Research Article
A Weighted Voting Classifier Based on Differential Evolution
Algorithm 1
DE algorithm for DEWVote’s model selection.
Input: The control parameters of DE: mutation factor F, crossover rate CR, and population size N. | (1) Initialization(); {Generate uniformly distributed random population of N individuals | , where is a | vector representing the weights (, , …, ,…, ) of D base classifiers.} | (2) Set the generation iterator . | (3) while the stopping criterion is not satisfied do | (4) for do | (5) Select random indexes, , and to be different from each other and from the index i. | (6) Compute a mutant vector using (1). | (7) Generate random number . | (8) for do | (9) Decide trial individual using (6). | (10) end for | (11) Compute the fitness of the vector and using 10-fold cross validation, and | update the vector of the next generation () using (7). | (12) end for | (13) Update generation iterator . | (14) end while | Output: The optimal weights (, , …, , …, ) for DEWVote. |
|