Currently, in the ReasonerProcessor, we call the includeMetaKIs(...) method to determine whether to use ENTRY or NORMAL reasoning mode. However, this is wrong, because this should not depend on whether the meta KIs should be included, it should rather be depending on whether the KI isMeta(). However, if we do that, the MockedKnowledgeBase and KnowledgeNetwork usage fails because they use the meta Graph Pattern (but are not meta KIs), but still would not like to use NORMAL reasoning mode since this causes an OutOfMemoryException.
So, we need to think about how we want the reasoner configuration to work. Maybe we should introduce the levels ENTRY, NORMAL, ADVANCED, ULTIMATE and SUPREME in the knowledge-engine artifact. In the reasoner artifact we can only use the MatchFlags in the reasoner.
Currently, in the
ReasonerProcessor
, we call theincludeMetaKIs(...)
method to determine whether to useENTRY
orNORMAL
reasoning mode. However, this is wrong, because this should not depend on whether the meta KIs should be included, it should rather be depending on whether the KIisMeta()
. However, if we do that, theMockedKnowledgeBase
andKnowledgeNetwork
usage fails because they use the meta Graph Pattern (but are not meta KIs), but still would not like to useNORMAL
reasoning mode since this causes anOutOfMemoryException
.So, we need to think about how we want the reasoner configuration to work. Maybe we should introduce the levels
ENTRY
,NORMAL
,ADVANCED
,ULTIMATE
andSUPREME
in theknowledge-engine
artifact. In thereasoner
artifact we can only use theMatchFlag
s in the reasoner.