Adaptive Bayes 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.

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. 

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.

Model Type

You can choose from three different modes for building the model.

• Multi-feature. Builds and compares a number of models, including an NB model and single and multifeature product probability models. This is the most exhaustive mode and typically takes the longest to compute as a result. Rules are produced only if the single feature model turns out to be best. If a multifeature or NB model is chosen, no rules are produced.
• Single-feature. Creates a simplified decision tree based on a set of rules. Each rule contains a condition together with probabilities associated with each outcome. The rules are mutually exclusive and are provided in a format that can be read by humans, which may be a significant advantage over Naive Bayes and multifeature models.
• Naive Bayes. Builds a single NB model and compares it with the global sample prior (the distribution of target values in the global sample). The NB model is produced as output only if it turns out to be a better predictor of the target values than the global prior. Otherwise, no model is produced as output.