Graph-to-Graph Inference Rules (GGIR)

> >
> >    I think with this kind of approach, the interpretation of bNodes as
> >    file-scope existential variables works perfectly.  In the premise,
> >    they are just anonymous universal variables (because they are inside
> >    the "if"), and in the conclusion they become skolem functions of the
> >    universal variables, which is very useful. 
> >
> >I think you pulled a fast one there.
> 
> Not if you interpret 'premis' and 'conclusion' as applying to an 
> inference rule (or sequent) rather than an implication. I think that 
> is what Sandro had in mind (yes?)

Yes.

> >If the variables are really
> >file-scope existentials, then a rule would look like this:
> >
> >ex x ((P x) -> (Q x))  
> >   [the 'ex x' might actually be way outside the implication,
> >    which is just one of many things in the file]
> >
> >This is not equivalent to
> >
> >ex x ((fa x (P x)) -> (Q x))
> >
> >or to
> >
> >fa x ((P x) -> (Q x))
> 
> But think of the rule instead as a relation between two RDF graphs 
> that indicates an extra-logical entailment licence, and allows one 
> graph to be transformed into another. Then the claim makes sense, 
> since none of the *logical* scopes go across the arrow (which really 
> ought be written |- rather than -> )

Yeah.   

So, is this list (rdf-rules) the place to talk about inference rules
like this, and www-rdf-logic the place to be if you want to talk about
logical implication (with generalized modus ponens, or resolution, or
something as your inference rule).....?  That is, is the focus of
www-rdf-rules extra-logical (in the sense Pat means)?

I imagine using a set of these rules as a kind of filter, a stage in
a pipeline of processing an RDF Graph.

What I can't figure out is whether or when it's better to use these
Graph->Graph Inference Rules and when it's better to just use logical
implication.  On the surface, GGIR looks faster, but less flexible,
and not naturally transmitted along with the data.  It's more like a
program that processes data (and via Prolog becomes one).

But this seems like a false dichotomy when I look deeper, since an
agent can fairly easily map back and for between the two (in at least
the areas where the expressiveness is the same).

    -- sandro

Received on Tuesday, 9 October 2001 03:55:10 UTC