[seek-kr-sms] KR Ontology Needs

[Matt Jones]

Hi Susan,

That sounds great.  I'll ask Bertram, Mark, Shawn, and Rich to be sure 
to include Joana in any future discussions regarding this issue.  Will 
she be attending the KR/SMS meeting in September?  Maybe she should -- I 
think some of this stuff will be discussed then, although probably not 
resolved.

As far as a corpus goes -- we're not really trying to search the text. 
We're really trying to search data, metadata descriptions of those data, 
and metadata on analytical processes, using formally structured semantic 
queries.  Of course there's text in there, but its highly structured.  I 
think Shawn was proposing that users might not like to compose queries 
using formal terms from our ontologies, but that keywords that were 
formally defined in terms of the ontologies would provide a bridge that 
the users could use more effectively.  Its pretty much an open question 
about what we do, but the goal is to provide the benefits of semantic 
linkages during search and integration of data and models.  If we need a 
corpus to mine for terms and their semantic associations, there are the 
journals like Ecology that might work.  I'm not sure what exactly would 
be needed, or how specific the text would need to be.  But I'm sure 
something could be found that is relevant and rich.

There are a set of use cases minimally outlined in our notes from Santa 
Barbara in cvs.  The notes are somewhat cryptic (I took them so can help 
clarify if you have questions), but they would make a good starting 
point.  See pages 6, 10, 11, and 12 of:
http://cvs.ecoinformatics.org/cvs/cvsweb.cgi/~checkout~/seek/projects/ams/docs/meeting-notes-sms-20031026.pdf?rev=1.2&content-type=application/pdf

Matt

sgauch at ittc.ku.edu wrote:
> The interactions between ontologies and keywords is one of great interest
> to me.  I would be happy to contribute to this discussion and development
> in any way.  The general area of conceptual search is my particular
> favorite.  Joana Trajkova is looking for Ph.D. topics that relate to SEEK
> and something tying ontologies to search to science would be an
> appropriate area.
> 
> The main difficulty I see is that I am not familiar enough with
> biodiversity applications to see where this ties in.  What fundamentally
> textual resources exist that need to searched?  So, the first step for me
> would be articulating some realistic (and hard!) use cases.
> 
> Susan.
> 
> 
> 
> 
> _______________________________________________
> seek-kr-sms mailing list
> seek-kr-sms at ecoinformatics.org
> http://www.ecoinformatics.org/mailman/listinfo/seek-kr-sms

-- 
-------------------------------------------------------------------
Matt Jones                                     jones at nceas.ucsb.edu
http://www.nceas.ucsb.edu/    Fax: 425-920-2439    Ph: 907-789-0496
National Center for Ecological Analysis and Synthesis (NCEAS)
University of California Santa Barbara
Interested in ecological informatics? http://www.ecoinformatics.org
-------------------------------------------------------------------