BEGIN:VCALENDAR VERSION:2.0 METHOD:PUBLISH X-WR-CALNAME: Calendar | Centre for Language Technology\, Gothenburg PRODID:-//Drupal iCal API//EN BEGIN:VEVENT UID:calendar:155:field_eventdate:0:0 SUMMARY:Guest lecture by Paola Merlo Geneve DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20080530T151500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20080530T170000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2008-05-30/guest-lecture-paola-merlo-geneve LOCATION:Humanisten F236 DESCRIPTION:One of the currently debated topics in Natural Language Processing is the \n problem of semantic role labelling. Given a sentence like "I want to reserve \n a flight from Geneva to Gotenborg''\, how do we determine automatically that \n "from Geneva'' is the source of the flight and "to Gotenborg'' is the \n destination? The solution to this problem lies at the heart of all \n applications that require language understanding\, such as dialogue systems\, \n question answering\, or machine translation. Recent successes of machine \n learning methods in statistical parsing and lexical acquisition pave the way \n to a learning approach for this problem too. In this talk\, I will report \n experiments that explore learning of syntactic and semantic representations. \n First\, we extend a state-of-the-art statistical parser to produce a richly \n annotated tree that identifies and labels nodes with semantic role labels as \n well as syntactic labels. Secondly\, we explore rule-based and learning \n techniques to extract predicate-argument structures from this enriched \n output. The learning method is competitive with previous single-system \n proposals for semantic role labelling\, yields the best reported precision\, \n and produces a rich output. In combination with other high-recall semantic \n role labelling systems it yields an F-measure of 81%. END:VEVENT BEGIN:VEVENT UID:calendar:153:field_eventdate:0:1 SUMMARY:Robert Dale (Macquarie University\, Australia): Referring Expression \n Generation: What Can We Learn From Human Data? DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20090611T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20090611T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2009-06-11/robert-dale-macquarie-university-australia-referring-expression-generation-what-can-we-learn-human-d DESCRIPTION:Abstract: In this talk\, I reflect on what we can learn about the processes \n involved in the generation of referring expressions by looking at a corpus of \n human-produced data. We find that the data vastly underspecifies what might \n be involved algorithmically\, although it does rule out a number of popular \n algorithms for referring expression generation as candidates for models of \n what people do. We posit an alternative algorithmic schema which forces us to \n focus more clearly on the questions that need to be answered experimentally \n if we are to develop algorithms that emulate human behaviour on this task. END:VEVENT BEGIN:VEVENT UID:calendar:142:field_eventdate:0:2 SUMMARY:Visit by Ivan Sag DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20090922T143500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20090923T143500 URL;VALUE=URI:http://www.clt.gu.se/visit/2009-09-22/visit-ivan-sag LOCATION:Gothenburg DESCRIPTION:Ivan Sag\, Stanford University\, will visit Gothenburg on 22nd and 23rd \n September\n \n Ivan Sag\, [1]Sadie Dernham Patek Professor in Humanities [2] and Professor of \n Linguistics [3]\, Stanford University\,  [4]Senior Researcher\, Center for the \n Study of Language and Information [5] (CSLI)\, will visit Gothenburg 22nd-23rd \n September.  He will give two talks\, both listed at CLT events.\n \n  \n \n Ivan A. Sag (PhD\, MIT 1976) is the Sadie Dernham Patek Professor in\n Humanities and Professor of Linguistics and Symbolic Systems at\n Stanford University. A fellow of the American Academy of Arts and\n Sciences and the Linguistic Society of America\, in 2005 he received\n the LSA's Victoria Fromkin Prize for distinguished contributions to\n the field of linguistics.\n \n Sag's research focuses on syntax\, semantics\, and language processing.\n He is the author or coauthor of 10 books\, as well as over 100 articles\n in venues as diverse as /Linguistics & Philosophy/ and /Nature\n Neuroscience/.  A member of the research teams that invented and\n developed GPSG and HPSG\, Sag's current research primarily concerns\n constraint-based\, lexicalist models of grammar\, and their relation to\n theories of language processing.\n \n Additional facts:\n \n Director (2005-2009): Stanford's Symbolic Systems Program\n (interdisciplinary program in cognitive and information science).\n \n Senior Researcher: Stanford's Center for the Study of Language and\n Information (CSLI).\n \n Professor /Honoris Causa/\, University of Bucharest\n \n Fellow: Center for Advanced Study in the Behavioral Studies\n (2002-2003) Research School Logica - the Netherlands (1994)\,\n Ameritch Fellow: University of Chicago (1987-88).\n \n He began his linguistic career studying Indo-European and Sanskrit at\n The University of Rochester (BA - 1971) and The University of\n Pennsylvania (MA - 1973).\n \n He also serves on the editorial boards of: /Journal of Linguistics/\,\n /Linguistics and Philosophy/\, /Research on Language and Computation/\,\n /Constructions/\, /Semantics and Pragmatics/.\n \n \n [1] http://lingo.stanford.edu/sag/\n [2] http://lingo.stanford.edu/sag/patek.html\n [3] http://www-linguistics.stanford.edu/linguistics/\n [4] http://www.stanford.edu/\n [5] http://www-csli.stanford.edu/csli/index.shtml END:VEVENT BEGIN:VEVENT UID:calendar:145:field_eventdate:0:3 SUMMARY:CLT seminar by Ivan A. Sag DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20090923T151500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20090923T171500 URL;VALUE=URI:http://www.clt.gu.se/seminar/2009-09-23/clt-seminar-ivan-sag LOCATION:L308 DESCRIPTION:*The Interplay of Grammatical Theory and Language Engineering: The Evolution \n of HPSG Ivan A. Sag (Stanford University) *\n \n In this talk I'll provide one person's perspective on the evolution of\n HPSG from its GPSG origins to the construction-based incarnation\n (Sign-Based Construction Grammar) that blends in aspects of Berkeley\n Construction Grammar. Since its inception\, the development of HPSG\n theory has been influenced by on-going attempts to make it\n computationally real. Language engineering\, where matters of detail\n and scale\, as well as the demand for analytic precision\, are of\n paramount concern\, has had an overall positive effect on the\n development of HPSG\; but it has had negative consequences as well.  In\n the process of telling this story\, I'll also discuss some current work\n in SBCG.\n \n home page [1]\n \n \n [1] http://lingo.stanford.edu/sag/ END:VEVENT BEGIN:VEVENT UID:calendar:144:field_eventdate:0:4 SUMMARY:CLT Seminar: Ruben the Robot DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20091001T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20091001T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2009-10-01/clt-seminar-ruben-robot LOCATION:T225 (The computer lab at Hovrätten) DESCRIPTION:*A demonstration of a speech-controlled LEGO robot\, built by students of the \n Language Technology Programme. NOTE the unusual location\, the computer lab \n (T225) in the FLOV department!*\n \n During the Project Course at the Language Technology Programme\, a group of \n students constructed and implemented a speech-controlled robot\, called Ruben \n the Robot. Ruben is built in LEGO Mindstorms NXT\, connected via Bluetooth to \n a computer with speech synthesis and speech recognition. The dialogue system\, \n synthesis\, recognition\, and LEGO controller\, are all implemented in Python. \n \n The seminar will be presented by a non-empty subset of the project members \n {Kim Demaret\, Jonatan Uppström\, Håkan Edqvist\, Martin Hammarstedt}. \n \n Ruben also has a Swedish webpage: http://www.cling.gu.se/~st7mhamm/ruben/ [1]\n \n \n [1] http://www.cling.gu.se/%7Est7mhamm/ruben/ END:VEVENT BEGIN:VEVENT UID:calendar:218:field_eventdate:0:5 SUMMARY:Searching for Good Search Keys using the MedEval Medical Test Collection DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20091015T100000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20091015T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2009-10-15/searching-good-search-keys-using-medeval-medical-test-collection LOCATION:L307 DESCRIPTION:I will present the intitial results of the experiments that will be\n reported in my thesis somtime next year. The MedEval test collection has\n been constructed at Språkdata\, GU\, during the last few years. It is an\n information retrieval test collection with documents from the medical\n domain. The test collection allows the user to state user group\, doctor\n or patient\, or none of these. The possibility to state user group is a\n unique feature for the MedEval test collection.\n In the initial experiments I have studied search terms with different\n features to see what features give good results\, and which features do\n not. The features studied are for example\, term broadness\, domain and\n doctor-patient language. I have also looked at compounds used as search\n keys and the compound constituents. Even though splitting compounds and\n using the constituents as search keys often is an effective strategy\, it\n is not alway the case. I will demonstrate examples of when decomposition\n works\, and when it does not.