Re: OAI as VO Harvesting Interface

From: Patricio F. Ortiz <pfo-at-star.le.ac.uk>
Date: Sat, 13 Sep 2003 10:09:04 +0100 (BST)


Hi Roy,

interesting approach. We've used the mu's with either 0 or 1 so far. Can you enlighten me when a catalog (oops resource) have a ranking of 0.9 in the example below? You've decomposed the query in the way I would've expected, it's just the "fuzzy part" which I'd like to understand more.

Cheers,

Patricio

On Fri, 12 Sep 2003, Roy Williams wrote:
> > (b) Which catalogues does Vizier have giving radio polarisation?
> >
> > select [Catalogs] where Manager/Title~'Vizier' and
> > VOTableColumns/FIELD/@UCD=[some ucd for radio polarization]
>
> It is no longer just an exact text match as it was in UCD1, but rather a
> matching function, mu(u1, u2) which is between 0 (no match) and 1 (excat
> match).
>
> Just as ADQL has a text-based "region" specifier, so it will have a UCD
> selection mechanism based on an appropriate mu function. I would replace
> your Xpath sketch above with a UCDmatch sketch more like this:
>
> select [Catalogs] where Manager/Title~'Vizier'
> and "Mu(VOTableColumns, "phot.radio") > 0.8"
> and "Mu(VOTableColumns, "pol") > 0.8"
>
>

---
Patricio F. Ortiz			pfo-at-star.le.ac.uk
AstroGrid project
Department of Physics and Astronomy
University of Leicester			Tel: +44 (0)116 252 2015
LE1 7RH, UK
Received on 2003-09-13Z11:10:21