Cannot change k-value for the sample kmean with plot.

kal12

hello community. I need to do k-means cluster using rapidminer and produce elbow method graph. i've tested the k-means clustering with plot sample. However, when I try to change the k value from 13 to 5 for the k-means operator and run, the k value does not change instead it turn back to 13 and produce 13 cluster. Can someone tell me what is the problem and solution?

Thank you in advance.

MartinLiebig

Hi,

this is exactly what i said. You use Loop Parameters to override the k of kmeans. Please change the parameters of Loop Parameters.

Best,

Martin

MartinLiebig

Hi @Kal12,

are you sure that you don't use any optimize around your kmeans? Or that you are using X-means?

Please post your process here, then we can have a look.

Best,

Martin

kal12

Thank you for your response. This is the xml, I also attach the .rmp file.

<?xml version="1.0" encoding="UTF-8"?><process version="9.7.002">
  <context>
    <input/>
    <output/>
    <macros/>
  </context>
  <operator activated="true" class="process" compatibility="9.7.002" expanded="true" name="Root" origin="GENERATED_SAMPLE">
    <parameter key="logverbosity" value="warning"/>
    <parameter key="random_seed" value="2001"/>
    <parameter key="send_mail" value="never"/>
    <parameter key="notification_email" value=""/>
    <parameter key="process_duration_for_mail" value="30"/>
    <parameter key="encoding" value="SYSTEM"/>
    <process expanded="true">
      <operator activated="true" class="retrieve" compatibility="9.7.002" expanded="true" height="68" name="Retrieve" origin="GENERATED_SAMPLE" width="90" x="45" y="34">
        <parameter key="repository_entry" value="../../data/Iris"/>
      </operator>
      <operator activated="true" class="loop_parameters" compatibility="9.7.002" expanded="true" height="82" name="ParameterIteration" origin="GENERATED_SAMPLE" width="90" x="179" y="34">
        <list key="parameters">
          <parameter key="KMeans.k" value="2,3,4,5,6,7,8,9,10,11,13"/>
        </list>
        <parameter key="error_handling" value="fail on error"/>
        <parameter key="synchronize" value="false"/>
        <process expanded="true">
          <operator activated="true" class="concurrency:k_means" compatibility="9.7.002" expanded="true" height="82" name="KMeans" origin="GENERATED_SAMPLE" width="90" x="45" y="34">
            <parameter key="add_cluster_attribute" value="true"/>
            <parameter key="add_as_label" value="false"/>
            <parameter key="remove_unlabeled" value="false"/>
            <parameter key="k" value="13"/>
            <parameter key="max_runs" value="10"/>
            <parameter key="determine_good_start_values" value="false"/>
            <parameter key="measure_types" value="BregmanDivergences"/>
            <parameter key="mixed_measure" value="MixedEuclideanDistance"/>
            <parameter key="nominal_measure" value="NominalDistance"/>
            <parameter key="numerical_measure" value="EuclideanDistance"/>
            <parameter key="divergence" value="SquaredEuclideanDistance"/>
            <parameter key="kernel_type" value="radial"/>
            <parameter key="kernel_gamma" value="1.0"/>
            <parameter key="kernel_sigma1" value="1.0"/>
            <parameter key="kernel_sigma2" value="0.0"/>
            <parameter key="kernel_sigma3" value="2.0"/>
            <parameter key="kernel_degree" value="3.0"/>
            <parameter key="kernel_shift" value="1.0"/>
            <parameter key="kernel_a" value="1.0"/>
            <parameter key="kernel_b" value="0.0"/>
            <parameter key="max_optimization_steps" value="100"/>
            <parameter key="use_local_random_seed" value="false"/>
            <parameter key="local_random_seed" value="1992"/>
          </operator>
          <operator activated="true" class="cluster_distance_performance" compatibility="9.7.002" expanded="true" height="103" name="Evaluation" origin="GENERATED_SAMPLE" width="90" x="179" y="34">
            <parameter key="main_criterion" value="Avg. within centroid distance"/>
            <parameter key="main_criterion_only" value="false"/>
            <parameter key="normalize" value="false"/>
            <parameter key="maximize" value="false"/>
          </operator>
          <operator activated="true" class="log" compatibility="9.7.002" expanded="true" height="103" name="ProcessLog" origin="GENERATED_SAMPLE" width="90" x="313" y="34">
            <list key="log">
              <parameter key="k" value="operator.KMeans.parameter.k"/>
              <parameter key="DB" value="operator.Evaluation.value.DaviesBouldin"/>
              <parameter key="W" value="operator.Evaluation.value.avg_within_distance"/>
            </list>
            <parameter key="sorting_type" value="none"/>
            <parameter key="sorting_k" value="100"/>
            <parameter key="persistent" value="false"/>
          </operator>
          <connect from_port="input 1" to_op="KMeans" to_port="example set"/>
          <connect from_op="KMeans" from_port="cluster model" to_op="Evaluation" to_port="cluster model"/>
          <connect from_op="KMeans" from_port="clustered set" to_op="Evaluation" to_port="example set"/>
          <connect from_op="Evaluation" from_port="performance" to_op="ProcessLog" to_port="through 1"/>
          <connect from_op="Evaluation" from_port="example set" to_op="ProcessLog" to_port="through 2"/>
          <connect from_op="ProcessLog" from_port="through 1" to_port="performance"/>
          <connect from_op="ProcessLog" from_port="through 2" to_port="result 1"/>
          <portSpacing port="source_input 1" spacing="0"/>
          <portSpacing port="source_input 2" spacing="0"/>
          <portSpacing port="sink_performance" spacing="0"/>
          <portSpacing port="sink_result 1" spacing="0"/>
          <portSpacing port="sink_result 2" spacing="0"/>
        </process>
      </operator>
      <connect from_op="Retrieve" from_port="output" to_op="ParameterIteration" to_port="input 1"/>
      <connect from_op="ParameterIteration" from_port="result 1" to_port="result 1"/>
      <portSpacing port="source_input 1" spacing="0"/>
      <portSpacing port="sink_result 1" spacing="0"/>
      <portSpacing port="sink_result 2" spacing="0"/>
    </process>
  </operator>
</process>

MartinLiebig

Hi,

this is exactly what i said. You use Loop Parameters to override the k of kmeans. Please change the parameters of Loop Parameters.

Best,

Martin

kal12

Ok, now I see. Thank you very much.