\n Welcome!\n Karin Friberg Heppin END:VEVENT BEGIN:VEVENT UID:calendar:160:field_eventdate:0:6 SUMMARY:Linguistics seminar on frame semantics and formal semantics DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20091027T131500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20091027T150000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2009-10-27/linguistics-seminar-frame-semantics-and-formal-semantics LOCATION:T340\, Gamla Hovrätten DESCRIPTION:Frames in formal semantics - Montague's temperature and Fernando's event \n strings Robin Cooper\n \n Abstract [1]\n \n More information about this event… [2]\n \n \n [1] http://docs.google.com/fileview?id=0B7zjC6M889NHMTQzYzRiMDAtZTIzMy00ZmJjLTg2MGUtMWI1ODRjZGI4OTAy&\;hl=en_GB\n [2] http://docs.google.com/Doc?docid=0AbzjC6M889NHZDlwNjhtNV8zNGY3NWZtcmRj&\;hl=en_GB END:VEVENT BEGIN:VEVENT UID:calendar:226:field_eventdate:0:7 SUMMARY:Between Abstract and Concrete Syntax: Parse Trees\, Dependency Trees\, and Word \n Alignment in GF DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20091029T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20091029T111500 URL;VALUE=URI:http://www.clt.gu.se/seminar/2009-10-29/between-abstract-and-concrete-syntax-parse-trees-dependency-trees-and-word-alignment-gf LOCATION:Department of Swedish Language\, Lennart Torstenssonsgatan 8\, L307 DESCRIPTION:Speaker: Aarne Ranta\n \n The leading idea of GF (Grammatical Framework) is the division between \n abstract and concrete syntax. The corresponding levels of representation are \n abstract syntax trees and strings. But other linguistic formalisms often work \n on levels that are between these two: phrase structure trees\, dependency \n trees\, and word alignments. In the talk\, we will show how these intermediate \n levels can be derived from GF and used in the visualization of grammars. We \n will also discuss some tentative ideas about their use in robust parsing \n (with Nivre's MaltParser) and statistical machine translation (with tools \n such as Moses). END:VEVENT BEGIN:VEVENT UID:calendar:250:field_eventdate:0:8 SUMMARY:Arto Mustajoki\, From Meaning to Form: A Model of Functional Syntax DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20091119T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20091119T101500 URL;VALUE=URI:http://www.clt.gu.se/seminar/2009-11-19/arto-mustajoki-meaning-form-model-functional-syntax LOCATION:Department of Swedish Language\, Lennart Torstenssonsgatan 8\, room L307 DESCRIPTION: \n \n  \n \n Linguists have been dreaming of the creation of a meaning-based grammar for \n generations. The first scholar to express the need for such a linguistic \n description was apparently Georg von der Gabelentz (1891). He drew a \n distinction between a synthetic and an analytic system of a language. He \n connected the synthetic system (das synthetische System) with grammatical \n synonymy: the speaker who wants to express a thought or a feeling has \n different possibilities for doing so. More or less the same aspiration was \n later expressed by dozens of linguists (Jespersen\, Bernštein\, Ščerba\, \n Bondarko\, Zolotova\, Mel’čuk\, Halliday\, Dik van Valin etc.) My own model \n of functional syntax (FS) is based on the following main principles:1. FS \n consequently follows the principle of going “from meaning to form”\; to be \n more specific\, “from semantic categories to linguistic means”. So\, the \n surface structures /John is cold\, //John has a hangover\, John is in a coma\, \n John is freezing /are all dealt with in the same chapter \n entitled /Physiological state/ because they have a similar semantic \n structure (a Physiogical state with an actant\, an Experiencer). 2. In the \n communicative situation\, FS reflects the point of view of the speaker who is \n searching for a suitable way to express his/her thoughts. However\, FS does \n not aim to describe the mental processes taking place in the speaker’s \n head. 3. The linguistic description in FS is based on *semantic \n structures*\, which reflect the state of affairs and the speaker’s comments \n on it. 4. The *states of affairs* are situations or fragments of reality \n (the real\, virtual\, or inner world) as they are interpreted by the speaker.\n \n \n ... Full abstract \n in http://tournesol.cs.chalmers.se/~aarne/talks/mustajoki.doc [1]More \n information:http://www.helsinki.fi/~mustajok/pdf/Functional_Syntax_RLJ.pdf \n [2]\n \n  \n \n \n [1] http://tournesol.cs.chalmers.se/~aarne/talks/mustajoki.doc\n [2] http://www.helsinki.fi/~mustajok/pdf/Functional_Syntax_RLJ.pdf END:VEVENT BEGIN:VEVENT UID:calendar:248:field_eventdate:0:9 SUMMARY:Mikael Roll (Lund) Grammatik\, prosodi och neurofysiologi på svenska DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20091207T151500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20091207T170000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2009-12-07/mikael-roll-lund-grammatik-prosodi-och-neurofysiologi-p%C3%A5-svenska LOCATION:L308 DESCRIPTION:Mikael Roll (Lund) kommer att tala om ERP-studier av svensk ordföljd\, \n prosodi och morfologi. Han disputerade nyligen på avhandlingen The \n Neurophysiology of Grammatical Constraints.\n \n Genom att titta på förändringar i hjärnaktivitet i realtid har Mikael \n Roll undersökt hur lyssnare använder prosodiska och pragmatiska signaler \n för att förutsäga syntaktisk och morfologisk struktur.\n En intonationsstigning i början av att-satser ökar t.ex. förväntningen \n på huvudsatsordföljd. I avsaknad av en stigning ser man en s.k. P600-effekt \n i att-satser med huvudsatsordföljd. P600 visar ombearbetning av den \n oväntade strukturen. Liknande P600-effekter ger ändelsen i ett ord där \n accent 2 bytts ut mot accent 1\, d.v.s. om t.ex. ordet 'fiskar' har den \n ordaccent som 'fisken' brukar ha. Detta tyder på att ordaccenten används \n för att för att förutsäga ordstrukturen.\n När man hör ett objekt som 'glass/glassen/den' i en sats som 'Han åt \n glass/glassen/den...'\, verkar både dess form och innehåll skapa olika typer \n av förväntning på kommande ord\, vilket ger upphov till olika effekter för \n olika sorters objekt i meningar av typen 'Han åt glass/glassen/den inte'.\n \n http://www.sol.lu.se/lingvistik/person/MikaelRoll [1]\n \n \n [1] http://www.sol.lu.se/lingvistik/person/MikaelRoll END:VEVENT BEGIN:VEVENT UID:calendar:222:field_eventdate:0:10 SUMMARY:Symbols\, Meaning and Statistics DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20091210T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20091210T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2009-12-10/symbols-meaning-and-statistics LOCATION:Department of Swedish Language\, Lennart Torstenssonsgatan 8\, room L307 DESCRIPTION:Presenter:     Richard Sproat\, Oregon Health & Science University\n \n Of all artifacts left behind by ancient cultures\, few are as evocative as \n symbols. When one looks at an inscription on stone or clay\, it is natural to \n ask: What does it mean? Was this a form of writing\, or some sort of \n non-linguistic system? If it was writing\, can we hope to decipher it?\n \n In this talk\, I examine these and related questions\, and the possible role of \n statistical methods in answering them. I start with a brief review of the \n history of successful decipherment\, and computational approaches to \n decipherment.  In the bulk of the talk\, I discuss a topic that made its \n rounds of the popular science press earlier this year: Namely\, the claimed \n "entropic evidence" that the 4000-year-old Indus Valley symbol system \n constituted a script. I present simple counterevidence to the usefulness of \n the proposed measure to support any such claim\; and I review the large amount \n of archaeological and comparative cultural evidence against the script \n hypothesis\, which must in any case be taken into account in any complete \n discussion of this subject.\n \n (Portions of this talk are based on joint work with Steve Farmer and Michael \n Witzel.) END:VEVENT BEGIN:VEVENT UID:calendar:221:field_eventdate:0:11 SUMMARY:Finite Automatas as a Model for Natural Languages DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20091210T150600 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20091210T170000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2009-12-10/finite-automatas-model-natural-languages LOCATION:EDIT Room\, 3364\, Chalmers\, DESCRIPTION:Kimmo Koskenniemi from Helsinki is visiting us and will give this talk END:VEVENT BEGIN:VEVENT UID:calendar:286:field_eventdate:0:12 SUMMARY:Seminar planning meeting DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100114T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100114T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-01-14/seminar-planning-meeting LOCATION:L307 DESCRIPTION:CLT seminar planning meeting for the spring semester END:VEVENT BEGIN:VEVENT UID:calendar:306:field_eventdate:0:13 SUMMARY:CLT Seminar: Ramona Enache DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100211T103000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100211T113000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-02-11/clt-seminar-ramona-enache LOCATION:EDIT Room\, 3364\, Hörsalsvägen\, Chalmers DESCRIPTION:Reasoning and Language Generation in the SUMO Ontology\n \n Abstract: All popular ontology description languages like OWL\, RDFS\n and KIF don't have type systems. This means that it is hard to\n guarantee that the information in the ontology is consistent. One of\n the contributions of the thesis is that we explore how to model\n ontologies in a strictly typed language like GF. The test case of the\n work is SUMO - the biggest open source ontology available today. This\n has several advantages: first inconsistencies are quickly discovered\,\n and second\, since GF is both a logical framework and a powerful natural\n language generation system\, this ontology has a direct verbalization.