About questions though… we still can’t just boil them down to GetLinks and PutLinks, because those do direct evaluation and sometimes questions should be answered via inference
It seems we want to map a question into the set of possible answers, and then link the question to “ConceptNode Interrogative” or whatever,
where the semantics of Interrogative is “the instruction or inclination to find out what is inside the set at hand”
So if we have “What do dogs eat?”
then we have
InheritanceLink
SatisfyingSet
EvaluationLink
PredicateNode “eat”
ListLink
ConceptNode “dog”
Variable $x
ConceptNode “interrogativeness_77”
InheritanceLInk
ConceptNode “interrogativeness_77"
ConceptNode “Interrogative
and then the meaning of “Interrogative” is that someone wants you to do inference or direct evaluation or something to find out what lies inside that SatisfyingSet…
This very nice paper
https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20050082005.pdf
defines a question in terms of the set of statements that answer it…
About questions though… we still can’t just boil them down to GetLinks and PutLinks, because those do direct evaluation and sometimes questions should be answered via inference
It seems we want to map a question into the set of possible answers, and then link the question to “ConceptNode Interrogative” or whatever,
where the semantics of Interrogative is “the instruction or inclination to find out what is inside the set at hand”
So if we have “What do dogs eat?”
then we have
InheritanceLink
SatisfyingSet
EvaluationLink
PredicateNode “eat”
ListLink
ConceptNode “dog”
Variable $x
ConceptNode “interrogativeness_77”
InheritanceLInk
ConceptNode “interrogativeness_77"
ConceptNode “Interrogative
and then the meaning of “Interrogative” is that someone wants you to do inference or direct evaluation or something to find out what lies inside that SatisfyingSet…
This very nice paper
https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20050082005.pdf
defines a question in terms of the set of statements that answer it…