Classification Node Models

Lists the models that are built when the node runs The default is build classification models using three different algorithms (SVM, NB, and DT). If the target is binary, the default is to build four models using SVM, NDB, DT, and GLM.

For multi-class targets, you can also specify the GLM algorithm if you add a model.

Model Setting lists the models that you want to build.

You can search the list for a particular model.

You can add, delete, or view models in the list. You can compare test results, if the models were tested. you can tune models is the model was tested correctly. You can also indicate which models are passed to subsequent nodes or not, as described in Classification Node Output Column.

To add a model, launch Add Model (Classification) by clicking Add model.

To delete one or more models, select them and click

delete selected

If models were tested you can compare test results by selecting two or more models and clicking

compare test results for selected models.

If a model built successfully, you can view the model (launch the model viewer) by selecting the model and clicking

view selected items

The model viewer is launched. The model viewer depends on the algorithm used to create the model; for more information about a viewer, see

You can tune classification models as described in Tuning Classification Models.