Optimize by Generation (YAGGA2)
Synopsis
This operator may select some attributes from the original attribute set and it may also generate new attributes from the original attribute set. YAGGA2 (Yet Another Generating Genetic Algorithm 2) does not change the original number of attributes unless adding or removing (or both) attributes proves to have a better fitness. This algorithm is an improved version of YAGGA.
Description
Sometimes the selection of features alone is not sufficient. In these cases other transformations of the feature space must be performed. The generation of new attributes from the given attributes extends the feature space. Maybe a hypothesis can be easily found in the extended feature space. This operator can be considered to be a blend of attribute selection and attribute generation procedures. It may select some attributes from the original set of attributes and it may also generate new attributes from the original attributes. The (generating) mutation can do one of the following things with different probabilities:
- Probability p/4: Add a newly generated attribute to the feature vector.
- Probability p/4: Add a randomly chosen original attribute to the feature vector.
- Probability p/2: Remove a randomly chosen attribute from the feature vector.
在addition to the usual YAGGA operator, this operator allows more feature generators and provides several techniques for redundancy prevention. This leads to smaller ExampleSets containing less redundant features.
A genetic algorithm (GA) is a search heuristic that mimics the process of natural evolution. This heuristic is routinely used to generate useful solutions to optimization and search problems. Genetic algorithms belong to the larger class of evolutionary algorithms (EA), which generate solutions to optimization problems using techniques inspired by natural evolution, such as inheritance, mutation, selection, and crossover. For studying the basic algorithm of a genetic algorithm please study the description of the
This operator is a nested operator i.e. it has a subprocess. The subprocess must return a performance vector. You need to have basic understanding of subprocesses in order to apply this operator. Please study the documentation of the
Differentiation
Optimize by Generation (YAGGA)
The YAGGA2 operator is an improved version of the usual YAGGA operator, this operator allows more feature generators and provides several techniques for redundancy prevention. This leads to smaller ExampleSets containing less redundant features.
Input
example set in
This input port expects an ExampleSet. This ExampleSet is available at the first port of the nested chain (inside the subprocess) for processing in the subprocess.
Output
example set out
The genetic algorithm is applied on the input ExampleSet. The resultant ExampleSet is delivered through this port.
attribute weights out
The attribute weights are delivered through this port.
performance out
This port delivers the Performance Vector for the selected attributes. A Performance Vector is a list of performance criteria values.
Parameters
Limit max total number of attributes
This parameter indicates if the total number of attributes in all generations should be limited. If set to true, the maximum number is specified by themax total number of attributesparameter.
Max total number of attributes
This parameter is only available when thelimit max total number of attributesparameter is set to true. This parameter specifies the maximum total number of attributes in all generations.
Use local random seed
This parameter indicates if alocal random seedshould be used for randomization. Using the same value oflocal random seedwill produce the same randomization.
Local random seed
This parameter specifies thelocal random seed. This parameter is only available if theuse local random seedparameter is set to true.
Show stop dialog
This parameter determines if a dialog with astopbutton should be displayed which stops the search for the best feature space. If the search for the best feature space is stopped, the best individual found till then will be returned.
Maximal fitness
This parameter specifies the maximal fitness. The optimization will stop if the fitness reaches this value.
Population size
This parameter specifies the population size i.e. the number of individuals per generation.
Maximum number of generations
This parameter specifies the number of generations after which the algorithm should be terminated.
Use plus
This parameter indicates if the summation function should be applied for a generation of new attributes.
Use diff
This parameter indicates if the difference function should be applied for a generation of new attributes.
Use mult
This parameter indicates if the multiplication function should be applied for a generation of new attributes.
Use div
This parameter indicates if the division function should be applied for a generation of new attributes.
Reciprocal value
This parameter indicates if the reciprocal function should be applied for a generation of new attributes.
Use early stopping
This parameter enables early stopping. If not set to true, always the maximum number of generations are performed.
Generations without improval
This parameter is only available when the使用早期停止parameter is set to true. This parameter specifies the stop criterion for early stopping i.e. it stops afterngenerations without improvement in the performance.nis specified by this parameter.
Tournament size
This parameter specifies the fraction of the current population which should be used as tournament members.
Start temperature
This parameter specifies the scaling temperature.
Dynamic selection pressure
If this parameter is set to true, the selection pressure is increased to maximum during the complete optimization run.
Keep best individual
If set to true, the best individual of each generation is guaranteed to be selected for the next generation.
P initialize
The initial probability for an attribute to be switched on is specified by this parameter.
P crossover
The probability for an individual to be selected for crossover is specified by this parameter.
Crossover type
可以选择交叉的类型parameter.
Use heuristic mutation probability
If this parameter is set to true, the probability for mutations will be chosen as1/nwherenis the number of attributes. Otherwise the probability for mutations should be specified through thep mutationparameter
P mutation
The probability for an attribute to be changed is specified by this parameter. If set to -1, the probability will be set to1/nwherenis the total number of attributes.
Use square roots
This parameter indicates if the square root function should be applied for a generation of new attributes.
Use power functions
This parameter indicates if the power (of one attribute to another attribute) function should be applied for a generation of new attributes.
Use sin
This parameter indicates if the sine function should be applied for a generation of new attributes.
Use cos
This parameter indicates if the cosine function should be applied for a generation of new attributes.
Use tan
This parameter indicates if the tangent function should be applied for a generation of new attributes.
Use atan
This parameter indicates if the arc tangent function should be applied for a generation of new attributes.
Use exp
This parameter indicates if the exponential function should be applied for a generation of new attributes.
Use log
This parameter indicates if the logarithmic function should be applied for a generation of new attributes.
Use absolute values
This parameter indicates if the absolute function should be applied for a generation of new attributes.
Use min
This parameter indicates if the minimum function should be applied for a generation of new attributes.
Use max
This parameter indicates if the maximum function should be applied for a generation of new attributes.
Use sgn
This parameter indicates if the signum function should be applied for a generation of new attributes.
Use floor ceil functions
This parameter indicates if the floor and ceiling functions should be applied for a generation of new attributes.
Restrictive selection
This parameter indicates if the restrictive generator selection should be used. Execution is usually faster if this parameter is set to true.
Remove useless
This parameter indicates if useless attributes should be removed.
Remove equivalent
This parameter indicates if equivalent attributes should be removed.
Equivalence samples
nnumber of samples are checked to prove equivalency wherenis the value of this parameter.
Equivalence epsilon
如果他们两个属性被认为是等价的difference is not bigger than epsilon.
Equivalence use statistics
If this parameter is set to true, attribute statistics are recalculated before equivalence check.
Unused functions
This parameter specifies the space separated list of functions which are not allowed in arguments for the attribute construction.
Constant generation prob
This parameter specifies the probability for a generation of random constant attributes.
Associative attribute merging
This parameter specifies if post processing should be performed after the crossover. It is only possible for runs with only one generator.