Decision tree

sshilderman

I'm trying to use a decision tree to predict user will leave.

My data include 4 regular attributes (2 nominal, 2 integer), and 1 special attribute (nominal label).

When using the Decision Tree operator I don't get a tree with all data, only one of the regular appear (as root) and the leafs contains the label data (which is OK).

 

What am I doing wrong?

bhupendra_patil

Hello, this may be simply happening because the data does not have patterns that fit the criteria you set.

 

I will suggest trying values for pruning, prepruning and confidence values.

 

A better way to find a right value for these would be using the "Optimize Parameters (Grid) operator and giving it a range to try combinations of some of these variables that affect your model.

 

You should be able to see a sample process in the help for "Optimize Parameters(Grid)" to see how this operator works

 

Good Luck

sshilderman

Followup question -

 

First of all, thank you for your answer.

I created a table with patterns (manually), first to check i'm doing it right.

 

Is there a way to know who is located in each leaf?

I would like to learn which users will have a specific value (the labell value) in the future.

 

Bests. 

MartinLiebig

Hi,

 

what you can do is use the tree to rules operator. As a result (see attached process) you get the paths as strings. That might be helpful in first place. There is no one operator solution to apply this rules to a dataset to get "leaf IDs" but it might be possible to find some working process with things like Write as Text and then parse the resulting text files.

 

Best,

Martin

<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<process version="7.1.001">
  <context>
    <input/>
    <output/>
    <macros/>
  </context>
  <operator activated="true" class="process" compatibility="7.1.001" expanded="true" name="Process">
    <process expanded="true">
      <operator activated="true" class="retrieve" compatibility="7.1.001" expanded="true" height="68" name="Retrieve Golf" width="90" x="112" y="85">
        <parameter key="repository_entry" value="//Samples/data/Golf"/>
      </operator>
      <operator activated="true" class="tree_to_rules" compatibility="7.1.001" expanded="true" height="82" name="Tree to Rules" width="90" x="246" y="85">
        <process expanded="true">
          <operator activated="true" class="parallel_decision_tree" compatibility="7.1.001" expanded="true" height="82" name="Decision Tree" width="90" x="45" y="34"/>
          <connect from_port="training set" to_op="Decision Tree" to_port="training set"/>
          <connect from_op="Decision Tree" from_port="model" to_port="model"/>
          <portSpacing port="source_training set" spacing="0"/>
          <portSpacing port="sink_model" spacing="0"/>
        </process>
      </operator>
      <operator activated="true" class="apply_model" compatibility="7.1.001" expanded="true" height="82" name="Apply Model" width="90" x="380" y="85">
        <list key="application_parameters"/>
      </operator>
      <connect from_op="Retrieve Golf" from_port="output" to_op="Tree to Rules" to_port="training set"/>
      <connect from_op="Tree to Rules" from_port="model" to_op="Apply Model" to_port="model"/>
      <connect from_op="Tree to Rules" from_port="example set" to_op="Apply Model" to_port="unlabelled data"/>
      <connect from_op="Apply Model" from_port="model" 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"/>
      <portSpacing port="sink_result 3" spacing="0"/>
    </process>
  </operator>
</process>