W3C home > Mailing lists > Public > www-rdf-interest@w3.org > February 2001

Re: About using distances or probabilities

From: David Allsopp <dallsopp@signal.dera.gov.uk>
Date: Thu, 08 Feb 2001 14:24:07 +0000
Message-ID: <3A82AC07.6758CD53@signal.dera.gov.uk>
CC: www-rdf-interest@w3.org


Bill de hOra wrote:
> 
> : > You could hang the probabilites off the property nodes:
> : >
> : > :Computer :pal20/30  :Printer.
> : > :pal20/30  :probability "20"; :inverseProbability "30"; a :Property.
> :
> : Presumably this requires a unique Property instance for each different
> : link between nodes, for example if I have
> :
> : John--believes-->SkyIsBlue
> : John--believes-->MagicPixies
> :
> : and want to assign 99% confidence to the first one and 1% to the second,
> : I'd need a 'believes99' property and a 'believes01' property (or reify
> : both statements and attach the confidence level to those) ?
> 
> Short answer:
> Don't put number typing in your schema. Put the burden on the application. So
> use Jena instead.
> 
> Long answer:
> Even if you did define believes01 through believes100 you'll still have to do a
> lookup on 100 legal values in a validating processor for every number or convert
> them to numbers in the application to do a range check. And it'll be pointless
> for reals numbers of course.
> 
> Given that, I suggest you just send the property's value as a string literal
> (that will let them pass through any RDF processor). So shift the burden onto
> the apps to *produce* good literals in the first place. Brian McBride's Jena has
> facilities for handling numbers if you're prepared to use java. If not you could
> clone his API calls for the language of your choice (sorry Brian :)

I wasn't suggesting that we use property names for encoding integers; I
guess labelling them 99 and 01 was misleading.

My point was that you have to have (I think?) different instance of the
relevant subclass of Property, with different URI's,  in order to attach
further triples to them, otherwise you'd be saying that the property
'believes' always has the confidence '99%' or whatever.  Perhaps this is
blindingly obvious from the previous example but I'm not familiar with
the N3 notation yet.  

This would presumably mean that in order to do any useful querying of
your model, the query engine would have to be schema-aware, to work out
that all your various properties are in fact instances of 'believes'.

How would you write 

    :Computer :pal20/30  :Printer.
    :pal20/30  :probability "20"; :inverseProbability "30"; a :Property.

in XML M&S syntax, anyway? Something like:

 <rdf:Property ID="pal2030">
	<a:probability>20</a:probability>
        <a:inverseProbability>30</a:inverseProbability>
 </rdf:Property>

<a:Computer ID="MyPC"/> 
<a:Printer ID="MyPrinter"/> 

<Description about="MyPC">
   <pal2030 resource="MyPrinter"/>
</Description>

Regards,

David.



Regards,

David.

-- 
The peason-molesworth space ship threatened by wild mercurian maths 
masters lies disabled in the onion bed of the kitchen garden. 
Meanwhile in the master's common room, Sigismund arbuthnot the mad 
maths master musters his rhomboids...
Received on Thursday, 8 February 2001 09:26:00 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:51:47 GMT