Bootstrap analysis of multiple repetitions of experiments using an interval-valued multiple comparison procedure

作者：

Highlights：

• Bootstrap tests improve ANOVA or Friedman when comparing multiple classifiers.

• Interval-valued tests improve nonparametric tests for stochastic algorithms.

• Friedman and Bootstrap tests produce different conclusions for certain ML tasks.

摘要

•Bootstrap tests improve ANOVA or Friedman when comparing multiple classifiers.•Interval-valued tests improve nonparametric tests for stochastic algorithms.•Friedman and Bootstrap tests produce different conclusions for certain ML tasks.

论文关键词：Cross validation,Statistical comparisons of algorithms,Tests for interval-valued data

论文评审过程：Received 23 July 2012, Revised 5 December 2012, Accepted 14 March 2013, Available online 21 March 2013.

论文官网地址：https://doi.org/10.1016/j.jcss.2013.03.009