Oracle SVM Model Options

Model name. You can generate the model name automatically based on the target or ID field (or model type in cases where no such field is specified) or specify a custom name.

Unique field. Specifies the field used to uniquely identify each case. For example, this might be an ID field, such as CustomerID. IBM® SPSS® Modeler imposes a restriction that this key field must be numeric.

Note: This field is optional for all Oracle nodes except Oracle Adaptive Bayes, Oracle O-Cluster and Oracle Apriori.

Auto Data Preparation. (11g only) Enables (default) or disables the automated data preparation mode of Oracle Data Mining. If this box is checked, ODM automatically performs the data transformations required by the algorithm. For more information, see Oracle Data Mining Concepts.

Active Learning. Provides a way to deal with large build sets. With active learning, the algorithm creates an initial model based on a small sample before applying it to the complete training dataset and then incrementally updates the sample and model based on the results. The cycle is repeated until the model converges on the training data or the maximum allowed number of support vectors is reached.

Kernel Function. Select Linear or Gaussian, or leave the default System Determined to allow the system to choose the most suitable kernel. Gaussian kernels are able to learn more complex relationships but generally take longer to compute. You may want to start with the linear kernel and try the Gaussian kernel only if the linear kernel fails to find a good fit. This is more likely to happen with a regression model, where the choice of kernel matters more. Also, note that SVM models built with the Gaussian kernel cannot be browsed in IBM SPSS Modeler. Models built with the linear kernel can be browsed in IBM SPSS Modeler in the same manner as standard regression models.

Normalization Method. Specifies the normalization method for continuous input and target fields. You can choose Z-Score, Min-Max, or None. Oracle performs normalization automatically if the Auto Data Preparation check box is selected. Uncheck this box to select the normalization method manually.