The most direct way to approach this sort of problem in OpenCog would
be just as a backward chainer query.... Nil is currently a few hours
away from having the backward chainer ready for you to try on these
sorts of problems...
Alternately, one could use ECAN, and one could write a function (say a
loop in a Scheme function) that repeatedly stimulates the Atom
A(X) :=
LambdaLink
EvaluationLink
PredicateNode "eats"
ListLink
VariableNode $X
ConceptNode "chicken"
with ShortTermImportance (STI) ...
Then one could create a Goal Atom of "learn new things" ... and then a
Psi implication saying, basically,
PredictiveImplication
finding X so that A(X)
fulfilling "learn new things" goal
This is somewhat elaborate for a simple thing, but would be a good
exercise in using the architecture...
-- Ben
> --
> You received this message because you are subscribed to the Google Groups
> "opencog" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to
opencog+u...@googlegroups.com.
> To post to this group, send email to
ope...@googlegroups.com.
> Visit this group at
https://groups.google.com/group/opencog.
> To view this discussion on the web visit
>
https://groups.google.com/d/msgid/opencog/8f34f57e-5c67-459b-9c09-74de4cb5a00b%40googlegroups.com.
> For more options, visit
https://groups.google.com/d/optout.
--
Ben Goertzel, PhD
http://goertzel.org
“I tell my students, when you go to these meetings, see what direction
everyone is headed, so you can go in the opposite direction. Don’t
polish the brass on the bandwagon.” – V. S. Ramachandran