API tools faq
paste
Advertisement
Guest User

Weight of Evidence Transformation - RapidMiner

a guest
Jul 29th, 2015
209
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
XML 18.99 KB | None | 0 0
raw download clone embed print report
  1. <?xml version="1.0" encoding="UTF-8" standalone="no"?>
  2. <process version="6.4.000">
  3.   <context>
  4.     <input/>
  5.     <output/>
  6.     <macros/>
  7.   </context>
  8.   <operator activated="true" class="process" compatibility="6.4.000" expanded="true" name="Process">
  9.     <process expanded="true">
  10.       <operator activated="true" class="retrieve" compatibility="6.4.000" expanded="true" height="60" name="Get Data" width="90" x="45" y="30">
  11.         <parameter key="repository_entry" value="../data/large_set"/>
  12.       </operator>
  13.       <operator activated="true" class="discretize_by_entropy" compatibility="6.4.000" expanded="true" height="94" name="Discretize" width="90" x="179" y="30">
  14.         <description align="center" color="transparent" colored="false" width="126">Testing Entropy Discretization.</description>
  15.       </operator>
  16.       <operator activated="true" class="generate_attributes" compatibility="6.4.000" expanded="true" height="76" name="Create Pos &amp; Neg labels as attributes" width="90" x="380" y="30">
  17.         <list key="function_descriptions">
  18.           <parameter key="label_pos" value="if(LABELFIELD == &quot;1.0&quot;,1,0)"/>
  19.           <parameter key="label_neg" value="if(LABELFIELD == &quot;0.0&quot;,1,0)"/>
  20.         </list>
  21.         <description align="center" color="transparent" colored="false" width="126">If not created already create two attributes holding values for positive and negative labels.</description>
  22.       </operator>
  23.       <operator activated="true" class="filter_examples" compatibility="6.4.000" expanded="true" height="94" name="Filter Examples" width="90" x="514" y="30">
  24.         <parameter key="condition_class" value="all"/>
  25.         <list key="filters_list">
  26.           <parameter key="filters_entry_key" value="CTRYCD.equals."/>
  27.         </list>
  28.         <description align="center" color="transparent" colored="false" width="126">Use this filter to remove example values you do not want to appear in the WoE calculations. For example, values that deliver infinite results.</description>
  29.       </operator>
  30.       <operator activated="true" class="set_macros" compatibility="6.4.000" expanded="true" height="76" name="Set Macros for Loop Att" width="90" x="648" y="30">
  31.         <list key="macros">
  32.           <parameter key="label_pos" value="label_pos"/>
  33.           <parameter key="label_neg" value="label_neg"/>
  34.         </list>
  35.         <description align="center" color="transparent" colored="false" width="126">Set the macro values for your positive and negative labels.</description>
  36.       </operator>
  37.       <operator activated="true" class="subprocess" compatibility="6.4.000" expanded="true" height="112" name="Weight of Evidence Calculator" width="90" x="782" y="30">
  38.         <process expanded="true">
  39.           <operator activated="true" class="loop_attributes" compatibility="6.4.000" expanded="true" height="112" name="WoE Table" width="90" x="45" y="30">
  40.             <parameter key="attribute_filter_type" value="value_type"/>
  41.             <parameter key="value_type" value="nominal"/>
  42.             <parameter key="iteration_macro" value="woeAttribute"/>
  43.             <process expanded="true">
  44.               <operator activated="true" class="select_attributes" compatibility="6.4.000" expanded="true" height="76" name="Select Attributes" width="90" x="45" y="30">
  45.                 <parameter key="attribute_filter_type" value="subset"/>
  46.                 <parameter key="attributes" value="%{woeAttribute}|%{label_pos}|%{label_neg}"/>
  47.                 <parameter key="include_special_attributes" value="true"/>
  48.               </operator>
  49.               <operator activated="true" class="subprocess" compatibility="6.4.000" expanded="true" height="76" name="WoE Steps" width="90" x="179" y="120">
  50.                 <process expanded="true">
  51.                   <operator activated="true" class="rename" compatibility="6.4.000" expanded="true" height="76" name="Rename label classes" width="90" x="45" y="30">
  52.                     <parameter key="old_name" value="%{label_pos}"/>
  53.                     <parameter key="new_name" value="label_pos"/>
  54.                     <list key="rename_additional_attributes">
  55.                       <parameter key="%{label_neg}" value="label_neg"/>
  56.                     </list>
  57.                     <description align="center" color="transparent" colored="false" width="126">Rename label classes to label_pos &amp;amp; label_neg</description>
  58.                   </operator>
  59.                   <operator activated="true" class="aggregate" compatibility="6.4.000" expanded="true" height="76" name="ForceAgg" width="90" x="179" y="30">
  60.                     <list key="aggregation_attributes">
