Synopsis
This operator applies given association rules on an example set.
Description
This operator creates a new confidence attribute for each item occurring in at least one conclusion of an association rule. Then it checks for each example and for each rule, if the example fulfills the premise of the rule. An example fulfills a premise, if the example covers all items in the premise. An example covers an item, if the attribute representing the item contains the positive value. If the check is positive, a confidence value for each item in the conclusion is derived. Which value is used, depends on the selected confidence aggregation method. There are two types: The binary choice will set a 1, for any item contained inside a fulfilled rule's conclusion. This is independent of how confident the rule was. Any aggregation choice will select the maximum of the previous and the new value of the selected confidence function.
Input
- example set: expects: ExampleSet
- association rules: expects: AssociationRules
Output
- example set:
Parameters
- confidence aggregation method: This selects the method to aggregat the confidence on the items in each fulfilled conclusion.
- positive value: This parameter determines, which value of the binominal attributes is treated as positive. Attributes with that value are considered as part of a transaction. If left blank, the example set determines, which is value is used.