"fitness function of Genetic algorithm"

IatiiIatii Member Posts: 8 Contributor I
edited June 2019 in Help

Dear all,


Can anyone tell me what is the "fitness function" of "Genetic algorithm" in " evolutionary selection" ? 




Best Answer

  • Options
    IngoRMIngoRM Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, Community Manager, RMResearcher, Member, University Professor Posts: 1,751 RM Founder
    Solution Accepted



    It is the function you want to optimize for.  Typically you would use some validation scheme like cross validation inside of the genetic algorithm and the performance estimation delivered by the cross validation becomes the fitness function.  This means that if you for example calculate the "accuracy" with the cross validation, this is what should be maximized.  If it is root mean squared error, this should be minimized.  You don't need to worry about this though since RapidMiner will automatically chose the correct direction (minimization or maximization).


    If you double click on the selection operator, you will see the inside where you can do the performance calculation.  You can also do this in a different way as long as you deliver a performance vector to the inner output port.


    Here is the process from the "Samples" directory showing you how this looks in detail:


    <?xml version="1.0" encoding="UTF-8"?><process version="7.2.000">
    <operator activated="true" class="process" compatibility="7.2.000" expanded="true" name="Root">
    <parameter key="random_seed" value="1976"/>
    <process expanded="true">
    <operator activated="true" class="generate_data" compatibility="7.1.001" expanded="true" height="68" name="ExampleSetGenerator" width="90" x="45" y="30">
    <parameter key="target_function" value="sum"/>
    <parameter key="number_examples" value="200"/>
    <parameter key="number_of_attributes" value="4"/>
    <parameter key="attributes_lower_bound" value="0.0"/>
    <operator activated="true" class="add_noise" compatibility="7.1.001" expanded="true" height="103" name="NoiseGenerator" width="90" x="180" y="30">
    <parameter key="random_attributes" value="8"/>
    <parameter key="label_noise" value="0.0010"/>
    <list key="noise"/>
    <operator activated="true" class="optimize_selection_evolutionary" compatibility="7.2.000" expanded="true" height="103" name="GeneticAlgorithm" width="90" x="313" y="30">
    <parameter key="population_size" value="6"/>
    <parameter key="maximum_number_of_generations" value="12"/>
    <parameter key="show_stop_dialog" value="true"/>
    <parameter key="selection_scheme" value="Boltzmann"/>
    <process expanded="true">
    <operator activated="true" class="x_validation" compatibility="7.2.000" expanded="true" height="112" name="XValidation" width="90" x="45" y="30">
    <parameter key="sampling_type" value="shuffled sampling"/>
    <process expanded="true">
    <operator activated="true" class="linear_regression" compatibility="7.2.000" expanded="true" name="LinearRegression">
    <parameter key="feature_selection" value="none"/>
    <connect from_port="training" to_op="LinearRegression" to_port="training set"/>
    <connect from_op="LinearRegression" from_port="model" to_port="model"/>
    <portSpacing port="source_training" spacing="0"/>
    <portSpacing port="sink_model" spacing="0"/>
    <portSpacing port="sink_through 1" spacing="0"/>
    <process expanded="true">
    <operator activated="true" class="apply_model" compatibility="7.1.001" expanded="true" name="ModelApplier">
    <list key="application_parameters"/>
    <operator activated="true" class="performance_regression" compatibility="7.2.000" expanded="true" name="RegressionPerformance">
    <parameter key="main_criterion" value="root_relative_squared_error"/>
    <parameter key="root_relative_squared_error" value="true"/>
    <connect from_port="model" to_op="ModelApplier" to_port="model"/>
    <connect from_port="test set" to_op="ModelApplier" to_port="unlabelled data"/>
    <connect from_op="ModelApplier" from_port="labelled data" to_op="RegressionPerformance" to_port="labelled data"/>
    <connect from_op="RegressionPerformance" from_port="performance" to_port="averagable 1"/>
    <portSpacing port="source_model" spacing="0"/>
    <portSpacing port="source_test set" spacing="0"/>
    <portSpacing port="source_through 1" spacing="0"/>
    <portSpacing port="sink_averagable 1" spacing="0"/>
    <portSpacing port="sink_averagable 2" spacing="0"/>
    <operator activated="true" class="log" compatibility="7.2.000" expanded="true" height="76" name="ProcessLog" width="90" x="425" y="48">
    <list key="log">
    <parameter key="gen" value="operator.GeneticAlgorithm.value.generation"/>
    <parameter key="perf" value="operator.GeneticAlgorithm.value.performance"/>
    <parameter key="best" value="operator.GeneticAlgorithm.value.best"/>
    <connect from_port="example set" to_op="XValidation" to_port="training"/>
    <connect from_op="XValidation" from_port="averagable 1" to_op="ProcessLog" to_port="through 1"/>
    <connect from_op="ProcessLog" from_port="through 1" to_port="performance"/>
    <portSpacing port="source_example set" spacing="0"/>
    <portSpacing port="source_through 1" spacing="0"/>
    <portSpacing port="sink_performance" spacing="0"/>
    <connect from_op="ExampleSetGenerator" from_port="output" to_op="NoiseGenerator" to_port="example set input"/>
    <connect from_op="NoiseGenerator" from_port="example set output" to_op="GeneticAlgorithm" to_port="example set in"/>
    <connect from_op="GeneticAlgorithm" from_port="example set out" to_port="result 1"/>
    <connect from_op="GeneticAlgorithm" from_port="weights" to_port="result 2"/>
    <connect from_op="GeneticAlgorithm" from_port="performance" to_port="result 3"/>
    <portSpacing port="source_input 1" spacing="0"/>
    <portSpacing port="sink_result 1" spacing="0"/>
    <portSpacing port="sink_result 2" spacing="0"/>
    <portSpacing port="sink_result 3" spacing="0"/>
    <portSpacing port="sink_result 4" spacing="0"/>
    <description align="left" color="yellow" colored="false" height="174" resized="false" width="400" x="40" y="140">This process is very similar to the previous process. Again, a cross validation building block is used as fitness evaluation. In this case, a genetic algortihm for feature selection is used. This process demonstrates two other features of the feature operators of RapidMiner: the stop button which allows the abort of the process if the user was already satisfied and the ProcessLog operator which allows online plotting of current fitness values. Please refer to the visualisation sample processes and the RapidMiner operator manual for further details.</description>

    Hope this helps,


Sign In or Register to comment.