Algorithm 4 Pseudocode of imbalanced metric-driven models based on AUE.

Input: S: new data chunk

C: ensemble of classifiers

K: size of the ensemble

Output: C: ensemble of updated classifiers with updated weights

X ← sampled S

Train new classifier $C^{\prime }$ na X;

Estimate weight of $C^{\prime }$ using cross-validation on S based on 5, 6 or 7;

for $C_i$ in C do

Calculate weight of $C_i$ on S based on 5, 6 or 7;

end for

Calculate weight $w_R$ of random classifier on S based on 5, 6 or 6 and a priori probabilities;

for $C_i$ in C do

if $w_i>w_R$ then

Update $C_i$ with S;

end if

end for

C ← K classifiers with the highest weights from $C\cup C^{\prime }$;

for $C_i$ in C do

$w_i\leftarrow \frac{w_i}{{\sum }_{C_{i}inC\cup C^{\prime }}{w}_i}$

end for

return C;