Re: WordNet Task Force - work outline

Some of the HP team were talking about the WordNet TF over lunch.
We found the survey of current work that Aldo produced a useful resource, 
but feel somewhat uncomfortable with it as scoping the TF.

In a sense, the survey showed what we think is the problem: too many 
approaches, all of which have merits; rather than one mapping of WordNet to 
RDF or OWL that is good enough for most users.

We felt that what would be most useful in the short term is to have a 
standard representation of WordNet in RDF that people can use.  We think 
this is what the community of implemetors needs and will most aid 
deployment in the short term.

Some specific deliverables that we think would be very useful and 
potentially could be achieved quickly are:

+ an RDF schema or OWL ontology with which to talk about the main wordnet 
relationships and concepts (e.g. words, senses, hyponyms, synonyms etc.)
+ a namespace URI for this schema
+ a version of WordNet converted into triples, using this schema and namespace
+ some illustrative examples of use

I believe that at least some of the approaches listed already provide at 
least some of these. Hopefully an understanding of these, and the expertise 
of the TF, will allow a best of breed proposal.


It would not matter if a first version only covered some core concepts 
(maybe the four above), and a later version added more sophistication.
The key problem facing a naive semantic web user, or a group producing 
tools for semantic web developers is making the choice - of which of the 
mappings to use, and hence which schema and which namespace URI.

An approach that would emphasize consensus and avoid blessing any one 
solution would be to provide a namespace URI and the elements shared across 
all prior solutions, and a forum in which the different wordnet mappers 
could agree amongst themselves how to resolve differences (or enabling a 
clear articulation of the differences, with their pros and cons)

When identifying the deliverables for this (or any) Task Force, we should 
also identify the target audience, and possible use cases in which that 
target audience may find the deliverables useful. It would be good to have 
a clear idea from the target audience what they want, so the work is based 
more on pull ("this is what you are asking for") than push ("take this 
because the doctor says it will be good for you").

We do not think that the target audience for the WordNet TF is people 
working on WordNet mappings, we think the target audience is any semantic 
web developer who might find a particular WordNet mapping useful.
For example, anyone creating an OWL or RDFS class might wish to annotate it 
with its intended meaning using *the* URI for a specific sense of an 
English word, as classified by wordnet. The main requirement from this use 
case is agreement over what that URI is, including the beginning bit (the 
namespace) and the end bit (the mapping from Wordnet's representation of 
senses)

Such basic agreement on the fundamentals will also help people doing more 
advanced work on Wordnet mappings, by giving them a baseline from which to 
start, and a shared vocabulary on which to build. When any research they do 
is complete, the Wordnet TF (possibly reincarnated), or whoever is 
maintaining the ontology recommended by the Wordnet TF, could then consider 
how to integrate such completed research into the best practice.

Jeremy

Received on Thursday, 25 March 2004 13:10:03 UTC