<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
</head>
<body bgcolor="#ffffff" text="#000000">
<p class="MsoNormal" style="text-align: center;" align="center">LDC2009T12<br>
- <a
href="http://www.ldc.upenn.edu/Catalog/CatalogEntry.jsp?catalogId=LDC2009T12"><b>2008
CoNLL Shared Task Data</b></a><a
href="http://www.ldc.upenn.edu/Catalog/CatalogEntry.jsp?catalogId=LDC2009T12"> </a>
-<br>
<br>
LDC2009T13<br>
- <b><a
href="http://www.ldc.upenn.edu/Catalog/CatalogEntry.jsp?catalogId=LDC2009T13">English
Gigaword Fourth Edition</a> -</b><br>
<br>
LDC2009T09<br>
<b>- <a
href="http://www.ldc.upenn.edu/Catalog/CatalogEntry.jsp?catalogId=LDC2009T09">GALE
Phase 1 Arabic Newsgroup Parallel Text - Part 2</a> -<o:p> <br>
<br>
</o:p></b></p>
<p class="MsoNormal" style="text-align: center;" align="center"><o:p>The
Linguistic Data Consortium (LDC) would like to announce the
availability of three new publications.</o:p><b><o:p><br>
</o:p></b></p>
<div class="MsoNormal" style="text-align: center;" align="center">
<hr align="center" size="2" width="100%"></div>
<div align="center"><b>New Publications</b><o:p></o:p>
</div>
<p>(1) <a
href="http://www.ldc.upenn.edu/Catalog/CatalogEntry.jsp?catalogId=LDC2009T12">2008
CoNLL Shared Task Data</a> contains the trial corpus, training corpus,
development and test data for the <a
href="http://www.yr-bcn.es/conll2008/">2008
CoNLL (Conference on Computational Natural Language Learning) Shared
Task
Evaluation</a>. The 2008 Shared Task developed syntactic dependency
annotations, including information such as named-entity boundaries and
the
semantic dependencies model roles of both verbal and nominal
predicates. The
materials in the Shared Task data consist of excerpts from the
following
corpora: <a
href="http://www.ldc.upenn.edu/Catalog/CatalogEntry.jsp?catalogId=LDC99T42">Treebank-3</a>
LDC99T42 , <a
href="http://www.ldc.upenn.edu/Catalog/CatalogEntry.jsp?catalogId=LDC2005T33">BNN
Pronoun Coreference and Entity Type Corpus</a> LDC2005T33, <a
href="http://www.ldc.upenn.edu/Catalog/CatalogEntry.jsp?catalogId=LDC2004T14">Proposition
Bank I</a> LDC2004T14 (PropBank) and <a
href="http://www.ldc.upenn.edu/Catalog/CatalogEntry.jsp?catalogId=LDC2008T23">NomBank
v 1.0</a> LDC2008T23. <o:p></o:p></p>
<p>The <a href="http://www.cnts.ua.ac.be/conll2008/">Conference on
Computational Natural Language Learning (CoNLL)</a> is accompanied
every year
by a shared task intended to promote natural language processing
applications
and evaluate them in a standard setting. The 2008 shared task employed
a
unified dependency-based formalism and merged the task of syntactic
dependency
parsing and the task of identifying semantic arguments and labeling
them with
semantic roles. <o:p></o:p></p>
<p>The 2008 shared task was divided into three subtasks: <o:p></o:p></p>
<ol start="1" type="1">
<li class="MsoNormal" style="">parsing syntactic dependencies <o:p></o:p></li>
<li class="MsoNormal" style="">identification and disambiguation of
semantic predicates <o:p></o:p></li>
<li class="MsoNormal" style="">identification of arguments and
assignment of semantic roles for each predicate <o:p></o:p></li>
</ol>
<p>Several objectives were addressed in this shared task:<o:p></o:p></p>
<ul type="disc">
<li class="MsoNormal" style="">Semantic Role Labeling (SRL) was
performed and evaluated using a dependency-based representation for
both syntactic and semantic dependencies. While SRL on top of a
dependency treebank has been addressed before, the approach of the 2008
Shared Task was characterized by the following novelties: <o:p></o:p></li>
<ol start="1" type="1">
<li class="MsoNormal" style="">The constituent-to-dependency
conversion strategy transformed all annotated semantic arguments in
PropBank and NomBank v 1.0, not just a subset; <o:p></o:p></li>
<li class="MsoNormal" style="">The annotations addressed
propositions centered around both verbal (PropBank) and nominal
(NomBank) predicates. <o:p></o:p></li>
</ol>
<li class="MsoNormal" style="">Based on the observation that a richer
set of syntactic dependencies improves semantic processing, the
syntactic dependencies modeled are more complex than the ones used in
the previous CoNLL shared tasks. For example, the corpus includes
apposition links, dependencies derived from named entity (NE)
structures, and better modeling of long-distance grammatical relations.
