An Approach for Identifying Cytokines Based on a Novel Ensemble Classifier

Algorithm 1

SMOTE over-sampling.

(1) Input: small sample set 𝑆, over-sampling magnification 𝑁;

(2) Output: the new small sample set 𝑆^′, sample number 𝑘;

(3) For each small class sample 𝑥, find 𝑦 nearest neighbors of the same kind with 𝑥;

(4) Choose 𝑁 samples according to the magnification of over-sampling randomly, then do random linear interpolation between 𝑥

   and each neighbor selected;