Using Modeler

Teradata® Analytic Apps User Guide

brand
Teradata Applications
prodname
Teradata Analytic Apps
vrm_release
1.0
category
User Guide
featnum
B035-3810-128K
See Teradata® Vantage Machine Learning Engine Analytic Function Reference, B700-4003 for additional information about Teradata Vantage MLE models and analytic functions.

    Prepare Datasets

  1. Create model input data.
  2. Set up the training and test models.
  3. Access Modeler

  4. Access AppCenter and select Vantage Modeler.
  5. Select the Vantage database and enter your database login credentials, then click SELECT.
  6. Define the Parameters and Run the Model

  7. Define the parameters for the model.
  8. Click RUN MODEL.
  9. Enter a unique prefix to identify the output tables.
    If you enter a prefix used for existing tables, those tables are overwritten.
  10. Click RUN MODEL.
  11. Analyze the Results

  12. When the predictive model completes running, review the results.
    Results Description
    Accuracy Reflects model accuracy using precision and recall values
    Confusion matrix Contrasts predicted values to actual values
    Lift Segments scoring results by decile, sorted by highest scoring decile
    Importance Lists events in order of importance to the model
  13. [Optional] Export Results

  14. Click EXPORT MODEL QUERIES.
    Exported results are saved to the database.
    1. Enter the Model Name.
    2. Enter a Prefix.
  15. Export Users from Lift Chart

  16. Export lift segment. Continuing with the example, you could export customers with highest score or propensity to a table and use the results to make targeted business decisions.