[This is preliminary documentation and is subject to change.]
Indicates whether or not a rule should be extracted from the Q-learning backpropagation network
Namespace: Clarion.FrameworkAssembly: CLARIONLibrary (in CLARIONLibrary.dll) Version: 6.1.0.7 (6.1.0.7)
Syntax
C# |
---|
public override bool CheckExtraction() |
Return Value
True if a rule should be extracted, otherwise FalseImplements
IExtractsRules..::..CheckExtraction()()()()
Remarks
This method is called internally by the system before attempting to extract a rule. By default, it uses the Q-learning variant (as specified in the CLARION tutorial) for calculating if a rule should be extracted.
Users can specify their own, customized, method for extraction checking by either of the following methods:
- By overriding the method in a subclass. This is the better option when implementing a new, fully-defined, custom Q-learning backpropagation network
- By specifying an ExtractionChecker delegate during the initialization of a Q-learning backpropagation network. This option should be used when a user wishes to alter the extraction checking method of an already existing Q-learning backpropagation network
![]() |
---|
The extraction checking delegate MUST return True if a rule should be extracted or False if a rule should not be extracted |