> Proximity is distance.. Within X characters.. Within X words.. within
> some metric. Same element is NOT a metric.
Does anyone else remember the old BRS Search Service? (traces of which were
still visible in Ovid's command line interface until fairly recently)
In addition to the usual Booleans, there were three "positional" operators:
ADJ (adjacent terms)
WITH (terms occur in the same sentence of an element)
SAME (terms occur in the same field/element)
There was no requirement to specify which element, though it was an option.
Basic training included an illustration of how the "inverted index" for a
database was actually constructed... so that searchers would understand how
the positional operators worked (i.e., they were evaluated by comparing the
field number, sentence-within-field number and word-within-sentence number
of the two operands).
---------------------------------------
Julie Blume Nye
Senior Product Designer -- Fretwell-Downing, Inc.
407 River Trace Dr.
Rougemont, NC 27572
Phone: 336-364-2607 Fax: 336-364-4224
E-mail: mailto:[log in to unmask]
Website: www.fdusa.com
~~~~~~~~~~~~~~ DISCLAIMER ~~~~~~~~~~~~~~
The information transmitted in this electronic mail message may contain
confidential and or privileged materials. For full details and restrictions
see http://www.fdgroup.com/emaildisclaimer.html
> -----Original Message-----
> From: SRU (Search and Retrieve Via URL) Implementors
> [mailto:[log in to unmask]] On Behalf Of Edward C. Zimmermann
> Sent: Tuesday, November 29, 2005 8:31 AM
> To: [log in to unmask]
> Subject: Re: Proximity search
>
> Quoting Mike Taylor <[log in to unmask]>:
>
> > I think you've missed an earlier part of the discussion.
> CQL already
> > has nice, general syntax for the general proximity searching. What
> > we're talking about here is whether a specific case of
> proxmity ("same
> > element" searching) is best expressed using proximity with
> a specific
>
> I don't sematically see "same element" (in the same leaf container)
> as proximity.
>
> Proximity is distance.. Within X characters.. Within X words.. within
> some metric. Same element is NOT a metric.
>
> Questions (that some people tend to, incorrectly I'd argue,
> associate with
> proximity) like terms in the same paragraph, same line, same
> sentence etc.
> are a result of content structure and not word metrics.
>
> In the same element... What element?
>
> We solved it in our own little "private" query space with the binary
> operator "peer".
>
> The RPN query: "search" "retrieval" PEER
> meaning that the terms "search" and "retrival" are in the
> same (any) leaf.
>
> We, of course, have want to ask for terms in a specific
> element (named, and
> not just the final leaves). We addressed this in our private
> language with
> a binary operator as pair Operator:Fieldpath, for example
> AND:RSS\CHANNEL\DESCRIPTION
> and in any "Description", not just those under channel under ..
> AND:DESCRIPTION
>
> same paragraph, line etc. are given content structure child's play..
>
> The trick we have is to then map whatever query language for
> an interface
> we want into our private language.
>
>
> > set of parameters, and if so whether it merits an alias
> such as "with"
> > or "where". For this, "near" would be totally inappropriate.
>
>
>
> --
> --
> Edward C. Zimmermann, Basis Systeme netzwerk, Munich
> Office Leo (R&D):
> Leopoldstrasse 53-55, D-80802 Munich,
> Federal Republic of Germany
> Telephone: Voice:= +49 (89) 385-47074 Corp.Fax:= +49 (89)
> 692-8150
> Nomadic (SMS/MMS/Fax):= +49 (176) 100-360-55 Alt.Mobile:=
> +49 (179) 205-0539
> http://www.nonmonotonic.net
>
|