Two-Stage Bagging Pruning for Reducing the Ensemble Size and Improving the Classification Performance

Algorithm 2

Accuracy based pruning for bagging algorithm.

Input:  -training set, -bootstrap subsets from , - number of base models or subsets, - a set of base models

Output: -a reduced set of base models, - a pruned bagging ensemble

1  Initialize

2  Collect the subsets of out-of-bag samples as .

3  Calculate the accuracy for each base model tested on the

4  Given a parameter , compute the threshold , which is the -th decile value of the set

5  for    do:

6    if :

7        

8 The outcome of a test sample predicted by the pruned ensemble is given as follows: