<div dir="ltr"><b id="internal-source-marker_0.937090368475765" style="color:rgb(0,0,0);font-family:Times"><p dir="ltr" style="text-align:center;margin-top:0pt;margin-bottom:0pt"><span style="font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap"><font size="4">Call For Participation</font></span></p>

<p dir="ltr" style="font-size:medium;text-align:center;margin-top:0pt;margin-bottom:0pt"><span style="font-size:15px;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap"><span style="font-weight:normal"><br class="">

</span>Multilingual Word Sense Disambiguation<br class=""><span style="font-weight:normal">SemEval 2013 - Task #12</span></span></p><span style="font-size:15px;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap;font-weight:normal"></span><br>

<p dir="ltr" style="font-size:medium;font-weight:normal;text-align:center;margin-top:0pt;margin-bottom:0pt"><span style="font-size:15px;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap"><a href="http://www.cs.york.ac.uk/semeval-2013/task12/">http://www.cs.york.ac.uk/semeval-2013/task12/</a><br class="">

</span></p><span style="font-size:15px;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap;font-weight:normal"></span><br><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap">The aim of this task is to evaluate Word Sense Disambiguation systems in an all-words multilingual setting.</span><br>

<span style="font-size:15px;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap;font-weight:normal"></span><br><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap"><br class="">

</span><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;text-decoration:underline;vertical-align:baseline;white-space:pre-wrap">INTRODUCTION</span><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap"><br class="">

</span><br><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap">Task 12 provides a traditional setup for evaluating Word Sense Disambiguation (WSD) systems in an all-words, multilingual setting by marking occurrences of potentially polysemous words in five different languages (English, French, German, Italian, Spanish) with sense labels provided by a multilingual sense inventory.  To enable multilinguality we make use of the BabelNet sense inventory [1], a wide-coverage semantic network built by merging WordNet with Wikipedia to provide an “encyclopedic dictionary.”  BabelNet concepts are lexicalized in many languages using Wikipedia’s inter-language links and the output of a state-of-the-art machine translation system.   Task 12 will use a validated version of BabelNet 1.1 (</span><a href="http://babelnet.org/" style="font-size:medium;font-weight:normal"><span style="font-size:15px;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap">http://babelnet.org</span></a><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap">) in which the Wikipedia-WordNet mappings of all senses of lemmas in the test data have been manually verified.</span><br>

<span style="font-size:15px;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap;font-weight:normal"></span><br><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;text-decoration:underline;vertical-align:baseline;white-space:pre-wrap">PARTICIPANT SENSE INVENTORY</span><br>

<span style="font-size:15px;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap;font-weight:normal"></span><br><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap">Participants are free to work on the full BabelNet sense inventory or to work on either of its inventory subsets, i.e. WordNet 3.0 or Wikipedia page titles. They are also free to participate using a single language of their choice or all five languages. </span><br>

<span style="font-size:15px;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap;font-weight:normal"></span><br><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;text-decoration:underline;vertical-align:baseline;white-space:pre-wrap">TASK DETAILS</span><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap">:</span><br>

<span style="font-size:15px;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap;font-weight:normal"></span><br><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap">Following the traditional WSD “all-words” experimental setting [2], systems will be expected to </span><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;font-style:italic;vertical-align:baseline;white-space:pre-wrap">link all occurrences of noun phrases within arbitrary texts in different languages to the most suitable senses in the sense inventory of their choice</span><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap">. For instance, given the sentence:</span><br>

<span style="font-size:15px;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap;font-weight:normal"></span><br><ol style="font-size:medium;font-weight:normal;margin-top:0pt;margin-bottom:0pt">

<li dir="ltr" style="list-style-type:decimal;font-size:15px;font-family:Arial;background-color:transparent;vertical-align:baseline"><span style="background-color:transparent;vertical-align:baseline;white-space:pre-wrap">The dramatic force of Miller's play derives in part from expressionistic techniques he used to portray Loman's psychological anguish and guilt-ridden fantasy life.</span></li>

</ol><span style="font-size:15px;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap;font-weight:normal"></span><br><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap">a disambiguation system should link “Miller” to any of (1) the BabelNet synset for </span><a href="http://lcl.uniroma1.it/babelnet/search.jsp?word=Arthur+Miller&lang=EN" style="font-size:medium;font-weight:normal"><span style="font-size:15px;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap">Arthur Miller</span></a><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap">, (2) the Wikipedia sense corresponding to the page </span><a href="http://en.wikipedia.org/wiki/Arthur_Miller" style="font-size:medium;font-weight:normal"><span style="font-size:15px;font-family:Arial;color:rgb(0,0,153);background-color:transparent;vertical-align:baseline;white-space:pre-wrap">http://en.wikipedia.org/wiki/Arthur_Miller</span></a><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap">, or (3) Miller#n#3 (i.e. the third WordNet sense for Miller), depending on the participant’s choice of sense inventory.  Note that the BabelNet synset will contain where applicable both the Wikipedia page and the WordNet synset in its representation.</span><br>

<span style="font-size:15px;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap;font-weight:normal"></span><br><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap">Participants will be evaluated in groups based on their choice of sense inventory and target language. All the information about the submitted systems (such as training data, resources, etc. used by the system) will be reported in the task paper.</span><br>

<span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap"><br class=""></span><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;text-decoration:underline;vertical-align:baseline;white-space:pre-wrap">DATASETS</span><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap">:</span><br>

<span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap"><br class="">No training data will be provided as a part of this task; however, participants are allowed to use any freely available training data for building their system.</span><br>

<span style="font-size:15px;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap;font-weight:normal"></span><br><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap">For annotating the test set, by mid-February we will provide a gold standard version of BabelNet 1.1 where all synsets used in the test data have been manually verified for correctness.<br class="">

 <br class=""><br class=""></span><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;text-decoration:underline;vertical-align:baseline;white-space:pre-wrap">IMPORTANT DATES</span><br>

<span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap"><br class="">February 15, 2013 - Registration Deadline<br class="">March 1, 2013 onwards - Start of evaluation period<br class="">

March 15, 2013 - End of evaluation period<br class="">April 9, 2013 - Paper submission deadline [TBC]<br class="">April 23, 2013 - Reviews Due [TBC]<br class="">May 4, 2013 - Camera ready Due [TBC]<br class=""><br class="">

<br class=""></span><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;text-decoration:underline;vertical-align:baseline;white-space:pre-wrap">MORE INFORMATION</span><br><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;text-decoration:underline;vertical-align:baseline;white-space:pre-wrap"><br class="">

</span><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap">The Semeval-2013 Task #12 website, for signup and details, is:<br class="">

<br class="">   </span><a href="http://www.cs.york.ac.uk/semeval-2013/task12/" style="font-size:medium;font-weight:normal"><span style="font-size:15px;font-family:Arial;background-color:transparent;font-weight:bold;vertical-align:baseline;white-space:pre-wrap">http://www.cs.york.ac.uk/semeval-2013/task12/</span></a><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap"><br class="">

  <br class="">If interested in the task please join our mailing list for updates:<br class=""><br class="">   </span><a href="http://groups.google.com/group/semeval13-multilingual-wsd/" style="font-size:medium;font-weight:normal"><span style="font-size:15px;font-family:Arial;vertical-align:baseline;white-space:pre-wrap">http://groups.google.com/group/semeval13-multilingual-wsd/</span></a><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap"><br class="">

<br class=""><br class=""></span><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;text-decoration:underline;vertical-align:baseline;white-space:pre-wrap">ORGANIZERS</span><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap"><br class="">

Roberto Navigli (<a href="mailto:lastname@di.uniroma1.it">lastname@di.uniroma1.it</a>), Sapienza University of Rome, Italy<br class="">David Jurgens (<a href="mailto:lastname@di.uniroma1.it">lastname@di.uniroma1.it</a>), Sapienza University of Rome, Italy<br class="">

<br class=""><br class=""></span><span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;text-decoration:underline;vertical-align:baseline;white-space:pre-wrap">REFERENCES</span><span style="font-size:15px;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap;font-weight:normal"></span><br>

<span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap">1. Roberto Navigli & Simone Paolo Ponzetto. </span><span style="font-size:15px;font-weight:normal;font-family:Arial;vertical-align:baseline;white-space:pre-wrap">BabelNet: The automatic construction, evaluation and application of a wide-coverage multilingual semantic network. Artificial Intelligence, 193, 2012, pp. 217-250.</span><span style="font-size:15px;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap;font-weight:normal"></span><br>

<span style="font-size:15px;font-weight:normal;font-family:Arial;background-color:transparent;vertical-align:baseline;white-space:pre-wrap">2. Roberto Navigli. Word Sense Disambiguation: A survey. ACM Computing Survey, 41(2), ACM Press, 2009, pp. 1-69.</span></b><br>

</div>