  61.                       <parameter key="label_neg" value="sum"/>
  62.                       <parameter key="label_pos" value="sum"/>
  63.                     </list>
  64.                     <parameter key="group_by_attributes" value="%{woeAttribute}"/>
  65.                   </operator>
  66.                   <operator activated="true" class="rename" compatibility="6.4.000" expanded="true" height="76" name="Rename (2)" width="90" x="313" y="30">
  67.                     <parameter key="old_name" value="sum(label_pos)"/>
  68.                     <parameter key="new_name" value="label_pos"/>
  69.                     <list key="rename_additional_attributes">
  70.                       <parameter key="sum(label_neg)" value="label_neg"/>
  71.                     </list>
  72.                   </operator>
  73.                   <operator activated="true" class="subprocess" compatibility="6.4.000" expanded="true" height="76" name="WoE Calculation" width="90" x="447" y="30">
  74.                     <process expanded="true">
  75.                       <operator activated="true" class="generate_id" compatibility="6.4.000" expanded="true" height="76" name="Generate ID" width="90" x="45" y="30">
  76.                         <description align="center" color="transparent" colored="false" width="126">target attribute made into IDs to allow for easy join following WoE calculation.</description>
  77.                       </operator>
  78.                       <operator activated="true" class="multiply" compatibility="6.4.000" expanded="true" height="94" name="Multiply" width="90" x="179" y="30"/>
  79.                       <operator activated="true" class="aggregate" compatibility="6.4.000" expanded="true" height="76" name="Aggregate (2)" width="90" x="447" y="75">
  80.                         <list key="aggregation_attributes">
  81.                           <parameter key="label_neg" value="sum (fractional)"/>
  82.                           <parameter key="label_pos" value="sum (fractional)"/>
  83.                         </list>
  84.                         <parameter key="group_by_attributes" value="id"/>
  85.                       </operator>
  86.                       <operator activated="true" class="rename_by_replacing" compatibility="6.4.000" expanded="true" height="76" name="tidy names 1" width="90" x="581" y="75">
  87.                         <parameter key="replace_what" value="fractional_sum"/>
  88.                         <parameter key="replace_by" value="distribution_"/>
  89.                       </operator>
  90.                       <operator activated="true" class="rename_by_replacing" compatibility="6.4.000" expanded="true" height="76" name="tidy names 2" width="90" x="715" y="75"/>
  91.                       <operator activated="true" class="generate_attributes" compatibility="6.4.000" expanded="true" height="76" name="WoE Generation" width="90" x="849" y="75">
  92.                         <list key="function_descriptions">
  93.                           <parameter key="WoE" value="ln(distribution_label_pos/distribution_label_neg)"/>
  94.                           <parameter key="%{woeAttribute}_MarginalContribution" value="(distribution_label_pos-distribution_label_neg)*WoE"/>
  95.                         </list>
  96.                       </operator>
  97.                       <operator activated="true" class="set_role" compatibility="6.4.000" expanded="true" height="76" name="Set Role" width="90" x="983" y="75">
  98.                         <parameter key="attribute_name" value="id"/>
  99.                         <parameter key="target_role" value="id"/>
  100.                         <list key="set_additional_roles"/>
  101.                         <description align="center" color="transparent" colored="false" width="126">Set id as id</description>
  102.                       </operator>
  103.                       <operator activated="true" class="rename" compatibility="6.4.000" expanded="true" height="76" name="Rename WoE" width="90" x="1117" y="75">
  104.                         <parameter key="old_name" value="WoE"/>
  105.                         <parameter key="new_name" value="%{woeAttribute}_WoE"/>
  106.                         <list key="rename_additional_attributes"/>
  107.                         <description align="center" color="transparent" colored="false" width="126">Rename here rather than calculation to ensure bad attribute names do not cause errors in the process.</description>
  108.                       </operator>
  109.                       <operator activated="true" class="select_attributes" compatibility="6.4.000" expanded="true" height="76" name="Select Attributes (2)" width="90" x="1251" y="75">
  110.                         <parameter key="attribute_filter_type" value="subset"/>
  111.                         <parameter key="attributes" value="distribution_label_neg|distribution_label_pos"/>
  112.                         <parameter key="invert_selection" value="true"/>
  113.                         <description align="center" color="transparent" colored="false" width="126">Remove unnecessary attributes</description>
  114.                       </operator>
