UPDF AI

Statistical Comparisons of Classifiers over Multiple Data Sets

J. Demšar

2006 · DBLP: journals/jmlr/Demsar06
Journal of machine learning research · 13,015 Citations

TLDR

A set of simple, yet safe and robust non-parametric tests for statistical comparisons of classifiers is recommended: the Wilcoxon signed ranks test for comparison of two classifiers and the Friedman test with the corresponding post-hoc tests for comparisons of more classifiers over multiple data sets.