Hi all,<br><br>"Speech and Language Technology for Minority Languages" (SALTMIL) will have its annual workshop on the 7th of Sept. 2009. The call for papers follows. To learn more about the group you can visit their website <a href="http://ixa2.si.ehu.es/saltmil/en/about-saltmil/aims/aims-of-isca-sig-speech-and-language-technology-for-minority-languages.html">http://ixa2.si.ehu.es/saltmil/en/about-saltmil/aims/aims-of-isca-sig-speech-and-language-technology-for-minority-languages.html</a><br>
<br><br><br> CALL FOR PAPERS<br>
Information Retrieval and Information Extraction<br>
for less resourced languages.<br>
IE-IR_LSL<br>
<br>
SEPLN 2009 pre-conference workshop<br>
University of the Basque Country<br>
Donostia-San Sebastián. Monday 7th September 2008<br>
<br>
<br>
Organised by the SALTMIL Special Interest Group of ISCA<br>
SALTMIL: <a href="http://ixa2.si.ehu.es/saltmil" target="_blank">http://ixa2.si.ehu.es/saltmil</a><br>
SEPLN 2009: <a href="http://ixa2.si.ehu.es/sepln2009" target="_blank">http://ixa2.si.ehu.es/sepln2009</a><br>
Paper submission: <a href="http://sepln.org/myreview-saltmil2009" target="_blank">http://sepln.org/myreview-saltmil2009</a><br>
Deadline for submission: 8 June 2009<br>
<br>
Papers are invited for the above half-day workshop, in the format<br>
outlined below. Most submitted papers will be presented in poster form,<br>
though some authors may be invited to present in lecture format.<br>
<br>
CONTEXT AND FOCUS<br>
<br>
The phenomenal growth of the Internet has led to a situation where, by<br>
some estimates, more than one billion words of text is currently<br>
available. This is far more text than any given person can possibly<br>
process. Hence there is a need for automatic tools to access and process<br>
this mass of textual information. Emerging techniques of this kind<br>
include Information Retrieval (IR), Information Extraction (IE), and<br>
Question Answering (QA)<br>
<br>
However, there is a growing concern among researchers about the<br>
situation of languages other than English. Although not all Internet<br>
text is in English, it is clear that non-English languages do not have<br>
the same degree of representation on the Internet. Simply counting the<br>
number of articles in Wikipedia, English is the only language with more<br>
than 20 percent of the available articles. There then follows a group of<br>
17 languages with between one and ten percent of the articles. The<br>
remaining 245 languages each have less than one percent of the articles.<br>
Even these low-profile languages are relatively privileged, as the<br>
total number of languages in the world is estimated to be 6800.<br>
<br>
Clearly there is a danger that the gap between high-profile and<br>
low-profile languages on the Internet will continue to increase, unless<br>
tools are developed for the low-profile languages to access textual<br>
information. Hence there is a pressing need to develop basic language<br>
technology software for less-resourced languages as well. In particular,<br>
the priority is to adapt the scope of recently-developed IE, IR and QA<br>
systems so that they can be used also for these languages. In doing so,<br>
several questions will naturally arise, such as:<br>
<br>
- What problems emerge when faced with languages having different<br>
linguistic features from the major languages?<br>
- Which techniques should be promoted in order to get the maximum<br>
yield from sparse training data?<br>
- What standards will enable researchers to share tools and techniques<br>
across several different languages?<br>
- Which tools are easily re-useable across several unrelated languages?<br>
<br>
It is hoped that presentations will focus on real-world examples, rather<br>
than purely theoretical discussions of the questions. Researchers are<br>
encouraged to share examples of best practice -- and also examples where<br>
tools have not worked as well as expected. Also of interest will be<br>
cases where the particular features of a less-resourced language raise a<br>
challenge to currently accepted linguistic models that were based on<br>
features of major languages.<br>
<br>
<br>
TOPICS<br>
<br>
Given the context of IR, IE and QA, topics for discussion may include,<br>
but are not limited to:<br>
<br>
- Information retrieval;<br>
- Text and web mining;<br>
- Information extraction;<br>
- Text summarization;<br>
- Term recognition;<br>
- Text categorization and clustering;<br>
- Question answering;<br>
- Re-use of existing IR, IE and QA data;<br>
- Interoperability between tools and data.<br>
- General speech and language resources for minority languages, with<br>
particular emphasis on resources for IR,IE and QA.<br>
<br>
IMPORTANT DATES<br>
<br>
8 June 2009 Deadline for submission<br>
1 July 2009 Notification<br>
15 July 2009 Final version<br>
7 September 2009 Workshop<br>
<br>
ORGANISERS<br>
<br>
Kepa Sarasola, University of the Basque Country<br>
Mikel Forcada, Universitat d'Alacant, Spain<br>
Iñaki Alegria. University of the Basque Country<br>
Xabier Arregi, University of the Basque Country<br>
Arantza Casillas. University of the Basque Country<br>
Briony Williams, Language Technologies Unit, Bangor University, Wales, UK<br>
<br>
<br>
PROGRAMME COMMITTEE<br>
<br>
Iñaki Alegria. University of the Basque Country.<br>
Atelach Alemu Argaw: Stockholm University, Sweden<br>
Xabier Arregi, University of the Basque Country.<br>
Jordi Atserias, Barcelona Media (yahoo! research Barcelona)<br>
Shannon Bischoff, Universidad de Puerto Rico, Puerto Rico<br>
Arantza Casillas. University of the Basque Country.<br>
Mikel Forcada, Universitat d'Alacant, Spain<br>
Xavier Gomez Guinovart. University of Vigo.<br>
Lori Levin, Carnegie-Mellon University, USA<br>
Climent Nadeu, Universitat Politècnica de Catalunya<br>
Jon Patrick, University of Sydney, Australia<br>
Juan Antonio Pérez-Ortiz, Universitat d'Alacant, Spain<br>
Bojan Petek, University of Ljubljana, Slovenia<br>
Kepa Sarasola, University of the Basque Country<br>
Oliver Streiter, National University of Kaohsiung, Taiwan<br>
Vasudeva Varma, IIIT, Hyderabad, India<br>
Briony Williams, Bangor University, Wales, UK<br>
<br>
SUBMISSION INFORMATION<br>
<br>
We expect short papers of max 3500 words (about 4-6 pages) describing<br>
research addressing one of the above topics, to be submitted as PDF<br>
documents by uploading to the following URL:<br>
<a href="http://sepln.org/myreview-saltmil2009" target="_blank">http://sepln.org/myreview-saltmil2009</a><br>
The final papers should not have more than 6 pages, adhering to the<br>
stylesheet that will be adopted for the SEPLN Proceedings (to be<br>
announced later on the Conference web site).