[rdfweb-dev] Re: Parsing FOAF in perl?

[Ian Davis]

On Tuesday, 10 December 2002 at 16:22, Danny Ayers wrote:

> Jena has Model classes to hold graphs, and it's easy enough to have
> a Model object containing Ian's foaf statements and another to hold Jim's.
> It's also easy enough to maintain provenance info, e.g. in another Model
> instance, containing statements referring to the documents' URIs. But what
> happens when you try and reason across these? Either you can do it
> programmatically, which seems clunky, for every operation you'd have
> 'foreach model {doStuff}', or you could merge all the triples together into
> a single model, then bigModel {doStuff}. This way you have to use the
> provenance info at merge time (perhaps one-off processing, or applying it to
> reified statements from the Models), but either way it seems like it needs
> more work that it should in principle... I'd very much like to hear more
> about how everyone's approaching this kind of situation.
Well one approach I've played with (in the thought experiment sense)
is a kind of evolutionary approach. In this scheme each batch of
statements is considered a source model. These source models are
combined to produce inferred models provided there are no logical
contradictions. As new source models are added they are merged into
new inferred models. All the models co-exist and are independent. When
the system is queried all models are consulted and some kind of
fitness function is applied so that inferred models that are
consistently wrong are gradually eliminated.

Its a sort of experience based reasoner. Like a human it will make
mistakes while it is learning who to trust.

> Cheers,
> Danny.
Ian