  115.                       <operator activated="true" class="join" compatibility="6.4.000" expanded="true" height="76" name="WoE Table (2)" width="90" x="1385" y="30">
  116.                         <list key="key_attributes"/>
  117.                       </operator>
  118.                       <connect from_port="in 1" to_op="Generate ID" to_port="example set input"/>
  119.                       <connect from_op="Generate ID" from_port="example set output" to_op="Multiply" to_port="input"/>
  120.                       <connect from_op="Multiply" from_port="output 1" to_op="WoE Table (2)" to_port="left"/>
  121.                       <connect from_op="Multiply" from_port="output 2" to_op="Aggregate (2)" to_port="example set input"/>
  122.                       <connect from_op="Aggregate (2)" from_port="example set output" to_op="tidy names 1" to_port="example set input"/>
  123.                       <connect from_op="tidy names 1" from_port="example set output" to_op="tidy names 2" to_port="example set input"/>
  124.                       <connect from_op="tidy names 2" from_port="example set output" to_op="WoE Generation" to_port="example set input"/>
  125.                       <connect from_op="WoE Generation" from_port="example set output" to_op="Set Role" to_port="example set input"/>
  126.                       <connect from_op="Set Role" from_port="example set output" to_op="Rename WoE" to_port="example set input"/>
  127.                       <connect from_op="Rename WoE" from_port="example set output" to_op="Select Attributes (2)" to_port="example set input"/>
  128.                       <connect from_op="Select Attributes (2)" from_port="example set output" to_op="WoE Table (2)" to_port="right"/>
  129.                       <connect from_op="WoE Table (2)" from_port="join" to_port="out 1"/>
  130.                       <portSpacing port="source_in 1" spacing="0"/>
  131.                       <portSpacing port="source_in 2" spacing="0"/>
  132.                       <portSpacing port="sink_out 1" spacing="0"/>
  133.                       <portSpacing port="sink_out 2" spacing="0"/>
  134.                     </process>
  135.                     <description align="center" color="transparent" colored="false" width="126">Requirements.&lt;br&gt;Dataset containing target variable name (define by macro), a label.</description>
  136.                   </operator>
  137.                   <connect from_port="in 1" to_op="Rename label classes" to_port="example set input"/>
  138.                   <connect from_op="Rename label classes" from_port="example set output" to_op="ForceAgg" to_port="example set input"/>
  139.                   <connect from_op="ForceAgg" from_port="example set output" to_op="Rename (2)" to_port="example set input"/>
  140.                   <connect from_op="Rename (2)" from_port="example set output" to_op="WoE Calculation" to_port="in 1"/>
  141.                   <connect from_op="WoE Calculation" from_port="out 1" to_port="out 1"/>
  142.                   <portSpacing port="source_in 1" spacing="0"/>
  143.                   <portSpacing port="source_in 2" spacing="0"/>
  144.                   <portSpacing port="sink_out 1" spacing="0"/>
  145.                   <portSpacing port="sink_out 2" spacing="0"/>
  146.                 </process>
  147.                 <description align="center" color="transparent" colored="false" width="126">This process takes a dataset of the form:&lt;br&gt;&lt;br&gt;[ attribute | class_pos | class_neg ]&lt;br&gt;and calculates the Weight of Evidence &amp;amp; Information Value for this attribute. This is a process that would normally be called inside a loop attributes and with additional aggregation steps applied prior.&lt;br&gt;</description>
  148.               </operator>
  149.               <operator activated="true" class="aggregate" compatibility="6.4.000" expanded="true" height="76" name="Aggregate" width="90" x="380" y="120">
  150.                 <list key="aggregation_attributes">
  151.                   <parameter key="%{woeAttribute}_MarginalContribution" value="sum"/>
  152.                 </list>
  153.               </operator>
  154.               <operator activated="true" class="subprocess" compatibility="6.4.000" expanded="true" height="76" name="Tidy up" width="90" x="581" y="75">
  155.                 <process expanded="true">
  156.                   <operator activated="true" class="rename_by_replacing" compatibility="6.4.000" expanded="true" height="76" name="Rename by Replacing" width="90" x="45" y="30">
  157.                     <parameter key="replace_what" value="(^.*)"/>
  158.                     <parameter key="replace_by" value="%{woeAttribute}"/>
  159.                   </operator>
  160.                   <operator activated="true" class="transpose" compatibility="6.4.000" expanded="true" height="76" name="Transpose" width="90" x="179" y="30"/>
  161.                   <operator activated="true" class="rename" compatibility="6.4.000" expanded="true" height="76" name="tidy att1 name" width="90" x="313" y="30">
  162.                     <parameter key="old_name" value="att_1"/>
  163.                     <parameter key="new_name" value="InformationValue"/>
  164.                     <list key="rename_additional_attributes"/>
