Creating and Editing a Mining Task

A mining task is the mechanism that searches for the collection of rules that make up a data model. Alongside the search criteria defined in the selected template, a task also defines the target (the actual question that motivated the analysis, such as how many customers are likely to respond to a mailing), and it identifies the datasets to be used. The goal of a mining task is to search for the best possible models.

Create a mining task

To create a mining task:

  1. Select the segment from which you want to mine additional segment conditions.
  2. Click Settings. The Create/Edit Mining Task dialog opens. This dialog provides options for defining the mining task.
  3. Make any necessary changes and click OK to return to the working model pane. Decision List Viewer uses the settings as the defaults to run for each task until an alternative task or settings is selected.
  4. Click Find Segments to start the mining task on the selected segment.

Edit a mining task

The Create/Edit Mining Task dialog provides options for defining a new mining task or editing an existing one.

Most parameters available for mining tasks are similar to those offered in the Decision List node. The exceptions are shown below. See the topic Decision List Model Options for more information.

Load Settings: When you have created more than one mining task, select the required task.

New... Click to create a new mining task based on the settings of the task currently displayed.

See the topic New Settings for more information.

Target

Target Field: Represents the field that you want to predict, whose value is assumed to be related to the values of other fields (the predictors).

Target value. Specifies the value of the target field that indicates the outcome you want to model. For example, if the target field churn is coded 0 = no and 1 = yes, specify 1 to identify rules that indicate which records are likely to churn.

Simple Settings

Maximum number of alternatives. Specifies the number of alternatives that will be displayed upon running the mining task. The minimum allowed setting is 1; there is no maximum setting.

Expert Settings

Edit... Opens the Edit Advanced Parameters dialog that enables you to define the advanced settings. See the topic Edit Advanced Parameters for more information.

Data

Build selection. Provides options for specifying the evaluation measure that Decision List Viewer should analyze to find new rules. The listed evaluation measures are created/edited in the Organize Data Selections dialog.

Available fields. Provides options for displaying all fields or manually selecting which fields to display.

Edit... If the Custom option is selected, this opens the Customize Available Fields dialog that enables you to select which fields are available as segment attributes found by the mining task. See the topic Customize Available Fields for more information.