RE: SUO: Re: Graphs and Labelled Graphs
I am not necessarily particularly interested in graph theory per se, it is
just that there are a lot of problems that I need to handle for which I
resort to using graphs. For me the importance of labelling a graph is
simply to provide the mapping between the graph and the system (or
representation of that system) that the graph represents. I therefore agree
wholeheartedly with the statement "you will typically abstract away from the
labels" as, from my perspective, that is the reason for having the labels.
Regards
Chris Angus
-----Original Message-----
From: owner-standard-upper-ontology@majordomo.ieee.org
[mailto:owner-standard-upper-ontology@majordomo.ieee.org]On Behalf Of
Chris Menzel
Sent: 30 June 2003 17:37
To: IEEE Standard Upper Ontology List
Subject: Re: SUO: Re: Graphs and Labelled Graphs
On Sun, Jun 29, 2003 at 01:08:26AM -0400, Patrick Cassidy wrote:
> I will repeat my original request with a slight
> modification so as not to be misunderstood:
>
> > Jon,
> > Could you tell us which field of *experimental*
> ? science or engineering (not mathematics) uses the
> > distinction of "graph" and "labeled graph",
> > and provide a specific example of how such a
> > distinction is used? I have never run into that
> > in my readings on scientific topics.
> ...
> Please note:
> (1) I excluded mathematics from the fields
> I am curious about that might use the distinction.
> Theory of computation is for this purpose a branch of
> mathematics. Actual computational practice -- say,
> C++ or JAVA standard classes in some commercial
> compiler that make such a distinction -- would be
> a legitimate example of an engineering use.
Hi Patrick. I guess I don't see quite the same clear divide between
the theory of computation and actual computational practice that you
seem to be presupposing here. Often the data we actually process
computationally takes the form of a labeled (or colored, or whatever)
graph. When you think about how to search or otherwise process such a
structure, however, you will typically abstract away from the labels and
consider the structure of the graph itself -- a binary tree, say -- in
order to identify the most efficient search algorithm for graphs with
that structure (an algorithm typically identified in a book like Knuth's
AoCP volume on sorting and searching).
By my admittedly weak lights, that sort of seems like an "engineering
use" of the distinction between labeled (colored, etc) graphs and graphs
per se to me.
Chris Menzel