Permutation tests for studying classifier performance

MaRDI QIDQ2896117zbMATH OpenFDO

Authors Markus Ojala, Gemma C. Garriga

Publication date 13 July 2012

Published in Journal of Machine Learning Research (JMLR) (Search for Journal in Brave)

Full work available at URL http://www.jmlr.org/papers/v11/ojala10a.html

zbMATH Keywords

restricted randomizationsignificance testinglabeled data

Mathematics Subject Classification ID

Nonparametric hypothesis testing (62G10) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Learning and adaptive systems in artificial intelligence (68T05)

Recommendations

• Learning Theory
• Machine Learning: ECML 2004
• A permutation based procedure for classification assessment
• Minimax optimality of permutation tests
• Classification accuracy as a proxy for two-sample testing

Cited in

(13)

• Global and local two-sample tests via regression
• A permutation test approach to the choice of sizekfor the nearest neighbors classifier
• Computing exact permutation \(p\)-values for association rules
• Minimax optimality of permutation tests
• Machine Learning: ECML 2004
• Robust ordinal regression for subsets comparisons with interactions
• Permutation Multiple Tests of Binary Features Do Not Uniformly Control Error Rates
• A permutation based procedure for classification assessment
• Wavelet-domain regression and predictive inference in psychiatric neuroimaging
• Another Look at the Lady Tasting Tea and Differences Between Permutation Tests and Randomisation Tests
• Classification on large networks: a quantitative bound via motifs and graphons (research)
• Perturbation-based classifier
• Learning Theory

Describes a project that uses

Uses Software

• OSCAR

This page was built for publication: Permutation tests for studying classifier performance

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2896117)