\n The original ontology has some NLG rules that are quite primitive\, but\n are used as the starting point for building the concrete syntax in GF for\n English. Furthermore experiments were done in the generation of natural\n language in Romanian and French. Since we don't have enough data for\n complete verbalization\, the experiments are just intended to show that\n our approach is substantially better for these languages. From the\n abstract syntax in GF it is possible also to generate axioms in TPTP\n (the language used in the state of the art automated theorem provers)\n which means that we can connect GF to theorem provers in order to check\n the ontology for logical consistency.\n \n   END:VEVENT BEGIN:VEVENT UID:calendar:289:field_eventdate:0:14 SUMMARY:CLT Seminar: Mats Lundälv\, DART DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100218T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100218T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-02-18/clt-seminar-mats-lund%C3%A4lv-dart LOCATION:L307 DESCRIPTION:This seminar is about people with communication disabilities\, and how \n language technology can be of help. More specifically\, Mats Lundälv will \n present the EU project ÆGIS – /Open Accessibility Everywhere: \n Groundwork\, Infrastructure\, Standards/:\n \n "The ÆGIS project seeks to determine whether 3rd generation access \n techniques will provide a more accessible\, more exploitable and deeply \n embeddable approach in mainstream ICT (desktop\, rich Internet and mobile \n applications)."\n \n Mats is working at DART – Centre for Augmentative and Alternative \n Communication (AAC) and Assistive Technology (AT).\n \n For more information:\n \n  \n \n * DART: http://www.dart-gbg.org [1]\n * AEGIS: http://www.aegis-project.eu [2]\n \n  \n \n \n [1] http://www.dart-gbg.org/\n [2] http://www.aegis-project.eu END:VEVENT BEGIN:VEVENT UID:calendar:312:field_eventdate:0:15 SUMMARY:LLT seminar DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100219T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100219T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-02-19/llt-seminar LOCATION:T116\, Gamla Hovrätten (FLoV)\, Olof Wijksgatan 6 DESCRIPTION:The first in a new series of seminars on Logic and Language Technology.  It \n will consist of an overview and discussion of logic's role in CLT and a \n number of short presentations of relevant work that is going on or could be \n planned for the future.  See http://clt.gu.se/page/llt-seminars [1]\n \n \n [1] http://clt.gu.se/page/llt-seminars END:VEVENT BEGIN:VEVENT UID:calendar:298:field_eventdate:0:16 SUMMARY:CLT Seminar: Swedish FrameNet++ DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100225T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100225T101500 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-02-25/clt-seminar-swedish-framenet LOCATION:L307 DESCRIPTION:In this seminar\, we will talk about our work on an integrated lexical \n resource for Swedish language technology. We call this resource the Swedish \n FrameNet++\, the "++" part of the name reflecting the fact that it will \n include much more information than is normally found in a framenet.\n \n Lars Borin\n Dana Dannélls\n Markus Forsberg\n Maria Toporowska Gronostaj\n Dimitrios Kokkinakis END:VEVENT BEGIN:VEVENT UID:calendar:293:field_eventdate:0:17 SUMMARY:CLT Seminar: Robin Cooper DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100304T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100304T101500 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-03-04/clt-seminar-robin-cooper LOCATION:room L307\, Lennart Torstenssonsgatan 8 DESCRIPTION:*Rising Frames*\n \n In this paper I will review arguments I have made before for using frame-like \n objects in a formal semantic account of Montague's puzzles about the rising \n of prices and temperatures.  I shall then look in a little detail at the \n view of semantic coordination that Staffan Larsson and I have been working on \n and suggest that the questions we should be asking are not "What is the \n meaning of /rise?" /but "Given certain resources how will an agent exploit \n her semantic resources for /rise/ when exposed to a new utterance of the \n word?" and "How will this modify the agent's resources?".  I will suggest \n that such a view might give a simple explanation for Zipf's law concerning \n the relationship of meanings and frequency.\n \n Slides [1].\n \n \n [1] http://www.ling.gu.se/~cooper/records/rise-slides.pdf END:VEVENT BEGIN:VEVENT UID:calendar:310:field_eventdate:0:18 SUMMARY:CLT Seminar: Shafqat Virk - Urdu Resource Grammar DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100318T103000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100318T113000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-03-18/clt-seminar-shafqat-virk-urdu-resource-grammar LOCATION:room L307\, Lennart Torstenssonsgatan 8 DESCRIPTION:I have been working on Urdu Resource Grammar in GF for some time now\, I would \n like to give a talk on my work in the seminar. END:VEVENT BEGIN:VEVENT UID:calendar:324:field_eventdate:0:19 SUMMARY:LLT seminar: Robin Cooper - Generalized quantifiers and clarification content DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100319T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100319T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-03-19/llt-seminar-robin-cooper-generalized-quantifiers-and-clarification-content LOCATION:T116\, Gamla Hovrätten (FLoV)\, Olof Wijksgatan 6 DESCRIPTION:Purver and Ginzburg introduce the\n Reprise Content Hypothesis (RCH) and use it to argue for a non-generalized\n quantifier approach to certain quantifiers.  Here we will contrast\n their approach with an approach which employs a more classical\n generalized quantifier analysis and examine what predictions it has\n for possible clarifications and reexamine the data which Purver and\n Ginzburg present in the light of this. END:VEVENT BEGIN:VEVENT UID:calendar:291:field_eventdate:0:20 SUMMARY:CLT Seminar: Paul Rayson (Univ. Lancaster) - Historical text mining: to bee \n or not to be? DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100325T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100325T120000 URL;VALUE=URI:http://www.clt.gu.se/visit/2010-03-25/clt-seminar-paul-rayson-univ-lancaster-historical-text-mining-bee-or-not-be LOCATION:room L307\, Lennart Torstenssonsgatan 8 DESCRIPTION:In this talk\, I will describe our research on spelling variation in Early \n Modern English (EmodE). Our original motivation was on retraining a semantic \n field annotation system (USAS\, first designed for modern English) for \n historical corpora. Now\, the research has taken a diversion into the \n detection of historical spelling variants as they cause significant problems \n for corpus-based computational linguistics techniques and tools. I will \n quantify the extent of the problem in a variety of EmodE corpora and show how \n it affects simple procedures such as key word comparisons. Our solution will \n be described\, a corpus pre-processing tool called VARD (the Variant \n Detector)\, which uses techniques adapted from modern spell checkers. VARD \n offers modern equivalents for historical variants with high levels of \n precision and recall. The evaluation will focus on how much training data is \n required for such a system. END:VEVENT BEGIN:VEVENT UID:calendar:290:field_eventdate:0:21 SUMMARY:CLT Seminar: Markus Forsberg - The young and the not-so-old DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100408T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100408T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-04-08/clt-seminar-markus-forsberg-young-and-not-so-old LOCATION:Department of Swedish Language\, Lennart Torstenssonsgatan 8\, room L307 DESCRIPTION: \n \n I will present our ongoing work in the project CONPLISIT\, where \n we investigate how semantic search methods can be used to detect changes in \n consumption patterns and life-style in 19th century Swedish literature. To \n reach this goal we are currently linking SALDO\, a Swedish semantic lexicon\, \n with Dalin\, a Swedish 19th century dictionary\, and moreover\, adapting the \n morphological description of SALDO to Dalin. An interesting aspect of this \n work is that it adds a diachronic facet to the lexical resource SweFN++.The \n current state of the work is updated daily \n at: http://spraakbanken.gu.se/eng/research/swefn/dalin [1]\n \n  \n \n \n [1] http://spraakbanken.gu.se/eng/research/swefn/dalin END:VEVENT BEGIN:VEVENT UID:calendar:329:field_eventdate:0:22 SUMMARY:LLT seminar: Fredrik Engström - Dependence and logicality DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100416T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100416T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-04-16/llt-seminar-fredrik-engstr%C3%B6m-dependence-and-logicality LOCATION:T219\, Gamla Hovrätten (FLoV)\, Olof Wijksgatan 6 DESCRIPTION:Statements about functional dependence can be added to first-order logic with \n a simple game theoretic semantics. But there is also a model theoretic \n semantics based on "teams" instead of assignments. An investigation of the \n logical constants in this model theoretic setting has just begun. I will give \n some background on dependence logic and some on logical constants\, and also \n say something on what the future might bring. END:VEVENT BEGIN:VEVENT UID:calendar:299:field_eventdate:0:23 SUMMARY:CLT Seminar: CANCELLED (---) DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100422T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100422T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-04-22/clt-seminar-cancelled END:VEVENT BEGIN:VEVENT UID:calendar:296:field_eventdate:0:24 SUMMARY:CLT Seminar: Josef Ruppenhofer - Linking Events and Their Participants in \n Discourse (using FrameNet and Salsa) DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100506T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100506T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-05-06/clt-seminar-josef-ruppenhofer-linking-events-and-their-participants-discourse-using-framenet-and LOCATION:room L307\, Lennart Torstenssonsgatan 8 DESCRIPTION:Josef Ruppenhofer (Saarland University).