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

Algorithm 4

Two-stage pruning for “AP+DP”.

Input:  -a reduced set of base models generated by the AP procedure, - number of base models in ,

- index set of base models in , -feature vector of a test sample

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

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

   .

2  Initialize

3  for    do:

4    if :

5      

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