"How to get the standard deviation from clustered data?"

stever1k

Hi,

after clustering my data, the data has the following format:

id A B C Cluster
a x y z  0
.. .... .... 1
.. .... .... 1
.. .... .... 2
.. .... .... 0
.. .... ....
.. .... .... N

So the cluster algorithm found several clusters and created a new column with the attribute cluster. I now want to calculate the standard deviation for Cluster 0 for the attributes A B and C, the same for cluster 1 up to N. Any ideas how this works?

cordially,
Stever

land

Hi Stever,
this is a typical situation for using the aggregation operator. You can group the examples by the cluster and then calculate a aggregation function over each attribute. I have done this in this process:

<operator name="Root" class="Process" expanded="yes">
    <operator name="ExampleSetGenerator" class="ExampleSetGenerator">
        <parameter key="target_function"	value="gaussian mixture clusters"/>
    </operator>
    <operator name="KMeans" class="KMeans">
        <parameter key="k"	value="3"/>
    </operator>
    <operator name="ChangeAttributeRole" class="ChangeAttributeRole">
        <parameter key="name"	value="cluster"/>
    </operator>
    <operator name="Aggregation" class="Aggregation">
        <list key="aggregation_attributes">
          <parameter key="att1"	value="standard_deviation"/>
          <parameter key="att2"	value="standard_deviation"/>
          <parameter key="att3"	value="standard_deviation"/>
          <parameter key="att4"	value="standard_deviation"/>
          <parameter key="att5"	value="standard_deviation"/>
        </list>
        <parameter key="group_by_attributes"	value="cluster"/>
    </operator>
</operator>

It should be easy to adapt it onto your needs.

Greetings,
  Sebastian

stever1k

thanks a lot Sebastian, that is EXACTLY what I'm looking for. My problem was, that I was searching for suitable operator inside the preprocession->attributres tree instead of the olap!

best regards,
Stever