\n \n In my presentation\, I will highlight the differences in the work of\n FrameNet and Salsa\, the most important of which is whether one\n proceeds through the lexicon frame-wise or lemma-wise. I will report\n on an experiment we carried out in Saarbrücken to study the usefulness\n of automatic pre-annotations for speeding up labeling.  I will also\n talk a bit about the recent Semeval shared Task that my colleague\n Caroline Sporleder and I organized on the topic of "Linking Events and\n their Participants in Discourse". This involved full-text annotation\n of extracts from Arthur Conan Doyle. These Sherlock Holmes-texts were\n annotated with frames\, coreference\, and information about the\n antecedents of null instantiated frame elements. Other recent work\n that I've been involved with is a tool for reconfiguring FrameNet\n semi-automatically\, in particular in the context of textual entailment\n recognition. END:VEVENT BEGIN:VEVENT UID:calendar:300:field_eventdate:0:25 SUMMARY:CLT Seminar: Michael McTear - Trends\, challenges and opportunities in spoken \n dialogue technology DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100520T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100520T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-05-20/clt-seminar-michael-mctear-trends-challenges-and-opportunities-spoken-dialogue-technology LOCATION:Department of Swedish Language\, Lennart Torstenssonsgatan 8\, room L307 DESCRIPTION:Michael McTear (Computer Science Research Institute\, University of Ulster)\n \n Spoken dialogue technology has developed considerably over the past thirty \n years both in terms of research activity as well as in terms of commercially \n deployed applications. In this talk I will begin with an overview of trends \n in dialogue research based on an analysis of papers that were presented at \n Eurospeech-Interspeech conferences in 1989\, 1999\, and 2009. Following this \n some challenges for dialogue researchers will be discussed\, in particular\, \n the issues faced by individual researchers and those in small groups who wish \n to build viable end-to-end dialogue systems and the problems involved in \n accessing the many resources and toolkits that are required for dialogue \n research. The talk will conclude with a brief overview of two recent PhD \n projects in dialogue at the University of Ulster and a discussion of some \n opportunities for future research involving the integration of dialogue \n technology into voice search applications and the application of spoken \n dialogue technology in ambient intelligence environments. END:VEVENT BEGIN:VEVENT UID:calendar:341:field_eventdate:0:26 SUMMARY:Rebecca Jonson thesis defence DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100522T131500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100522T150000 URL;VALUE=URI:http://www.clt.gu.se/event/2010-05-22/rebecca-jonson-thesis-defence LOCATION:T302 (Gamla hovrätten)\, Dept of Philosophy\, Linguistics and Theory of \n Science\, University of Gothenburg\, Olof Wijksgatan 6\, Gothenburg DESCRIPTION:Rebecca Jonson will defend her thesis Information state based speech \n recognition [1].  The Faculty Opponent will be Professor Michael McTear \n [2]\, University of Ulster at Jordanstown.\n \n \n [1] http://gupea.ub.gu.se/handle/2077/22169\n [2] http://www.infj.ulst.ac.uk/~cbdg23/ END:VEVENT BEGIN:VEVENT UID:calendar:292:field_eventdate:0:27 SUMMARY:CLT Seminar: Peter Ljunglöf - Grammar-based language learning DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100527T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100527T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-05-27/clt-seminar-peter-ljungl%C3%B6f-grammar-based-language-learning LOCATION:L307\, Lennart Torstenssonsgatan 8 DESCRIPTION:GRASP: Grammatikbaserad Språkinlärning (Grammar-based language learning)\n \n I am developing a pedagogical tool for children with communication \n disabilities. The tool will consist of a computer system which gives support \n for language learning and language training. The system is graphical\, where \n the child can point to words or phrases to get explanations in the form of \n symbols\, written text\, or spoken words. The child can also reformulate the \n sentences\, by moving words to new places or change the inflection of words. \n The system will then automatically reformulate the sentence\, so that it will \n always be grammatically correct. The tool is intended to stimulate - in a \n playful way - the child's will to explore the possibilities of language.\n \n GRASP is a small one-year project which started in January 2010. It is run by \n DART (Kommunikations- och dataresurscenter för personer med \n funktionshinder)\, and is financed by Sunnerdahls Handikappfond.\n \n In this talk I will report on the current status of the tool (less of this)\, \n and the underlying ideas (more of this). If you are lucky\, I might even show \n some pseudo-mathematical theories about tree similarity in type-theoretical \n grammars... END:VEVENT BEGIN:VEVENT UID:calendar:343:field_eventdate:0:28 SUMMARY:LLT seminar: Robin Cooper and Staffan Larsson - Compositionality and Dialogue \n (T116\, Gamla Hovrätten (FLoV)) DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100528T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100528T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-05-28/llt-seminar-robin-cooper-and-staffan-larsson-compositionality-and-dialogue-t116-gamla-hovr%C3%A4tten-flov LOCATION:T116\, Gamla Hovrätten\, Olof Wijksgatan 6 DESCRIPTION:We distinguish two kinds of information flow in semantic interpretation: that \n associated with compositional  interpretation (combining the meaning of \n syntactic constituents) and threading (the passing of information from left \n to right and top down in a manner related to dynamic semantics).  We shall \n argue that both kinds are relevant to sentence semantics and also possibly to \n discourse semantics.  However\, for dialogue only a threading or dynamic \n update flow seems appropriate.  The argument has to do with the fact that \n dialogues do not have semantic content in the way that sentences or \n discourses do and that many of the interesting semantic phenomena in dialogue \n are not best explained by a compositional approach. END:VEVENT BEGIN:VEVENT UID:calendar:297:field_eventdate:0:29 SUMMARY:CLT Seminar: Staffan Larsson - Accommodating innovative meaning in dialogue DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100603T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100603T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-06-03/clt-seminar-staffan-larsson-accommodating-innovative-meaning-dialogue LOCATION:L307\, Lennart Torstenssonsgatan 8 DESCRIPTION:Meaning accommodation occurs when a dialogue participant learns new \n meanings by observing utterances in context and  adapting to the way \n linguistic expressions are used. This paper sketches a formal account of \n meaning accommodation\, focusing on the detection of subjectively innovative \n uses of an expression. It also provides an example of adaptation of \n lexical meaning to accommodate innovation. END:VEVENT BEGIN:VEVENT UID:calendar:345:field_eventdate:0:30 SUMMARY:LLT seminar: Claes Strannegård - Anthropomorphic Artificial Intelligence DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100611T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100611T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-06-11/llt-seminar-claes-stranneg%C3%A5rd-anthropomorphic-artificial-intelligence LOCATION:T219 (Gamla hovrätten) DESCRIPTION:In psychology\, cognitive models\, i.e. computational models of the human mind\, \n are used for describing human cognition. In anthropomorphic artificial \n intelligence\, they are used for the purpose of problem-solving in artificial \n intelligence (AI). The basic idea is this:\n (i) start with a search-based AI algorithm \n (ii) use psychological experiments and cognitive models for approximating the \n set of solution candidates that are accessible to the human mind\, the \n anthropomorphic search space\n (iii) run the same AI algorithm restricted to the anthropomorphic search \n space. \n This yields an algorithm for finding anthropomorphic solutions to the \n original problem. While the original problem may be unsolvable\, the \n anthropomorphic version of it is solvable whenever the anthropomorphic search \n space is finite. Transposing to the anthropomorphic version of the problem \n  is adequate whenever there is a boundary condition on the problem that the \n solutions produced must be psychologically simple. \n I will show how anthropomorphic methods can be used for solving problems in a \n wide range of areas including formal verification\, number progression \n analysis\, progressive matrix analysis\, and natural language generation. I \n will also discuss potential applications to machine generated music and art. END:VEVENT BEGIN:VEVENT UID:calendar:351:field_eventdate:0:31 SUMMARY:Andrew Gargett visit DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100823T090000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100827T170000 URL;VALUE=URI:http://www.clt.gu.se/visit/2010-08-23/andrew-gargett-visit DESCRIPTION:Andrew Gargett (Saarland University) will be visiting us 23-27/8.  