Explore Results Improvements

You can redo clustering to explore the improvements to results from cleaning the SYMPTOM_TEXT variable. Complete the following steps:
  1. Verify that the Cluster property settings for the Text Miner — CST node are the same as in previous examples.
  2. Right-click the Text Miner — CST node, and select Run. Click Yes in the Confirmation dialog box. Click OK in the Run Status dialog box when the node has finished running.
  3. Select the Selector Button button for the Interactive property in the Text Miner — CST node property panel to open the Interactive Results window. Look at the Clusters table.
  4. Compare these results to the first Text Miner node results from View Interactive Results. Does the clustering seem to have improved with the cleaned SYMPTOM_TEXT data?
  5. Close the Interactive Results window.
  6. Right-click the Segment Profile node, and select Copy from the menu. Right-click on an empty space in the diagram workspace, and select Paste from the menu.
  7. Connect the Text Miner — CST node to the Segment Profile (2) node.
    Process Flow Diagram
  8. Click the Selector Button for the Variables property for the Segment Profile (2) node to open the Variables — Prof2 window. Make sure that the PROB variables and the _SVD_ variables have a Use value of No.
  9. Click OK to save the variables settings and close the Variables — Prof2 window.
  10. Right-click the Segment Profile (2) node and select Run. Click Yes in the Confirmation dialog box.
  11. Click Results in the Run Status dialog box to open the Results window when the node has finished running. Note the significant relationships in the table. Do the relationships appear clearer with the cleaned text than they did with the uncleaned text?
  12. Close the Results window.
Copyright © SAS Institute Inc. All rights reserved.