A former member
Post #: 1
Hey Sean,

After reading through Chapter 2 of PLN again, I have reconstructed the following BNF grammar for PLN:
Atom ::=first_order_relationship elementary_term elementary_term <truth_value> | higher_order_relationship higher_order_term higher_order_term <truth_value>
elementary_term ::= Individual | Class
fuzzy_set_membership ::= Member Individual Class <truth_value>
first_order_relationship ::= Member | IntensionalInheritance | Inheritance | Subset |ExtensionalSimilarity | Similarity | IntensionalSimilarity

higher_order_relationship ::= ExtensionalImplication | Implication | IntensionalImplication | ExtensionalEquivalence | Equivalence | IntensionalEquivalence

higher_order_term ::= Predicate | first_order_relationship | higher_order_relationship

Predicate ::= Evaluation Relation variables <truth_value> | SatisfyingSet (Member Relation)

truth_value ::= strength | SimpleTruthValue | IndefiniteTruthValue | DistributionalTruthValue

strength ::= [0..1] | 0, 1, 2, 3...

SimpleTruthValue ::= <strength, weight_of_evidence> | <strength, N>

weight_of_evidence ::= [0..1]

N ::= 0, 1, 2, 3...

IndefiniteTruthValue ::= <[L,U], credibility_level, lookahead>

0 <= L <= U <= 1.

credibility_level ::= [0..1]

lookahead ::= 0, 1, 2, 3,...



Sean
user 8162770
Cambridge, MA
Post #: 3
Great! I suppose my next questions would be: What is the list of inference rules? And what is a problem that reveals the power of this formalism?
A former member
Post #: 2
What are you guys referencing with "PLN" acronym? I thought maybe it was "Probabilistic Learning Network", but found this paper uses "ProNet" instead.

Sorry for the pedestrian question, I'm trying to follow the conversation at a bit of a distance.
Sean
user 8162770
Cambridge, MA
Post #: 4
Hi Bob,

It's "Probabilistic Logic Network" (http://en.wikipedia.o...­.

- Sean
Powered by mvnForum

Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy