One-Class SVM node Fields

The Fields tab specifies which fields are used in the analysis.

Use predefined roles. Select this option to select all fields with a defined role of Input.

Use custom field assignments. To manually select fields, select this option and choose input fields and split fields:

Inputs. Select the input fields to use in the analysis. All storage types and measurement types are supported, except for typeless or unknown. If a field has a String storage type, the values of this field will be binarized in a one-vs-all fashion via a one-hot encoding algorithm.

Split. Select which field or fields to use as split fields. All flag, nominal, ordinal, and discrete measurement types are supported.

Use partitioned data If a partition field is defined, this option ensures that data from only the training partition is used to build the model.