If you \n would like to organize time to meet with him contact Ellen Breitholtz \n (ellen@ling.gu.se [1]).\n \n \n [1] mailto:ellen@ling.gu.se END:VEVENT BEGIN:VEVENT UID:calendar:350:field_eventdate:0:32 SUMMARY:Linguistics seminar: Andrew Gargett - Can names be translated\, or must they \n be borrowed? DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100824T131500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100824T150000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-08-24/linguistics-seminar-andrew-gargett-can-names-be-translated-or-must-they-be-borrowed LOCATION:T340\, Gamla hovrätten DESCRIPTION:Can names be translated\, or must they be borrowed?\n Andrew Gargett\n Saarland University\n \n We might be tempted to answer 'borrowed'\, if our account presumed that\n names were a well-defined class of linguistic objects\, largely empty\n of content\, yet linked to (potentially rich) associated information\n about some individual name-bearer (e.g. Anderson 2004). Such an\n account would be consistent with the intuitively appealing\, and\n well-known idea that names are rigid designators (Kripke 1981)\, which\n is to say\, labels designating once and for all some unique\n object/entity. On the other hand\, the cross-linguistic diversity in\n forms of person reference (e.g. possessed kin terms rival personal\n names as the default form of naming across languages\, Stivers et al.\n 2007)\, suggests that languages offer a much wider range of strategies\n for naming than sometimes assumed\, suggesting some potential for\n translatability. Indeed\, forms of referring in linguistic interaction\n are far more dynamic and complex than orthodox accounts of names seem\n able to deal with\, with a wide variety of alternative forms of entity\n reference\, and choices between such alternative forms often driven by\n context-specific factors (even being conditioned by such\n partner-specific information as levels of expertise\, e.g. Isaacs &\n Clark 1987).\n \n All of this suggests the need for a more general account of person\n reference\, grounded in actual situations of language use. Blending\n ideas from anthropological linguistics (Enfield and Stivers 2007)\,\n cognitive psychology (Valentine et al. 1996\, Isaacs & Clark 1987)\, and\n formal linguistics (von Heusinger 2004\, McCawley 1996)\, in particular\,\n a dynamically oriented model of grammar\, Dynamic Syntax (Cann et al.\n 2005)\, Gargett (2010) presented a fresh take on this: (i) that\n enriching these underspecified forms of person reference is a\n potentially unbounded process\, drawing on either immediate\n (linguistic) context or else longer term (cognitive) context (such as\n images\, personal recollections\, and the like)\, and (ii) that\n enrichments may be stored and reused on subsequent occasions for\n referring to the same individual\, leading to the idea that names are\n "routine" designators. This talk will seek to further motivate this\n proposal by way of examining patterns across languages of switching\n between alternative forms of naming (e.g. during dialogue). END:VEVENT BEGIN:VEVENT UID:calendar:348:field_eventdate:0:33 SUMMARY:CLT Seminar: Andrew Gargett - Routines\, context\, and figurative language in \n dialogue DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100826T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100826T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-08-26/clt-seminar-andrew-gargett-routines-context-and-figurative-language-dialogue LOCATION:L307\, Lennart Torstenssonsgatan 8 DESCRIPTION:*Andrew Gargett (Saarland University): Routines\, context and figurative \n language in dialogue*\n \n Abstract: This talk presents ongoing work investigating the use of grammars \n for modelling core dialogue phenomena (clarifications\, reformulations\, \n corrections\, and the like)\, in particular\, employing the grammar formalism of \n Dynamic Syntax (DS\, Kempson et al. 2001). I will first outline an extension \n of the DS account of dialogue (e.g. Purver et al. 2006) to explain \n routinisation via a dual process model. Then I will present recent work on \n extending this to account for the interaction between the linguistic and \n conceptual levels\, focusing on the use of figurative language to resolve \n uncertainty in dialogue (e.g. Bavelas et al. 2008). Now\, if we presume that \n the linguistic level reflects an underlying conceptual model\, then the \n emergence of linguistic routinisation over the course of interaction reflects \n routinised conceptualisation of the task at hand. So it is interesting to \n consider how interlocutors might respond when difficulty arises with how a \n task is being conceptualised (difficulty may be environmental\, such as \n restricted visual cues as when on the telephone\, or interactional\, such as \n when an interlocutor misunderstands a partner's contribution). I will suggest \n that\, faced with such difficulty\, figurative language can be used for \n re-conceptualisation\, as a way of shifting the conceptual ground of the \n current focus of activity\, and consequently\, as a strategy for re-grounding \n understanding of the interaction. The resulting formal model invites \n reconsideration of what has been termed the "groundedness" of figurative \n language (Gibbs 2006)\, in terms of its role in dialogue. Along the way\, I \n connect DS to cognitive linguistics and artificial intelligence\, in \n particular\, demonstrating how combining the DS grammar formalism with \n ATT-Meta\, an AI system for modelling the use of figurative language as \n reasoning-by-simulation (e.g. Barnden 2009)\, goes some way toward modelling \n how interlocutors can employ conceptual metaphor to cope with uncertainty. END:VEVENT BEGIN:VEVENT UID:calendar:349:field_eventdate:0:34 SUMMARY:CLT Seminar: Kalervo Järvelin - Managing Morphologically Complex Languages \n in Information Retrieval DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100902T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100902T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-09-02/clt-seminar-kalervo-j%C3%A4rvelin-managing-morphologically-complex-languages-information-retrieval LOCATION:L307\, Lennart Torstenssonsgatan 8 DESCRIPTION:*Kalervo Järvelin (Tampere): Managing Morphologically Complex Languages in \n Information Retrieval*\n \n English is a morphologically simple\, phrase-oriented language. Its treatment \n in mono- and cross-language IR is fairly simple\, save for possible phrase \n recognition. Over the years\, much software has been developed for English IR. \n However\, many other languages have become\, or are becoming\, essential for IR \n for example in the Web. All of these languages are not as simple as English \n but may lack many computational resources available for English. The main \n challenges on IR posed by many languages relate to (a) lexicons: \n out-of-vocabulary words used in texts but not listed in dictionaries\; (b) \n morphology: word inflection and compounds. The talk discusses methods for \n managing these challenges in mono-lingual and cross-language IR in \n morphologically complex languages both at the indexing stage and the search \n stage. The methods include stemming\, lemmatization\, compound splitting\, \n compound translation\, inflectional form generation\, structured queries\, and \n approximate string matching. END:VEVENT BEGIN:VEVENT UID:calendar:355:field_eventdate:0:35 SUMMARY:Planning seminar DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100909T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100909T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-09-09/planning-seminar LOCATION:L307\, Lennart Torstenssonsgatan 8 DESCRIPTION:We get together and plan the CLT seminar series for the autumn semester. END:VEVENT BEGIN:VEVENT UID:calendar:360:field_eventdate:0:36 SUMMARY:CLT seminar: Report from LREC DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100916T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100916T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-09-16/clt-seminar-report-lrec LOCATION:L307\, Lennart Torstenssonsgatan 8 DESCRIPTION: \n \n Report from LREC 2010 [1]\n \n Dimitrios Kokkinakis\, Lars Borin and Markus Forsberg\n \n \n [1] http://www.lrec-conf.org/lrec2010/ END:VEVENT BEGIN:VEVENT UID:calendar:371:field_eventdate:0:37 SUMMARY:Trial lectures DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100930T132500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20100930T143500 URL;VALUE=URI:http://www.clt.gu.se/event/2010-09-30/trial-lectures LOCATION:Chalmers Johanneberg\, EDIT room DESCRIPTION:Public trial lectures by the applicants for the CLT sponsored position as \n assistant professor with tenure track at Computer Science and Engineering.