artificial intelligence - Representing Natural Language as RDF -
How capable is the representation of RDF / OWL in the concepts given in natural language? I am still learning RDF and other semantic technologies, but as I have understood it at present, the information is usually shown as a triple form (subject, forecast, object). So I can imagine that can be used to represent the "punishment of Bob's hat" However, you mall after the approval of "Bob, how would represent a much more complex sentences like over 42nd Street, Owner Will be a job in "? Tag conventions to represent nouns / verbs / ownership / causation / time / etc?
Note, I'm not asking how to automatically convert arbitrary natural language text for RDF (as this appears to be currently impossible). I am just trying to understand how RDF can be used to represent the same information representing the natural language.
It may be that the order is to define a piece of English, One part of this effort can be mapped to the mapping of Oval 2 DL. See example.
At the Boss 42nd Street, after the owner's approval, there will be a job in the mall
as a controlled natural English
(Ace) Can be rewritten in the owner of the mall, whose address is "42nd road" then it is approved by the mall.
(Or something similar, depending on what you really want to say.)
This sentence should be automatically mapped to an OWL2 can SubClassOf -axiom
SubClassOf (ObjectIntersectionOf (ObjectOneOf (: Mall) ObjectSomeValuesFrom (: owner ObjectSomeValuesFrom (: approval ObjectIntersectionOf (ObjectOneOf (John) DataHasValue (: address " 42nd street "^^ & LT; http: //www.w3.org/2001/XMLSchema#string>))))) ObjectSomeValuesFrom (: ObjectOneOf (employment: John)))
< P> This mapping applies some traditions about basic word classes Is: - Map names OWL class common names
- proper names map to the WL personal name
- verb, transitive verb , And of -constructions for the name of the property property owls: If the data names the property, then their argument is a number or string, the name of the object property is otherwise
in general, English its building blocks (noun, adjective different types, different types of actions, ...) o Very rich in comparison to the OWL ( Classes, individuals, objects and data properties, and (typed) data items such as strings and numbers). And this is just the "word versus institution" level. Things like stress are more complex because they have many surfaces in English Representation is there and no one is implied by OWL.
Comments
Post a Comment