========================================================= NODALIDA 2007 NEWSLETTER NO. 4 (May 11, 2007) ========================================================= ========================================================= ========================================================= Important Dates ========================================================= Deadline of late registration: May 15, 2007 Workshops: May 24, 2007 Main conference: May 25-26, 2007 ========================================================= Table of Contents ========================================================= 1. NODALIDA 2007 in Tartu, Estonia 2. Committees 3. Workshops 4. Invited Speakers 5. Language Tutorial: Estonian 6. Registration and Payment/Fees 7. Conference Program 8. Instructions for Presenters 9. Accommodation and Dining ========================================================= 1. NODALIDA 2007 in Tartu, Estonia ========================================================= NODALIDA 2007, the 16th Nordic Conference of Computational Linguistics, will take place 24-26 May 2007 in Tartu, Estonia. Tartu is a little town in the southern part of Estonia. The oldest in Estonia (established 1030), second by the number of inhabitants (100 000) and widely known by the University of Tartu (established 1632). The conference will be held in the heart of the city in two buildigs of the University of Tartu. The workshops and the short tutorial of Estonian (May 24) will be held in the building of the Faculty of Mathematics and Computer Science (J. Liivi 2). All sessions of the main conference (May 25-26) will take place in the University Library (W. Struwe 1). Wireless internet connection is available in all of the conference rooms. Tartu is accessible from Tallinn by coach. The coaches run very frequently (1-2 departures per hour) and the ride takes 2.5 hrs. There are direct flights to Tallinn from many European capital cities. Find out more on http://math.ut.ee/nodalida2007/ ========================================================= 2. Committees ========================================================= PROGRAM COMMITTEE Joakim Nivre (chair), Växjö University and Uppsala University Helena Ahonen-Myka, University of Helsinki Kristiina Jokinen, University of Helsinki and University of Tartu Pierre Nugues, Lund University Stephan Oepen, University of Oslo, NTNU Trondheim and Stanford University Patrizia Paggio, University of Copenhagen Torbj¸rn Svendsen, NTNU Trondheim LOCAL ORGANIZATION COMMITTEE Mare Koit (co-chair), University of Tartu Tiit Roosmaa (co-chair), University of Tartu Urve Talvik, University of Tartu Heli Uibo, University of Tartu Kadri Vider, University of Tartu ========================================================= 3. Workshops ========================================================= The following workshops will be held in conjunction with NODALIDA 2007, May 24, 2007 (J. Liivi 2, Faculty of Mathematics and Computer Science) - The 2nd International Workshop on Typed Feature Structure Grammars http://cst.dk/anders/tfsg/tfsg07.html Room 612 - Building Frame Semantics Resources for Scandinavian and Baltic Languages http://nlp.cs.lth.se/frame2007/ Room 611 - Workshop on Semantic Content Acquisition and Representation http://www.sics.se/~mange/scar2007/ Room 512 - New Developments in Constraint Grammar http://giellatekno.uit.no/cg/ee07/ Room 511 IMPORTANT: TO PARTICIPATE IN A WORKSHOP YOU HAVE TO REGISTER FOR THE MAIN CONFERENCE filling in the registration form on the http://math.ut.ee/nodalida2007/nodaregistration.html AND TO PAY THE CONFERENCE FEE. More information about the workshops is available at their respective websites. ========================================================= 4. Invited Speakers ========================================================= Keynote speakers at NODALIDA 2007 are - Diana F. McCarthy (May 25) - Walter Daelemans (May 26) --------------------------------------------------------------- Evaluating Automatic Approaches for Word Meaning Discovery and Disambiguation using Lexical Substitution (Diana F. McCarthy, University of Sussex) Abstract There has been a surge of interest in Computational Linguistics in word sense disambiguation (WSD). A major catalyst has been the SENSEVAL evaluation exercises which have provided standard datasets for the field over the past decade. Whilst researchers believe that WSD will ultimately prove useful for applications which need some degree of semantic interpretation, the jury is still out on this point. One significant problem is that there is no clear choice of inventory for any given task, other than the use of a parallel corpus for a specific language pair for a machine translation application. Most of the datasets produced, certainly in English, have used WordNet. Whilst WordNet is a wonderful resource it would be beneficial if systems using other inventories could enter the WSD arena without the need for mappings between the inventories which may mask results. As well as the work in disambiguation, there is a growing interest in automatic acquisition of inventories of word meaning. It would be useful to investigate the merits of predefined inventories themselves, aside from their use for disambiguation, and compare automatic methods of acquring inventories. In this talk I will discuss these issues and some results in the context of the English Lexical Substitution Task, organised by myself and Roberto Navigli (University of Rome, "La Sapienza") earlier this year under the auspices of SEMEVAL. --------------------------------------------------------------------- Text Analysis and Machine Learning for Stylometrics and Stylogenetics (Walter Daelemans, University of Antwerp) Abstract Automatic Text Categorization, learning to assign documents to specific categories (e.g. in topic assignment or spam filtering), has been an influential application in Natural Language Processing. These systems consist of two components: a first one that constructs representations of documents (mostly bags of words represented as binary or numeric vectors), and a second one that uses standard machine learning techniques to learn mappings between such document vectors and their topics. Recently, this general approach has been put to use for other, more linguistically interesting "stylometric" applications, such as assigning authorship to documents or determining the gender of the author of a document. Such applications need linguistically more sophisticated document representations and provide insight into which linguistic properties of documents are relevant for predicting the (gender of) the author. In my presentation, I will give a brief overview of results in this approach and describe a number of applications of the methodology we are currently investigating in the CNTS research group. For creating linguistically more interesting document representations, we use a memory-based shallow parser that analyzes documents at the levels of morphology, part of speech, phrases, and grammatical relations. More specifically I will describe results on authorship attribution in the context of journalists writing about the same topic (politics). A more challenging task is personality assignment on the basis of text. We constructed a corpus consisting of 145 documents describing the contents of the same documentary, written by 145 different students who also took a personality test. We show which linguistic features correlate with different dimensions of personality and the predictability of personality from these features. Finally, I will describe work on what we dubbed "stylogenetics", stylistic analysis of literary works based on the same general architecture, but using clustering as a machine learning technique rather than supervised learning. ========================================================= 5. Language Tutorial (May 24): Typological Singularities of Estonian and Their Role in Computer Analysis (Haldur Oim, University of Tartu) ========================================================= In the hope of establishing a new tradition (suggested by Koenraad de Smedt), NODALIDA 2007 will feature a tutorial on Estonian in the afternoon of May 24, immediately preceding the welcome reception. The tutorial will be free of charge for everyone who has registered for NODALIDA 2007. ========================================================= 6. Registration and Payment/Fees ========================================================= Only ELECTRONIC REGISTRATION on the conference web-site is available, use registration form on http://math.ut.ee/nodalida2007/nodaregistration.html. CONFERENCE FEE (includes a CD with conference proceedings, refreshments during coffee breaks, a conference reception on May 24 and a conference dinner on May 25). Accommodation is NOT included. IMPORTANT: TO PARTICIPATE IN A WORKSHOP YOU HAVE TO REGISTER FOR THE MAIN CONFERENCE AND TO PAY THE CONFERENCE FEE. FEES Regular participants: Regular registration (before 20 April): 130 EUR or 2035 EEK Late registration (before 15 May): 150 EUR or 2345 EEK On-site registration: 170 EUR or 2660 EEK Student participants: 70 EUR or 1095 EEK Payment is by bank transfer or credit card until May 15, see details on http://math.ut.ee/nodalida2007/. IMPORTANT: When you are paying ON-SITE then we can accept ONLY Estonians Kroons and ONLY CASH. Nationals of some countries must have a visa to enter Estonia, check visa information on http://math.ut.ee/nodalida2007/. ======================================================= 7. Conference Program ======================================================= Conference program can be found on the conference website http://math.ut.ee/nodalida2007/ (Program) ======================================================== 8. Instructions for Presenters ======================================================== Instructions for presenters can be found on the http://math.ut.ee/nodalida2007/ (Information for authors) Oral presentations The time alotted to each presentation is 20 minutes, followed by 5 minutes for questions and discussion, and 5 minutes to allow people to change rooms. The session chairs will be instructed to enforce time constraints strictly, so make sure to be well prepared. Make sure to show up at least 15 minutes before your session starts, so that you can copy your presentation onto the computer in the room or (if this is necessary for some reason) to connect your own laptop and check that everything works. If you need other equipment than a computer and beamer, or plan to use presentations in other formats than PowerPoint or PDF, please contact the local organizers well in advance by sending an e-mail. Poster presentations The poster sessions are scheduled at the end of the first day, and you are required to be available for questions and discussion during one hour, either 16.00-17.00 or 17.00-18.00. Please check the program. However, we recommend you to put up your poster already in the morning, so that people attending the conference can check it out before the session starts. You can also leave it up during the second day, although you are responsible for taking it down before you leave. Please make sure to place your poster at the correct board - the boards will be labelled with the name(s) of author(s) and the title. The recommended size of a poster is A1 (59x84 cm), orientation: portrait. If you have questions about practical matters, please contact the local organizers by sending an e-mail. ========================================================= 9. Accommodation and Dining ========================================================= A list of Tartu hotels can be found on the webpage http://www.visittartu.com/1424 A list of places where to eat: http://www.visittartu.com/1434 Maps: http://www.tartu.ee/?lang_id=2 Weather in Tartu: http://www.tartu.ee/?lang_id=2 http://www.weather.ee/ ========================================================= Looking forward to seeing you in Tartu! NODALIDA 2007 local organizers