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: domain of pt.activemq@gmail.com
 designates 209.85.160.47 as permitted sender)
MIME-Version: 1.0
Date: Fri, 13 Apr 2012 22:06:49 +0200
Message-ID: 
 <CAKapaZ-AknSjg_taC-m4UJuUS1RFKxte7kjiwYtTf38bjrDUGg@mail.gmail.com>
Subject: CAS Editor: Sentence Detector shows strange suggestions...
From: Peter Thygesen <pt.activemq@gmail.com>
To: user@uima.apache.org
Content-Type: multipart/alternative; boundary=047d7b15ae5d466dfb04bd95031e

--047d7b15ae5d466dfb04bd95031e
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Guess this one is for you J=F6rn ;-)

Can I use the trunk version of the sentence detector tool?

I have downloaded and built the entire uimaj trunk. Configured Eclipse and
have no problems annotating and building models. However the "new" sentence
suggesting view does not seem to work correctly. First I annotate
headlines, subheadlines and paragraphs. The sentence detector is configured
to use my paragraph type and exclude headline and subheadline.

While annotating the suggestions gets worse and worse... The sentences are
repeated... and does not seem to be able to look further if it encounters a
subheadline, because there are not sentence suggestions after the first
subheadline even though there are more paragraphs.

thx,
Peter Thygesen

(also using opennlp-tools 1.5.3-SNAPSHOT, opennlp-uima 1.5.3-SNAPSHOT,
opennlp-maxent.3.0.3-SNAPSHOT is this is important)

--047d7b15ae5d466dfb04bd95031e--