Package edu.udo.cs.yale.operator.features.selection

Provides operators for automatic feature selection.

See:
Description

Class Summary

AbstractGeneticAlgorithm	Genetic algorithms are general purpose optimization / search algorithms that are suitable in case of no or little problem knowledge.
AttributeWeightSelection	This operator selects all attributes which have a weight fulfilling a given condition.
BackwardElimination	This PopulationOperator realises backward elimination, i.e. creates a list of clones of each individual and switches of one attribute in each of the clones.
BestSelection	Selects the best individual and build a new population.
BoltzmannSelection	Like RouletteWheel this population operator selects a given fixed number of individuals by subdividing a roulette wheel in sections of size proportional to the individuals' fitness values.
BruteForceSelection	This feature selection operator selects the best attribute set by trying all possible combinations of attribute selections.
CutSelection	Creates a new population by a deterministical selection of the best individuals.
FeatureSelectionOperator	This operator realizes the two deterministic greedy feature selection algorithms forward selection and backward elimination.
ForwardSelection	This PopulationOperator realises forward selection, i.e. creates a list of clones of each individual and switches on one attribute in each of the clones.
GeneticAlgorithm	A genetic algorithm for feature selection (mutation=switch features on and off, crossover=interchange used features).
IterativeFeatureAdding	Adds iteratively the next feature according to given attribute name array.
NonDominatedSortingSelection	Performs the non dominated sorting selection from NSGA II.
NonDominatedSortingSelection.CriteriaComparator	The comparator for aggregation individuals using the fitness values of the m-th criterion.
NonDominatedSortingSelection.CrowdingComparator	The comparator for aggregation individuals using the crowding distance.
RankSelection	Selects a given fixed number of individuals by subdividing a roulette wheel in sections of size proportional to the individuals' rank based on their fitness values.
RemoveCorrelatedFeatures	Removes (un-) correlated features due to the selected filter relation.
RemoveUselessFeatures	Removes useless attribute from the example set.
RouletteWheel	Selects a given fixed number of individuals by subdividing a roulette wheel in sections of size proportional to the individuals' fitness values.
SelectionCrossover	Crossover operator for the used bitlists of example sets.
SelectionMutation	Inverts the used bit for every feature of every example set with a given fixed probability.
StochasticUniversalSampling	Similar to a the roulette wheel selection the fitness values of all individuals build a partition of the 360 degrees of a wheel.
SwitchingForwardSelection	This PopulationOperator realises forward selection, i.e. creates a list of clones of each individual and switches on one attribute in each of the clones.
TournamentSelection	Performs tournaments with k participants.
UniformSelection	Selects a given fixed number of individuals by uniformly sampling from the current population until the desired population size is reached.
WeightGuidedSelectionOperator	This operator uses input attribute weights to determine the order of features added to the feature set starting with the feature set containing only the feature with highest weight.

Package edu.udo.cs.yale.operator.features.selection Description

Provides operators for automatic feature selection.