- From: Peter <meancity@gmail.com>
- Date: Tue, 28 Sep 2004 12:15:00 +0800
- To: Harry Halpin <hhalpin@ibiblio.org>
- Cc: www-rdf-interest@w3.org
Dear Harry, thank u, For triple relation , i think i can understand, and in RDF specification 1999 they have Subject->Preidcate->Object this kind of triple relation. well, these days, i read some paper in 2002 IEEE intelligent systems named "DAML+OIL: An Ontology Lanuage for the Semantic Web" for some survey study. in that paper it mentions how to map DAML-OIL to FOL. but i am confused when i read that part. it shows following steps 1) translate DAML-OIL knowledge base from its concrete syntax into collection of RDF statement; well, i am not really sure what is so-called "concrete syntax" 2)Translate RDF statement with Property P, subject S and Object O. into FOL sentense of the form (PropertyValue, P S O) where PropertyVaule denotes a ternary relation that relates a property and an entity(a subject) to a value(an object) that they property has for that entity. i can not really understand this ... so maybe u can enlighten me on this. 3)Add the axioms that constrain the allowable interpretation of the properties, classes and constraints include in RDF, RDFS adn Daml-OIL. ... so i am very clear of step 1 and 2, so .. step 3 is ... more difficult for me to understand ... On Mon, 27 Sep 2004 12:59:02 -0400 (EDT), Harry Halpin <hhalpin@ibiblio.org> wrote: > Peter, > Note that FOL is a higher-subset of DL (description logics), DL > being chosen for its ability to remain decidable and having tractable > complexity (i.e. if you send a query you get it back). However, DL is a > bit more constrained, especially OWL/RDF DL. > > In FOL book has the color red would be: > > Elambda x, lambda y,(book(x), has(x,y), red(y), colour(red(y)) > which makes some assumptions about handling red and color. Not also the > exists over lambda x and y, which meanss we have only ONE book with ONE > color, not ALL Books. > > Coding it adjectives is tricky, but I'd guess you mean that "red is a color". It > might be different if you said "The book is bright red". > > There's a million variants of FOL but that gets you the idea. And god > knows a million ways (type/token) distinctions that get blurry, never > mind the adjectives and subclass relations. > > In DL it's less clear. > You have to phrase it as triples: > subject: book > predicate: has > object: red > The class red is string value of the class color. > Note you could have made it a class red is a subclass of color, but > let's keep it simple. > > So you could say (using N3 Notation) > http://www.w3.org/2000/10/swap/Primer.html > > So you want do divide things into classes and indivduals. > Books are a class of thing, colors are a class, and red > is a kind of color denoted by the string "red". > > In the land of N3 Notation.... > > :book :hasColor "red" > > which in full bloody glory should be qualified thus: > > @prefix :<http://www.example.org/> > @prefix rdfs: <http://www.w3.org/2000/01/rdf-schema> > :ColorableThing a rdfs:Class > :book a rdfs:Class > :hasColor rdfs:domain :ColorableThing > rdfs:range rdfs:Literal > > which could then be transformed into RDF/XML via: > > <rdf:RDF > xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" > xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#" > xmlns:example="www.example.org" > > <rdf:Description rdf:about="example:book"> > <example:hasColor>red</example:hasColor> > </rdf:Description> > > <rdfs:Class rdf:about="example:book" /> > <rdfs:Class rdf:about="example:ColorableThing" /> > <rdfs:Property rdf:about="example:hasColor> > <rdfs:domain="example:ColorableThing" /> > <rdfs:range="rdfs:Literal /> > </rdfs:Property> > > </rdf:RDF> > > You could do without the colorable thing idea, but why not? You might > want to color some things, and not color others, such as the "idea > of loyalty". > > Read: > http://www.w3.org/TR/rdf-primer/ > > Cheers, > > -Harry > > > > On Mon, 27 Sep 2004, Peter wrote: > > > > > dear all, > > > > given a sentence like this > > "book has color red" > > > > how can i map it into FOL format > > subject, object, property and property value > > which is which. thank u > > sorry for newbie quesiton. > > > > yours > > peter. > > > > > >
Received on Tuesday, 28 September 2004 04:15:01 UTC