\n \n Schedule:\n \n 13:25 Maria Liakata\n \n 13:50 Peter Ljunglöf\n \n 14:15 Kimmo Kettunen\n \n   END:VEVENT BEGIN:VEVENT UID:calendar:370:field_eventdate:0:38 SUMMARY:CLT Seminar: Stefan Schulz - Automated codification of medical documents DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101008T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101008T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-10-08/clt-seminar-stefan-schulz-automated-codification-medical-documents LOCATION:L307\, Lennart Torstenssonsgatan 8 DESCRIPTION:Stefan Schulz (Freiburg) - Automated codification of medical documents\n \n Abstract:\n \n The advance of electronic health records not only in industrialized countries \n but also in the developing world has made large amounts of medical narratives \n electronically available. These texts exhibit special characteristics\, such \n as un- and paragrammatical sentences\, the use of acronyms and abbreviations\, \n and the abundance of typing errors.\n \n In a case study from a Brazilian university hospital\, selected content of \n medical discharge summaries in Portuguese was automatically mapped to SNOMED \n CT\, a comprehensive medical terminology. SNOMED CT provides (chiefly \n multi-word) English and Spanish terms. In order to optimally align text \n sequences and terms\, token chains that exhibited term-typical POS tags were \n selected from the source. These term candidates\, together with all SNOMED \n terms\, were  submitted to a morphosemantic normalization process\, which \n consists of the extraction of meaningful subwords and their mapping to a \n language-independent interlingua.\n \n The accuracy of the mapping\, measured against a gold standard built by two \n domain experts (kappa 0.89) ranged from 0.66 for two-word terms to 0.89 for \n five-word terms.\n \n Schulz on the web [1]\n \n \n [1] http://www.imbi.uni-freiburg.de/biom/index.php?showEmployee=schulz END:VEVENT BEGIN:VEVENT UID:calendar:357:field_eventdate:0:39 SUMMARY:PhD thesis defence - Karin Friberg Heppin DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101009T100000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101009T130000 URL;VALUE=URI:http://www.clt.gu.se/event/2010-10-09/phd-thesis-defence-karin-friberg-heppin LOCATION:Lilla hörsalen\, Humanisten\, Renströmsgatan 6 DESCRIPTION:*Karin Friberg Heppin will defend her PhD thesis:*\n \n /*Resolving power of search keys in MedEval\, a Swedish medical text \n collection with user groups: doctors and patients*/\n \n Opponent: \n \n Professor Stefan Schulz\, Freiburg \n \n Committee: \n Professor Barbara Gawronska\, Agder \n Professor Karin Kjellgren\, Göteborg \n Docent Magnus Merkel\, Linköping \n \n Chairman: \n Professor Lars Borin END:VEVENT BEGIN:VEVENT UID:calendar:359:field_eventdate:0:40 SUMMARY:CLT Seminar: Lars Borin & Markus Forsberg - Beyond the synset: Swesaurus - a \n fuzzy Swedish wordnet DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101014T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101014T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-10-14/clt-seminar-lars-borin-markus-forsberg-beyond-synset-swesaurus-fuzzy-swedish-wordnet LOCATION:L307\, Lennart Torstenssonsgatan 8 DESCRIPTION:/Swesaurus/ is a free Swedish wordnet currently under construction in our \n research unit. Swesaurus is made up in part by a combination of a number of \n pre-existing freely available lexical resources. Two central resources for \n the purposes of this presentation are SALDO and Synlex.\n \n SALDO is a full-scale Swedish lexical-semantic resource with non-classical\, \n associative relations among word and multiword senses. The senses in SALDO \n are identified by carefully designed persistent formal identifiers\, and for \n this reason\, SALDO has become the “pivot” resource of all our \n computational lexicon activities\, including Swesaurus. We will say something \n about the relationship of SALDO’s associative relations to classical \n lexical-semantic relations and how we intend to incorporate the SALDO \n relations in Swesaurus.\n \n Synlex is a graded Swedish synonym list created by asking members of the \n public – users of an online Swedish-English dictionary – to judge the \n degree of synonymy of a random\, automatically generated synonym pair \n candidate.\n \n We will describe our experiments turning the graded synonymy relations of \n Synlex into /fuzzy synsets/ in Swesaurus. The introduction of fuzziness into \n a wordnet raises many intricate methodological and theoretical questions\, \n e.g.\, if w1 is a graded synonym of w2 \, and w1 is a hyponym of w3 \, what is \n the relation between w2 and w3. Similarly\, if wa is a synonym of degree 0.75 \n of wb and wb is a synonym of degree 0.9 of wc \, what – if any – is the \n degree of synomymy between wa and wc?\n \n If synonymy is seen as an all-or-none transitive and symmetric relation\, the \n construction of synsets from synonym pairs is arguably straightforward: We \n can simply compute the transitive closure of the synonymy relation.\n \n When graded synonymy enters the picture\, which method to use for collecting \n synonym pairs into synsets becomes much less obvious\, and especially how to \n assign a degree of synonymy to “derived” pairs\, i.e.\, pairs not in the \n original list. We have experimented with two different approaches for turning \n Synlex synonym pairs into fuzzy synsets in Swesaurus\, transitive closure and \n clique formation. We will present the outcomes of these experiments and \n discuss the merits and disadvantages of the two methods. END:VEVENT BEGIN:VEVENT UID:calendar:356:field_eventdate:0:41 SUMMARY:CLT Seminar: Srikant Sarangi - Communicatively managing risk and uncertainty \n in healthcare encounters DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101021T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101021T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-10-21/clt-seminar-srikant-sarangi-communicatively-managing-risk-and-uncertainty-healthcare-encounters LOCATION:K333\, Lennart Torstenssonsgatan 6 DESCRIPTION:Srikant Sarangi (Health Communication Research Centre\, Cardiff University)\n \n *Communicatively managing risk and uncertainty in healthcare encounters*\n \n The notion of risk in the healthcare setting is intimately tied up with the \n notion of uncertainty. More generally\, in contemporary societies\, experts no \n longer solve problems through a straightforward application of \n theoretical/scientific knowledge\, but increasingly engage in assessment of \n problems\, and hence by extension\, in generation of risks associated with \n events that might or might not occur. The language of risk/likelihood thus \n forms a basis for decision-making.\n \n In this presentation I examine the communicative aspects of risk/uncertainty \n formulation in the context of oncology and genetic counselling. Beginning \n with a characterisation of uncertainty in medical settings\, I first consider \n the cancer clinic where risk and uncertainty are simultaneously managed with \n regard to therapeutic intervention. Here I draw particular attention to \n ‘risks of procedure’ and the subtle ways in which ‘risk vs. risk’ and \n ‘risk vs benefit’ scenarios are navigated. I then focus on how genetic \n counsellors frame risk information in order to balance the need for accurate \n transmission of factual information as probability statements and the \n psychosocial goal of avoiding the creation of inappropriate anxieties. I \n argue that in genetic counselling both counsellors and clients orient to \n (de)escalating the overall risks – of going through the testing process and \n of developing the disease – although the counsellors’ immediate focus is \n on explicating the ‘risks of knowing’ (and disclosing) one’s genetic \n status rather than the ‘risks of occurrence’ of a genetic condition.\n \n (Link to Sarangi on the web [1])\n \n \n [1] http://cardiff.ac.uk/encap/contactsandpeople/profiles/sarangi-srikant.html END:VEVENT BEGIN:VEVENT UID:calendar:384:field_eventdate:0:42 SUMMARY:Monday seminar: Bernard Comrie - Typology of Ditransitive Constructions DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101025T151500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101025T170000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-10-25/monday-seminar-bernard-comrie-typology-ditransitive-constructions LOCATION:Lilla hörsalen\, Humanisten DESCRIPTION:*Monday seminar at the Department of Swedish*:\n \n *Bernard Comrie*\n \n Max Planck Institute for Evolutionary Anthropology [1] and\n University of California Santa Barbara [2]\n \n *Typology of Ditransitive Constructions*\n \n A ditransitive construction can be defined as a construction including an \n agent (A) argument\, a recipient (R) argument\, and a theme (T) argument\, as in \n "Mary (A) gives a book (T) to John (R)". While some languages express \n corresponding content without using a three-place predicate (e.g. by using a \n serial verb construction)\, the vast majority of the world's languages have \n three-place predicates. Such languages present different alignment types (the \n relation between ditransitive and monotransitive constructions)\, the three \n main ones being indirective (T = P ≠ R)\, secundative (T ≠ P = R)\, and \n neutral (T = P = R)\, where P is the patient-like argument of a monotransitive \n predicate. Differences in alignment can be seen clearly in flagging (marking \n by case/adpositions) and verb indexing\, i.e. coding properties\, but a variety \n of other syntactic constructions are considered that provide evidence of \n alignment of behavioral properties\, including passivization and \n relativization. Finally\, some issues of lexical variation among ditransitive \n predicates will be considered\, including from the perspective of semantic \n maps.