<o:p></o:p></li>
<li class="MsoNormal" style="">A practical framework is provided for
the joint learning of syntactic and semantic dependencies. <o:p></o:p></li>
</ul>
<br>
<div align="center"><b>*</b><br>
</div>
<p><o:p></o:p></p>
<p>(2) <a
href="http://www.ldc.upenn.edu/Catalog/CatalogEntry.jsp?catalogId=LDC2009T13">English
Gigaword Fourth Edition</a>.<span style=""> </span>English
Gigaword, now being released in its fourth edition, is a comprehensive
archive
of newswire text data that has been acquired over several years by the
LDC at
the University of Pennsylvania. The fourth edition includes all of the
contents
in English Gigawaord Third Edition (<a
href="http://www.ldc.upenn.edu/Catalog/CatalogEntry.jsp?catalogId=LDC2007T07">LDC2007T07</a>)
plus new data covering the 24-month period of January 2007 through
December
2008. <o:p></o:p></p>
<p>The six distinct international sources of English newswire included
in this
edition are the following:<o:p></o:p></p>
<ul type="disc">
<li class="MsoNormal" style="">Agence France-Presse, English Service
(afp_eng) <o:p></o:p></li>
<li class="MsoNormal" style="">Associated Press Worldstream, English
Service (apw_eng) <o:p></o:p></li>
<li class="MsoNormal" style="">Central News Agency of Taiwan, English
Service (cna_eng) <o:p></o:p></li>
<li class="MsoNormal" style="">Los Angeles Times/Washington Post
Newswire Service (ltw_eng) <o:p></o:p></li>
<li class="MsoNormal" style="">New York Times Newswire Service
(nyt_eng) <o:p></o:p></li>
<li class="MsoNormal" style="">Xinhua News Agency, English Service
(xin_eng) <o:p></o:p></li>
</ul>
<p><small><big> New in the Fourth Edition:</big><o:p></o:p></small></p>
<ul type="disc">
<li class="MsoNormal" style="">Articles with significant Spanish
language content have now been identified and documented. <o:p></o:p></li>
<li class="MsoNormal" style="">Markup has been simplified and made
consistent throughout the corpus. <o:p></o:p></li>
<li class="MsoNormal" style="">Information structure has been
simplified. <o:p></o:p></li>
<li class="MsoNormal" style="">Character entities have been
simplified. <o:p></o:p></li>
</ul>
<div align="center"><b>*</b><br>
</div>
<p><o:p></o:p></p>
<p>(3) <a href="cid:part1.03030002.01000209@ldc.upenn.edu"><b><span
style="color: windowtext; text-decoration: none;"><span style=""> </span></span></b><span
style="color: windowtext; text-decoration: none;"><span style=""><u><span
style="color: blue;">GALE Phase 1 Arabic Newsgroup Parallel Text -
Part 2</span></u></span></span></a>
contains a total of 145,000 words (263 files) of Arabic newsgroup text
and its
translation selected from thirty-five sources. Newsgroups consist of
posts to
electronic bulletin boards, Usenet newsgroups, discussion groups and
similar
forums. This release was used as training data in Phase 1 (year 1) of
the
DARPA-funded GALE program. This is the second of a two-part release. <a
href="http://www.ldc.upenn.edu/Catalog/CatalogEntry.jsp?catalogId=LDC2009T03">GALE
Phase 1 Arabic Newsgroup Parallel Text - Part 1</a> was released in
early 2009.<o:p></o:p></p>
<p>Preparing the source data involved four stages of work: data
scouting, data
harvesting, formating and data selection.<o:p></o:p></p>
<p class="MsoNormal" style="">Data
scouting involved manually searching the web for suitable newsgroup
text. Data
scouts were assigned particular topics and genres along with a
production
target in order to focus their web search. Formal annotation guidelines
and a
customized annotation toolkit helped data scouts to manage the search
process
and to track progress. <o:p></o:p></p>
<p>Data scouts logged their decisions about potential text of interest
to a
database. A nightly process queried the annotation database and
harvested all
designated URLs. Whenever possible, the entire site was downloaded, not
just
the individual thread or post located by the data scout. Once the text
was
downloaded, its format was standardized so that the data could be more
easily
integrated into downstream annotation processes. Typically, a new
script was
required for each new domain name that was identified. After scripts
were run,
an optional manual process corrected any remaining formatting problems.<br>
<br>
The selected documents were then reviewed for content-suitability using
a
semi-automatic process. A statistical approach was used to rank a
document's
relevance to a set of already-selected documents labeled as "good."
An annotator then reviewed the list of relevance-ranked documents and
selected
those which were suitable for a particular annotation task or for
annotation in
general. These newly-judged documents in turn provided additional input
for the
generation of new ranked lists. <o:p></o:p></p>
<p class="MsoNormal" style="">Manual
sentence units/segments (SU) annotation was also performed as part of
the
transcription task. Three types of end of sentence SU were identified:
statement SU, question SU, and incomplete SU. After transcription and
SU
annotation, files were reformatted into a human-readable translation
format and
assigned to professional translators for careful translation.
Translators
followed LDC's GALE Translation guidelines which describe the makeup of
the
translation team, the source data format, the translation data format,
best
practices for translating certain linguistic features and quality
control
procedures applied to completed translations.<br>
</p>
<hr size="2" width="100%">
<p class="MsoNormal" style=""> <o:p></o:p></p>
<span style="color: black;"></span>
<div align="center"><font face="Courier New, Courier, monospace"><small><small><big>Ilya
Ahtaridis<br>
Membership Coordinator</big><br>
<br>
</small>--------------------------------------------------------------------</small><small><br>
</small></font></div>
<div align="center">
<pre class="moz-signature" cols="72"><font
face="Courier New, Courier, monospace">Linguistic Data Consortium Phone: (215) 573-1275
University of Pennsylvania Fax: (215) 573-2175
3600 Market St., Suite 810 <a
class="moz-txt-link-abbreviated" href="mailto:ldc@ldc.upenn.edu">ldc@ldc.upenn.edu</a>
Philadelphia, PA 19104 USA <a
class="moz-txt-link-freetext" href="http://www.ldc.upenn.edu">http://www.ldc.upenn.edu</a></font></pre>
</div>
<br>
<pre class="moz-signature" cols="72">
</pre>
</body>
</html>