[Corpora-List] help with WS concordance search by syntax

[Smith, Nicholas]

Hi Linda 
It will soon be possible also to use the new version of BNCweb with CQP,
searching for:

_NP0 _NP0 of _NP0

See Stefan Evert and Sebastian Hoffmann's ICAME 2006 talk, at
http://www.eng.helsinki.fi/varieng/icame/programme/Demos.pdf, for more
info.

Nick Smith,
Lancaster.


-----Original Message-----
From: owner-corpora at lists.uib.no [mailto:owner-corpora at lists.uib.no] On
Behalf Of Lou's Laptop
Sent: 16 June 2006 14:45
To: Mark Davies
Cc: Linda Bawcom; corpora at uib.no
Subject: Re: [Corpora-List] help with WS concordance search by syntax

The query can also be solved very easily using the Xaira querybuilder, 
of course -- but I'm sorry to say that the Xaira-indexed full BNC is 
still only available for inhouse testing...

Lou

Mark Davies wrote:
> Hi Linda,
>  
> You may need to use WordSmith, but in case you can consider another
option, there are at least three possibilities:
>  
> 1) Use Bill Fletcher's Phrase in English site (pie.usna.edu) and
search for << np0 np0 of np0 >> . It's fast and you can link to KWIC
entries, but it will only display those that occur three times or more
in the BNC (173 / 4211 types, or about 4% of all types)
>  
> 2) Use the VIEW interface to the BNC (http://view.byu.edu) and search
for << [np0] of [np0] >> and then click on the entries that look like
they might have another [np0] before the first [np0]. Queries with four
sequential high frequency "words" are disallowed via the interface, for
reasons of server load. Sorry.
>  
> 3) I just ran the query directly against the BNC/VIEW database (server
load doesn't matter if it's me doing the query :-), and found all cases
of [np0] [np0] of [np0]. I've placed a file with all matching strings
at: http://view.byu.edu/download/np0_np0_of_np0.txt.
>