Editing Bootstrap aggregating (section)

== Algorithm (classification) ==
[[File:Bagging for Classification with descripitons.png|thumb|Flow chart of the bagging algorithm when used for classification]]
For classification, use a training set <math>D</math>, Inducer <math>I</math> and the number of bootstrap samples <math>m</math> as input. Generate a classifier <math>C^*</math> as output<ref name="Bauer">{{cite journal|last1=Bauer|first1=Eric|last2=Kohavi|first2=Ron|date=1999|title=An Empirical Comparison of Voting Classification Algorithms: Bagging, Boosting, and Variants|journal=Machine Learning|volume=36|pages=108–109|doi=10.1023/A:1007515423169|s2cid=1088806|doi-access=free}}</ref>

# Create <math>m</math> new training sets  <math>D_i</math>, from <math>D</math> with replacement
# Classifier <math>C_i</math> is built from each set <math>D_i</math> using <math>I</math> to determine the classification of set <math>D_i</math>
# Finally classifier <math>C^*</math> is generated by using the previously created set of classifiers <math>C_i</math> on the original dataset <math>D</math>, the classification predicted most often by the sub-classifiers <math>C_i</math> is the final classification 
<pre>
for i = 1 to m {
    D' = bootstrap sample from D    (sample with replacement)
    Ci = I(D')
}
C*(x) = argmax #{i:Ci(x)=y}         (most often predicted label y)
         y∈Y   
</pre>