\n \n \n [1] http://www.eva.mpg.de/lingua/staff/comrie/home.php\n [2] http://www.linguistics.ucsb.edu/people/comrie.html END:VEVENT BEGIN:VEVENT UID:calendar:372:field_eventdate:0:43 SUMMARY:Workshop on using standardized word lists in linguistic data collection DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101026T100000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101026T163000 URL;VALUE=URI:http://www.clt.gu.se/event/2010-10-26/word-lists LOCATION:Lilla hörsalen\, Humanisten DESCRIPTION:The aim of this workshop is to discuss methodogical issues connected with the \n use of standardized word lists - and more generally: questionnaires - in \n primary data collection for linguistic research. The workshop is a joint \n arrangement by the project partners (University of Gothenburg\, Uppsala \n University and the Max Planck Institute for Evolutionary Anthropology\, \n Leipzig).\n \n For details and information about how to register\, see the workshop webpage:\n \n \n \n Welcome!\n \n Anju Saxena and Lars Borin\n \n \n [1] http://spraakbanken.gu.se/eng/research/all-projects/digital-areal-linguistics/workshop-october-2010 END:VEVENT BEGIN:VEVENT UID:calendar:361:field_eventdate:0:44 SUMMARY:Swedish Language Technology Conference (SLTC-2010) DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101028T000000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101029T000000 URL;VALUE=URI:http://www.clt.gu.se/event/2010-10-28/swedish-language-technology-conference-sltc-2010 LOCATION:Linköping University DESCRIPTION:Welcome to the Third Swedish Language Technology Conference (SLTC-2010 [1])\, \n to be held at Linköping University\, October 28-29.\n \n \n [1] http://www.ida.liu.se/conferences/sltc2010/ END:VEVENT BEGIN:VEVENT UID:calendar:383:field_eventdate:0:45 SUMMARY:GF meets SMT DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101101T000000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101102T000000 URL;VALUE=URI:http://www.clt.gu.se/event/2010-11-01/gf-meets-smt LOCATION:Chalmers DESCRIPTION:The UPC [1] team experts in Statistical Machine Translation are visiting the \n GF [2] group to work on the MOLTO [3] project's task of making GF more robust \n by SMT methods. The first days of this visit are devoted to introductory \n lectures\, suitable to a general audience and are therefore open to the \n public.\n \n Please follow the web page for forthcoming information:\n \n http://www.molto-project.eu/event/gf-smt-2010 [4]\n \n \n [1] http://www.molto-project.eu/Universitat%20Polit%C3%A8cnica%20de%20Catalunya\n [2] http://www.grammaticalframework.org/\n [3] http://www.molto-project.eu/\n [4] http://www.molto-project.eu/event/gf-smt-2010 END:VEVENT BEGIN:VEVENT UID:calendar:347:field_eventdate:0:46 SUMMARY:CLT Seminar: Ellen Breitholtz - Enthymematic inference in dialogue DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101104T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101104T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-11-04/clt-seminar-ellen-breitholtz-enthymematic-inference-dialogue LOCATION:L307\, Lennart Torstenssonsgatan 8 DESCRIPTION:Abstract:\n \n In dialogue we usually neither spell out our arguments\, nor present every \n step necessary to reach a logical conclusion or justify a proposal. These \n semi–logical arguments\, which are fundamental to rhetoric\, are what \n Aristotle calls enthymemes. The few linguistic studies that  have been \n carried out on the topic indicate that enthymemes also play an important role \n in spontaneous dialogue. However\, there are no precise accounts of what that \n role actually is.\n \n We would like to provide such an account\, and thereby add a rhetorical \n dimension to dialogue semantics. We hope to show that an analysis taking the \n enthymeme as its point of departure elucidates some aspects of meaning in \n spoken language that cannot be explained in terms of\, for example\, questions\, \n as has been suggested by Ginzburg et al\, who introduced the idea that \n questions under discussion  provide coherence in dialogue and drive it \n forward (cf. 1).\n \n First\, we give some background on enthymemes in general and enthymemes in \n dialogue illustrated by dialogue examples. Then we move on to discuss how \n enthymemes can play a role in explaining why we need informational redundancy \n in dialogue and how the enthymeme can help us choose which apparently \n redundant utterances we should add to make a dialogue run smoothly (cf. 2 & \n 3).\n \n Finally\, we give a tentative formal account of the enthymeme using an \n information state update approach\, where an agent’s information state is \n described after each dialogue move (cf. 4). We will also discuss the idea of \n enthymemes as instantiations of topoi which an agent involved in dialogue has \n at its disposal. These could be regarded as a rhetorical resource\, similar to \n the way grammatical and lexical competence has been described as resources \n available to an agent\, as envisaged by Cooper & Ranta (cf. 5).\n \n  \n (1) Ginzburg\, J. The Interactive Stance: Meaning for Conversation. 2010\, \n http://www.dcs.kcl.ac.uk/staff/ginzburg/papers-new.html [1]\n \n (2)  Walker\, M. The effect of resource limits and task complexity on \n collaborative planning in dialogue.  Artificial Intelligence 85. 1996.\n \n (3) Breitholtz\, E. & Villing\, J. Can Aristotelian Enthymemes Decrease the \n Cognitive Load of a Dialogue System User? Proceedings of LonDial 2008\, the \n 12th SEMDIAL workshop. 2008.\n \n (4) Larsson\, S. & Traum\, D. The Information State Approach to Dialogue \n Management. p. 325-353 Smith & Kuppevelt (eds.): Current and New Directions \n in Discourse & Dialogue. Kluwer Academic Publishers. 2003.\n \n (5) Cooper\, R & Ranta\, A. Natural Languages as Collections of Resources. p. \n 109-120 Language in Flux: Dialogue Coordination\, Language Variation\, Change \n and Evolution. Eds. Cooper\, R. & Kempson\, R. 2008.\n \n \n [1] http://www.dcs.kcl.ac.uk/staff/ginzburg/papers-new.html END:VEVENT BEGIN:VEVENT UID:calendar:354:field_eventdate:0:47 SUMMARY:CLT Workshop DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101108T083000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101109T150000 URL;VALUE=URI:http://www.clt.gu.se/event/2010-11-08/clt-workshop LOCATION:Gullmarsstrand DESCRIPTION:------------------------------------------------------------------------------\n The first annual workshop for CLT members will take place at Gullmarsstrand \n [1] November 8-9. We will leave by bus from Humanisten at 08.30 on Monday \n morning\, and expect to be back in Gothenburg by 3 pm on Tuesday. Below\, you \n can find a preliminary schedule and titles for the presentations. If you have \n any questions\, please contact Martin Kaså [2].\n \n ------------------------------------------------------------------------------\n *Schedule*\n \n *Monday* 08.30-10.15 Travelling 10.15-10.30 Arrival and check-in 10.30-11.00 \n Coffee 11.00-12.30 Introduction and information 12.30-13.30 Lunch 13.30-16.00 \n Talks (9 + break) 16.00-17.00 Coffee and poster session 17.00-17.30 Jeremiah \n Foster: Open source technology strategy 17.30-18.00 Discussion 18.00-19.00 \n CLT board meets the advisors 19.30 Dinner *Tuesday* 09.00-09.50 Talks (3) \n 09.50-10.40 Recording\, transcribing and analysing spoken dialogue 10.40-11.10 \n Coffee 11.10-11.40 Talks (2) 11.50-12.30 Concluding discussion and \n information 12.30-13.30 Lunch 13.30-15.15 Travelling\n ------------------------------------------------------------------------------\n *Order and titles of talks*\n \n *Monday* *Speaker* *Title* 13.30-13.45 Aarne Ranta Grammars vs. statistics: \n Some pros and cons in machine translation 13.45-14.00 Torbjörn Lager \n Martin's Little Helper: Exploring Tropo's cloud-based dialogue technology \n 14.00-14.15 Markus Forsberg One for all\, all for one: Lexical resources \n starting a union 14.15-14.30 Robin Cooper Factoring semantic grammars using \n record types 14.30-14.45 Maria Toporowska Gronostaj SweFN 15.00-15.15 \n Elisabet Engdahl The Scandinavian syntactic judgment database 15.15-15.30 \n Lars Borin SALDO version 2.0 15.30-15.45 Peter Ljunglöf Assistive language \n technology 15.45-16.00 Krasimir Angelov Online parsing\, type checking and \n advanced editor for controlled languages in GF *Tuesday* *Speaker* *Title* \n 09.00-09.15 Leif-Jöran Olsson Free software and LT mashups 09.15-09.30 \n Fredrik Engström A formal framework for branching quantifiers 09.30-09.45 \n Elena Volodina On SALT\, its routines and challenges 11.10-11.25 Bengt \n Nordström English is not a formal language 11.25-11.40 Harald Hammarström \n Defining intelligibility on formal languages\n ------------------------------------------------------------------------------\n *Poster session*\n \n Ann-Marie Eklund Medical querying behaviour and an application Björn Haglund \n Cognitive models for text processing Dimitrios Kokkinakis Is data scrubbing \n capable of protecting the confidentiality and integrity of sensitive data? \n Grégoire Détrez A framework for multilingual applications on the Android \n platform Jonas Lindh A simple voice verification