Package unifeat.featureSelection.wrapper.GABasedMethods.HGAFS

Class HGAFS

java.lang.Object
unifeat.featureSelection.FeatureSelection
unifeat.featureSelection.wrapper.WrapperApproach
unifeat.featureSelection.wrapper.GABasedMethods.BasicGA<Population>
unifeat.featureSelection.wrapper.GABasedMethods.HGAFS.HGAFS
public class HGAFS extends BasicGA<Population>
This java class is used to implement feature selection method based on hybrid genetic algorithm for feature selection using local search (HGAFS) in which the type of Population is extended from Population class.
Author:
Sina Tabakhi
See Also:

  • Constructor Details

    • HGAFS

      public HGAFS(Object... arguments)
      Initializes the parameters
      Parameters:
      arguments - array of parameters contains ( path, numFeatures, classifierType, selectedClassifierPan, selectionType, crossoverType, mutationType, replacementType, numIteration populationSize, crossoverRate, mutationRate, kFolds , epsilon , mu ) in which path is the path of the project, numFeatures is the number of original features in the dataset, classifierType is the classifier type for evaluating the fitness of a solution, selectedClassifierPan is the selected classifier panel, selectionType is used for selecting parents from the individuals of a population according to their fitness, crossoverType is used for recombining the parents to generate new offsprings based on crossover rate, mutationType is used for mutating new offsprings by changing the value of some genes in them based on mutation rate, replacementType is used for handling populations from one generation to the next generation, numIteration is the maximum number of allowed iterations that algorithm repeated, populationSize is the size of population of candidate solutions, crossoverRate is the probability of crossover operation, mutationRate is the probability of mutation operation, kFolds is the number of equal sized subsamples that is used in k-fold cross validation, epsilon is the epsilon parameter used in the subset size determining scheme, and mu is the mu parameter used in the local search operation

  • Method Details

    • createSelectedFeatureSubset

      protected int[] createSelectedFeatureSubset()
      This method creates the selected feature subset based on the fittest individual in the population.
      Specified by:
      createSelectedFeatureSubset in class BasicGA<Population>
      Returns:
      the array of indices of the selected feature subset

    • evaluateFeatures

      public void evaluateFeatures()
      Starts the feature selection process by hybrid genetic algorithm for feature selection using local search (HGAFS)
      Specified by:
      evaluateFeatures in class FeatureSelection