Association Rule
A PMML association rule model represents rules where some set of items is associated to another set of items. For example a rule can express that a certain product or set of products is often bought in combination with a certain set of other products, also known as Market Basket Analysis. An Association Rule model typically has two variables: one for grouping records together into transactions and another that uniquely identifies each record.
Model Element
<AssociationModel functionName="associationRules" ...Unsupported Features
Non-string field types for the field(s) identifying the item are not supported.
Having more than one field for grouping records is not supported.
Association Rule models with the <MiningSchema> element containing a reference to a <DerivedField> element are not supported.
Model Outputs
| Supported Model Output Features | Description |
|---|---|
|
transformedValue |
A value generated via a transformation expression applied to the predicted model output. |
|
decision |
A value generated via an expression applied to the predicted model output resulting in a categorized value. |
|
entityId |
The id of the winning rule (default), or the rule specified by the rank value. If the selected rule does not provide an id, a 1-based index is returned. |
|
ruleId |
This is identical to the entityId option and has been deprecated as of PMML 4.2. Although its use is currently supported it is recommended to use entityId. |
|
affinity |
The affinity of the winning rule (default), or the rule specified by the rank value. |
|
antecedent |
The antecedent of the winning rule (default), or the rule specified by the rank value. This output will be formatted as a comma separated string of values. |
|
consequent |
The consequent of the winning rule (default), or the rule specified by the rank value. This output will be formatted as a comma separated string of values. |
|
rule |
The winning rule (default), or the rule specified by the rank value. This output will return a description of the rule, formatted in the following way: {antecedent}->{consequent}. |
|
confidence |
The confidence of the winning rule (default), or the rule specified by the rank value. |
|
support |
The support of the winning rule (default), or the rule specified by the rank value. |
|
lift |
The lift of the winning rule (default), or the rule specified by the rank value. |
|
leverage |
The leverage of the winning rule (default), or the rule specified by the rank value. |