Pages: [1]
  Print  
Author Topic: Recursive partitioning regression tree?  (Read 312 times)
Ted
Newbie
*
Posts: 14


« on: January 16, 2014, 03:56:51 AM »

I would like to discover explanatory variables for a numeric dependent variable (which conditions result in the shortest commute times on any given week day). Some of the explanatory variables are numeric, others are categoric. Does RapidMiner have any algorithms that can handle that?

I understand that recursive partitioning regression trees can. Is there an equivalent or alternative in RM?
Logged
Marius
Administrator
Hero Member
*****
Posts: 1753



WWW
« Reply #1 on: January 16, 2014, 10:06:13 AM »

Hi Ted,

RapidMiner's Decision Trees can only be used for classification problems, not for regression tasks. But they can deal with both nominal an numeric explanatory attributes.

If you want to stick to decision trees you can discretize the label (as we call the dependent variable) with one of the Discretize operators.

However, decision trees are probably not the best way to define the explanatory power of variables. For that you could e.g. try the Linear Regression and have a look at p-value and attribute weights of the resulting model, or use a Forward Selection around a decision tree.

Best regards,
Marius
Logged

Please add [SOLVED] to the topic title when your problem has been solved! (do so by editing the first post in the thread and modifying the title)
Please click here before posting.
Ralf Klinkenberg
Administrator
Jr. Member
*****
Posts: 65



WWW
« Reply #2 on: January 16, 2014, 08:20:32 PM »

Hi Ted,

if you install the Weka Extension for RapidMiner, you can use regression tree learners from Weka seamlessly within RapidMiner:

Code:
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<process version="6.0.001">
  <context>
    <input/>
    <output/>
    <macros/>
  </context>
  <operator activated="true" class="process" compatibility="6.0.001" expanded="true" name="Process">
    <process expanded="true">
      <operator activated="true" class="generate_data" compatibility="6.0.001" expanded="true" height="60" name="Generate Data" width="90" x="45" y="30">
        <parameter key="target_function" value="polynomial"/>
        <parameter key="number_of_attributes" value="3"/>
        <parameter key="attributes_lower_bound" value="0.0"/>
      </operator>
      <operator activated="true" class="multiply" compatibility="6.0.001" expanded="true" height="94" name="Multiply" width="90" x="45" y="120"/>
      <operator activated="true" class="weka:W-M5P" compatibility="5.3.001" expanded="true" height="76" name="W-M5P" width="90" x="179" y="120"/>
      <operator activated="true" class="weka:W-REPTree" compatibility="5.3.001" expanded="true" height="76" name="W-REPTree" width="90" x="179" y="30"/>
      <connect from_op="Generate Data" from_port="output" to_op="Multiply" to_port="input"/>
      <connect from_op="Multiply" from_port="output 1" to_op="W-REPTree" to_port="training set"/>
      <connect from_op="Multiply" from_port="output 2" to_op="W-M5P" to_port="training set"/>
      <connect from_op="W-M5P" from_port="model" to_port="result 3"/>
      <connect from_op="W-REPTree" from_port="model" to_port="result 1"/>
      <connect from_op="W-REPTree" from_port="exampleSet" to_port="result 2"/>
      <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"/>
    </process>
  </operator>
</process>

For installing the Weka Extension, start RapidMiner, go to the Help menu, and then to the "Updates and Extensions (Marketplace)" submenu.

Cheers,
Ralf
Logged

Pages: [1]
  Print  
 
Jump to: