Constructing a Classification Model

The Classification Model is a nonlinear model, used for classification of targets, as opposed to prediction of targets.

The output of the Classification Model is a discrete data type, used to classify the target into discrete classes.

To construct a Classification Model:

On the Troubleshooting Project Bar, click on the modeling button. Note this button is only available after the previous steps have been completed.
From the Modeling view, scroll down to the Classification Model group, and click [Construct].
The configuring dialog for constructing the Classification Model will show, requesting you select fields as inputs to the model, as well as selecting a target field.
Use the (>) and (<) buttons to make your field selections.
- Select the field(s) to add and click on the add (>) button to add the field to the selection.
- Remove field(s) by selecting the a field and clicking on the remove (<) button.
- To add all fields to the selection, click on the add all (>>) button.
- To remove all fields from the selection, click on the remove all (<<) button.

NOTE:

Fields with more than 100 classes will not be selected, such as a "Date" field for example.

The target field selected must have discrete data, to be used as classification classes.

Select the data to be modeled, from [Use all data], [Use brushed data] and [Use non-brushed data]. The latter two options will only be available if you brushed data during Step 2: Preparation.
Click Ok when done.
Modeling will take place, and a Decision Tree will be generated.
The Classification Model can be viewed in the Knowledge Extraction View.

Reconstructing the Classification Model

This option allows the user to reconstruct the Classification Model with different inputs fields, a different target field or different brushing area.

Click on the [Reconstruct] button.
Reconfigure the data to be used for the Classification Model.
Click on Ok when done.