example

A dataset containing software metrics of 1,000 calculation of Breiman's variable importance scores

datasets

The Scott-Knott Effect Size Difference (ESD) test is a mean comparison approach that leverages a hierarchical clustering to partition the set of treatment means (e.g., means of variable importance scores, means of model performance) into statistically distinct groups with non-negligible difference [Tantithamthavorn et al., (2018) <doi:10.1109/TSE.2018.2794977>].

Chakkrit Tantithamthavorn

ScottKnottESD

The Scott-Knott Effect Size Difference (ESD) Test

example function

A data frame with 1,000 rows and 9 variables:<dl class="dl-horizontal">
 <dt>LOC</dt><dd>lines of code</dd>
 <dt>Components</dt><dd>the numbers of components</dd>
 <dt>Subsystem</dt><dd>the numbers of subsystems</dd>
 <dt>Files</dt><dd>the numbers of files</dd>
 <dt>Commit</dt><dd>the numbers of commits</dd>
 <dt>Churn</dt><dd>the numbers of churns</dd>
 <dt>Ownership</dt><dd>ownership</dd>
 <dt>Authorship</dt><dd>authorship</dd>
 <dt>Experience</dt><dd>developer's experience</dd>
 ...
</dl>

Format

An example dataset of Breiman's variable importance scores — example

A data frame with 1,000 rows and 9 variables:<dl class='dl-horizontal'>
 <dt>LOC</dt><dd>lines of code</dd>
 <dt>Components</dt><dd>the numbers of components</dd>
 <dt>Subsystem</dt><dd>the numbers of subsystems</dd>
 <dt>Files</dt><dd>the numbers of files</dd>
 <dt>Commit</dt><dd>the numbers of commits</dd>
 <dt>Churn</dt><dd>the numbers of churns</dd>
 <dt>Ownership</dt><dd>ownership</dd>
 <dt>Authorship</dt><dd>authorship</dd>
 <dt>Experience</dt><dd>developer's experience</dd>
 ...
</dl>

example: An example dataset of Breiman's variable importance scores

Description

Usage

Arguments

Format