Research Article
enDNA-Prot: Identification of DNA-Binding Proteins by Applying Ensemble Learning
Algorithm 1
The pseudocode of Unbalanced-AdaBoost.
Input: | positive train dataset ; | negative train dataset ; | Base learning algorithm ; | Number of learning rounds . | Output: | | Process: | (1) //Initialize the weight distribution on | (2) For : | (3) ; //sampling negative samples from the negative train dataset with weight | distribution | (4) ; //combine the positive train dataset and sampled dataset into a dataset | (5) ; //train the base learner on the dataset | (6) ; //test on the negative dataset and calculate its predicted error | (7) ; //calculate the voting weight of the base learner | (8) //calculate the weight distribution for the next learning round, | where size denotes the number of samples in and used to ensure that is a distribution. | (10) End |
|