Mailing-List: contact user-help@uima.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@uima.apache.org
Received-SPF: pass (nike.apache.org: local policy)
Date: Tue, 15 Apr 2014 19:32:19 +0200
From: Petr Baudis <pasky@ucw.cz>
To: user@uima.apache.org
Subject: Re: Complex architectures with multiple CASes - how to?
Message-ID: <20140415173219.GV6156@machine.or.cz>
References: <20140409023434.GP6156@machine.or.cz>
 <CAGRYgEMA619SVXcKYyyMt0JzqerX8cNucx6ODBi6rssZrOx30g@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: 
 <CAGRYgEMA619SVXcKYyyMt0JzqerX8cNucx6ODBi6rssZrOx30g@mail.gmail.com>
User-Agent: Mutt/1.5.21 (2010-09-15)

  Hi!

On Wed, Apr 09, 2014 at 04:14:53PM -0400, Eddie Epstein wrote:
> One approach is for the cas multiplier to put the question View in each
> search CAS, using the CasCopier, and create a second view for search result
> processing. Down stream annotators would then be multi-view, getting
> question analysis results from one view and doing search analysis in the
> other.

  Thank you, that was just what I was looking for!  This works pretty
well, and on careful re-reading of the Making Watson Fast paper,
I realize that this is what they were in fact doing as well.

  P.S.: I had much trouble finding any actual code examples for the cases
where non-trivial CAS multipliers and views manipulation is required,
UIMAfit is used, etc.  My code that juggles multiple views and CASes
with UIMAfit is open source at https://github.com/brmson/yodaqa, perhaps
someone will find that useful.

				Petr "Pasky" Baudis