- From: James Cerra <jfcst24_public@yahoo.com>
- Date: Sun, 12 Jun 2005 00:18:45 -0700 (PDT)
- To: semantic-web@w3.org
I did some chicken-scratch calculations on the density [1] of a basic rdf:List
and rdf:Seq (i.e. rdfs:Collection subclasses) graph. Say that we represent
each rdf statement as subgraph with three vertices and two edges like so:
Subject -- Predicate -- Object
Each predicate is considered a different node even if they have the same URI
Ref, with one exception. For a rdf:List I made the assumption that each
rdf:List instance connects to the same rdf:type->rdf:List subgraph like so:
_:node1 -- rdf:type -- rdf:List
|
_:node2 ------|
|
_:node3 ------|
I figure that's an obvious optimization for rdf:List instances.
Well, as a result of my calculations I found that a rdf:Seq of 62 items or more
would be sparse, while a rdf:List of 21 items or more is sparse. Is that
correct? Is that one reason for the existance of rdf:List in RDF? Is it
common to have rdf:List and rdf:Seq graphs that big? And in general, what kind
of RDF graphs are sparse?
[1] http://en.wikipedia.org/wiki/Adjacency_matrix
--
Jimmy Cerra
https://nemo.dev.java.net
__________________________________
Discover Yahoo!
Get on-the-go sports scores, stock quotes, news and more. Check it out!
http://discover.yahoo.com/mobile.html
Received on Sunday, 12 June 2005 07:19:15 UTC