Apply the Best Model to New Data

tell me more icon Tell me more

Now apply the best table to new data, that is, use the model for predictions.

  1. Specify the apply data: Create a Data Source Node on the workflow; select the view MINING_DATA_APPLY_V.

  2. Open the Evaluate and Apply section of the Components Palette. Drag an Apply node to the workflow; name the node PredictHigh.

  3. Link ClassBuild to PredictHigh, the apply node; also link MINING_DATA_APPLY_V to PredictHigh.

    tell me more icon Tell me more about what I see in the GUI

  4. Double-click PredictHigh to open the Apply editor. By default the apply output consists of the prediction and the probability of the prediction and nothing else.

    To add the CUST_ID column to the Apply output to identify the prediction, follow these steps:

    1. Click the Data Columns tab.
    2. Click add to list.
    3. Move CUST_ID from Available Attributes to Selected Attributes.
    4. Click OK; click OK again.

      tell me more icon Tell me more about what I see in the GUI

  5. To apply the selected NB model to the new data, right-click PredictHigh and select Run.

  6. Monitor the apply in the Task Manager.

  7. The NB model is applied to the new data. By default, the apply is performed with automatic settings; CUST_ID is passed through. For details, see Apply Node.

tell me more icon Tell me more about what I see in the GUI

The Next step is to create a table containing the apply results.