  165.                   </operator>
  166.                   <connect from_port="in 1" to_op="Rename by Replacing" to_port="example set input"/>
  167.                   <connect from_op="Rename by Replacing" from_port="example set output" to_op="Transpose" to_port="example set input"/>
  168.                   <connect from_op="Transpose" from_port="example set output" to_op="tidy att1 name" to_port="example set input"/>
  169.                   <connect from_op="tidy att1 name" from_port="example set output" to_port="out 1"/>
  170.                   <portSpacing port="source_in 1" spacing="0"/>
  171.                   <portSpacing port="source_in 2" spacing="0"/>
  172.                   <portSpacing port="sink_out 1" spacing="0"/>
  173.                   <portSpacing port="sink_out 2" spacing="0"/>
  174.                 </process>
  175.               </operator>
  176.               <connect from_port="example set" to_op="Select Attributes" to_port="example set input"/>
  177.               <connect from_op="Select Attributes" from_port="example set output" to_op="WoE Steps" to_port="in 1"/>
  178.               <connect from_op="Select Attributes" from_port="original" to_port="example set"/>
  179.               <connect from_op="WoE Steps" from_port="out 1" to_op="Aggregate" to_port="example set input"/>
  180.               <connect from_op="Aggregate" from_port="example set output" to_op="Tidy up" to_port="in 1"/>
  181.               <connect from_op="Aggregate" from_port="original" to_port="result 2"/>
  182.               <connect from_op="Tidy up" from_port="out 1" to_port="result 1"/>
  183.               <portSpacing port="source_example set" spacing="0"/>
  184.               <portSpacing port="sink_example set" spacing="0"/>
  185.               <portSpacing port="sink_result 1" spacing="0"/>
  186.               <portSpacing port="sink_result 2" spacing="0"/>
  187.               <portSpacing port="sink_result 3" spacing="0"/>
  188.             </process>
  189.             <description align="center" color="transparent" colored="false" width="126">This will work for every nominal attribute in your exampleset.</description>
  190.           </operator>
  191.           <operator activated="true" class="append" compatibility="6.4.000" expanded="true" height="76" name="InformationValues" width="90" x="380" y="75">
  192.             <description align="center" color="transparent" colored="false" width="126">Appends IV into a table</description>
  193.           </operator>
  194.           <connect from_port="in 1" to_op="WoE Table" to_port="example set"/>
  195.           <connect from_op="WoE Table" from_port="example set" to_port="out 1"/>
  196.           <connect from_op="WoE Table" from_port="result 1" to_op="InformationValues" to_port="example set 1"/>
  197.           <connect from_op="WoE Table" from_port="result 2" to_port="out 3"/>
  198.           <connect from_op="InformationValues" from_port="merged set" to_port="out 2"/>
  199.           <portSpacing port="source_in 1" spacing="0"/>
  200.           <portSpacing port="source_in 2" spacing="0"/>
  201.           <portSpacing port="sink_out 1" spacing="0"/>
  202.           <portSpacing port="sink_out 2" spacing="54"/>
  203.           <portSpacing port="sink_out 3" spacing="126"/>
  204.           <portSpacing port="sink_out 4" spacing="0"/>
  205.         </process>
  206.       </operator>
  207.       <connect from_op="Get Data" from_port="output" to_op="Discretize" to_port="example set input"/>
  208.       <connect from_op="Discretize" from_port="example set output" to_op="Create Pos &amp; Neg labels as attributes" to_port="example set input"/>
  209.       <connect from_op="Create Pos &amp; Neg labels as attributes" from_port="example set output" to_op="Filter Examples" to_port="example set input"/>
  210.       <connect from_op="Filter Examples" from_port="example set output" to_op="Set Macros for Loop Att" to_port="through 1"/>
  211.       <connect from_op="Set Macros for Loop Att" from_port="through 1" to_op="Weight of Evidence Calculator" to_port="in 1"/>
  212.       <connect from_op="Weight of Evidence Calculator" from_port="out 1" to_port="result 1"/>
  213.       <connect from_op="Weight of Evidence Calculator" from_port="out 2" to_port="result 2"/>
  214.       <connect from_op="Weight of Evidence Calculator" from_port="out 3" to_port="result 3"/>
  215.       <portSpacing port="source_input 1" spacing="0"/>
  216.       <portSpacing port="sink_result 1" spacing="36"/>
  217.       <portSpacing port="sink_result 2" spacing="0"/>
  218.       <portSpacing port="sink_result 3" spacing="0"/>
  219.       <portSpacing port="sink_result 4" spacing="0"/>
  220.       <description align="center" color="yellow" colored="false" height="415" resized="true" width="757" x="10" y="10"/>
  221.       <description align="center" color="red" colored="true" height="105" resized="false" width="180" x="10" y="303">Make edits in the yellow area.</description>
  222.     </process>
  223.   </operator>
  224. </process>
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement
Public Pastes
Advertisement
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!