingency tables in the 5$\times$2 BCV to form an effective contingency table. Then, we define a 5$\times$2 BCV McNemar's test on the basis of the effective contingency table. We demonstrate the reasonable type I error and the promising power of the proposed 5$\times$2 BCV McNemar's test on multiple simulated and real-world data sets.
format Preprint
id arxiv_https___arxiv_org_abs_2304_03990
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Block-regularized 5$\times$2 Cross-validated McNemar's Test for Comparing Two Classification Algorithms
Yang, Jing
Wang, Ruibo
Song, Yijun
Li, Jihong
Machine Learning
In the task of comparing two classification algorithms, the widely-used McNemar's test aims to infer the presence of a significant difference between the error rates of the two classification algorithms. However, the power of the conventional McNemar's test is usually unpromising because the hold-out (HO) method in the test merely uses a single train-validation split that usually produces a highly varied estimation of the error rates. In contrast, a cross-validation (CV) method repeats the HO method in multiple times and produces a stable estimation. Therefore, a CV method has a great advantage to improve the power of McNemar's test. Among all types of CV methods, a block-regularized 5$\times$2 CV (BCV) has been shown in many previous studies to be superior to the other CV methods in the comparison task of algorithms because the 5$\times$2 BCV can produce a high-quality estimator of the error rate by regularizing the numbers of overlapping records between all training sets. In this study, we compress the 10 correlated contingency tables in the 5$\times$2 BCV to form an effective contingency table. Then, we define a 5$\times$2 BCV McNemar's test on the basis of the effective contingency table. We demonstrate the reasonable type I error and the promising power of the proposed 5$\times$2 BCV McNemar's test on multiple simulated and real-world data sets.
title Block-regularized 5$\times$2 Cross-validated McNemar's Test for Comparing Two Classification Algorithms
topic Machine Learning
url https://arxiv.org/abs/2304.03990