SVM Model Nugget
The SVM model creates a number of new fields. The most important of these is the $S-fieldname field, which shows the target field value predicted by the model.
The number and names of the new fields created by the model depend on the measurement level of the target field (this field is indicated in the following tables by fieldname).
To see these fields and their values, add a Table node to the SVM model nugget and execute the Table node.
| New field name | Description |
|---|---|
| $S-fieldname | Predicted value of target field. |
| $SP-fieldname | Probability of predicted value. |
| $SP-value | Probability of each possible value of nominal or flag (displayed only if Append all probabilities is checked on the Settings tab of the model nugget). |
| $SRP-value | (Flag targets only) Raw (SRP) and adjusted (SAP) propensity scores, indicating the likelihood of a "true" outcome for the target field. These scores are displayed only if the corresponding check boxes are selected on the Analyze tab of the SVM modeling node before the model is generated. See the topic Modeling Node Analyze Options for more information. |
| $SAP-value |
| New field name | Description |
|---|---|
| $S-fieldname | Predicted value of target field. |
Predictor Importance
Optionally, a chart that indicates the relative importance of each predictor in estimating the model may also be displayed on the Model tab. Typically you will want to focus your modeling efforts on the predictors that matter most and consider dropping or ignoring those that matter least. Note this chart is only available if Calculate predictor importance is selected on the Analyze tab before generating the model. See the topic Predictor Importance for more information.
Note: Predictor importance may take longer to calculate for SVM than for other types of models, and is not selected on the Analyze tab by default. Selecting this option may slow performance, particularly with large datasets.