demo using open source tools \n Karin Cavallin Exploring lexical sets over time Olga Caprotti MOLTO - Tools \n for multilingual grammar based translation on the web Sofie Johansson \n Kokkinakis Lexical profiling as means for characterising subject-specific \n school book texts Staffan Larsson Demo of voice cursor technology in an \n in-vehicle dialogue system Taraka Rama Evaluation of linguistic similarity \n measures\n \n [1] http://gullmarsstrand.se/\n [2] mailto:martin.kasa@gu.se END:VEVENT BEGIN:VEVENT UID:calendar:362:field_eventdate:0:48 SUMMARY:CLT seminar: Mike Scott - Lexical Software Design Constraints\, or the perils \n of amateur gardening DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101111T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101111T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-11-11/clt-seminar-mike-scott-lexical-software-design-constraints-or-perils-amateur-gardening LOCATION:K333\, Lennart Torstenssonsgatan 6 DESCRIPTION:Abstract:\n \n This presentation will consider the principles underlying the construction of \n lexical analysis software with particular emphasis on MicroConcord (1993) and \n WordSmith Tools (1996 to date). Both endeavours have benefitted from numerous \n suggestions and advice but have been implemented mostly individually\, almost \n completely self-taught. Aspects to be explored include language-independence\, \n separation of task and display\, relevant senses\, the nature of the underlying \n units (from the individual character up through affixes\, words\, sentences\, \n paragraphs\, texts etc.)\, keyness\, statistical inference\, character \n representation\, colour and\, especially important\,  pattern generation and \n interpretation. These will be illustrated in part by reference to studies of \n 17th and 19th Century English literature.\n \n Web: Aston University [1] and lexically.net [2]\n \n \n [1] http://www1.aston.ac.uk/lss/staff/mike-scott/\n [2] http://www.lexically.net/ END:VEVENT BEGIN:VEVENT UID:calendar:387:field_eventdate:0:49 SUMMARY:Introduction to WordSmith Tools DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101111T133000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101111T163000 URL;VALUE=URI:http://www.clt.gu.se/event/2010-11-11/introduction-wordsmith-tools LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Mike Scott will give an introduction to WordSmith Tools \n (http://lexically.net/wordsmith/index.html [1]) 13.30-14.30 and then (after a \n coffee break) there will be a hands-on workshop follow-up 15.00-16.30. \n Participants should bring their own computers. Corpus data and WS-licenses \n may be provided.\n \n If you like to participate in this workshop\, please send an e-mail to Sofie \n Johansson Kokkinakis\n \n  \n \n \n [1] http://lexically.net/wordsmith/index.html END:VEVENT BEGIN:VEVENT UID:calendar:363:field_eventdate:0:50 SUMMARY:CLT seminar: Torbjörn Lager - Exploring Tropo - Part Two DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101118T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101118T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-11-18/clt-seminar-torbj%C3%B6rn-lager-exploring-tropo-part-two LOCATION:L307\, Lennart Torstenssonsgatan 8 DESCRIPTION:Abstract:\n \n At the Gullmarsstrand conference I presented a small application built on the \n Tropo [1] platform. During my seminar I will report on a couple of other \n Tropo experiments that I have performed\, in attempts to pave the way to more \n sophisticated applications.\n \n \n [1] https://www.tropo.com/home.jsp END:VEVENT BEGIN:VEVENT UID:calendar:364:field_eventdate:0:51 SUMMARY:CLT Seminar: Alexandra Weilenmann - Micro-sociological studies of the use of \n mobile IT: Examples from wild boar hunting and science centers DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101125T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101125T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-11-25/clt-seminar-alexandra-weilenmann LOCATION:L307\, Lennart Torstenssonsgatan 8 DESCRIPTION:Apart from giving a general overview of my research\, I will present some \n preliminary findings from two ongoing projects. The first deals with hunters' \n communication and organization using tecnology\, the second with the use of \n mobile phones among young science center visitors. The collected material \n consists of ethnographic fieldwork in these two environments\, including \n video- and audio recordings\, as well as interviews of participants. It would \n be interesting to get your comments on these studies.\n \n Alexandra Weilenmann is associate professor ('docent') in Applied Information \n Technology and works at the IT Faculty\, University of Gothenburg.  She is a \n member of University of Gothenburg's priorirty research areas Learning and \n Language Technology.  During 2010-2010 she is financed by Vinnova for a \n project titled Mobile services for interaction\, communication and learning. \n Partners in this project are Mobile Life VinnExcellence Center\, Stockholm \n University\, and the Linnaeus Excellence center LinCS at GU. END:VEVENT BEGIN:VEVENT UID:calendar:358:field_eventdate:0:52 SUMMARY:CLT Seminar: Adam Kilgarriff - Corpora: measurement\, comparison\, evaluation DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101202T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101202T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-12-02/clt-seminar-adam-kilgarriff LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Corpora now play a central role in many varieties of language research\, and \n most varieties of computational language research.  Yet our vocabulary for \n talking about them is terribly weak: corpora are usually described by giving \n their language\, size\, perhaps number of constituent texts\, and a free-text \n account of how they were gathered and what they have in them\, perhaps \n mentioning the genres or domains covered.  This is pre-scientific.  As we \n all come to realise that our findings\, in linguistics research\, need to be \n corpus-based\, so our inability to talk scientifically about the corpus they \n are based on becomes a central flaw in the whole research area.\n \n Providing a scientific vocabulary for describing corpora has been my core \n research agenda for more than a decade now\, and the talk will cover work on \n comparing corpora\, developing corpora for different domains and genres\, \n defining domains according to corpora of them\, and evaluating corpora.\n \n Adam Kilgarriff on the web: http://www.kilgarriff.co.uk/ [1]\n \n \n [1] http://www.kilgarriff.co.uk/ END:VEVENT BEGIN:VEVENT UID:calendar:365:field_eventdate:0:53 SUMMARY:CLT seminar: Ljunglöf\, Kronlid & Ericsson - Lekbot DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101209T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101209T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-12-09/clt-seminar-lekbot DESCRIPTION:*LekBot - A talking and playing robot for children with communicative \n disabilities*\n \n http://clt.gu.se/research/lekbot [1]\n Peter Ljunglöf\, Fredrik Kronlid & Stina Ericsson\n \n Lekbot is a collaboration between DART\, Talkamatic and the Dept. of \n philosophy\, linguistics and theory of science\, University of Gothenburg. It \n is funded by VINNOVA\, and runs from March 1\, 2010 to August 31\, 2011.\n \n The project uses current theory and technology in human communication\, \n computer communication and dialogue systems to develop a toy that is fun and \n interesting for young people with communicative disabilities. The toy helps \n the development of dialogical communication\, an area that is normally \n problematic for these children. Children with severe disabilities often have \n few opportunities to play independently and to interact on equal terms with \n children without disabilities\, and here Lekbot enables children with and \n without disabilities to interact and learn from each other.\n \n The Lekbot toy developed in the project is a radio-controlled robot that can \n be used by children with severe physical and/or cognitive disabilities\, such \n as cerebral palsy or autism. The robot is controlled by the child through \n touch-screen symbols. The symbols are translated into spoken language\, so \n that the touch screen "talks" to the robot and acts as the child's voice. The \n robot can\, in turn\, talk to the child using spoken language\, and the child \n can again answer using the touch screen.\n \n The robot is built using Lego Mindstorms NXT [2]\, and the dialogue system is \n developed using GoDiS. The project is supported by Acapela [3]\, whose speech \n synthesis is used for both the touch screen and the robot.\n \n We will present the current state of the project and give a demo.\n \n \n [1] http://clt.gu.se/research/lekbot\n [2] http://en.wikipedia.org/wiki/Lego_Mindstorms_NXT\n [3] http://www.acapela-group.com/ END:VEVENT BEGIN:VEVENT UID:calendar:366:field_eventdate:0:54 SUMMARY:CLT seminar: Ann-Marie Eklund\, Carola Ribeck\, Grégoire Détrez and Taraka \n Rama DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101216T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20101216T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2010-12-16/clt-seminar-ann-marie-eklund-carola-ribeck-gr%C3%A9goire-d%C3%A9trez-and-taraka-rama LOCATION:L307\, Lennart Torstenssonsgatan 8 DESCRIPTION: \n \n ------------------------------------------------------------------------------\n Presentations by our new PhD-students.\n \n  \n \n *Ann-Marie*:\n \n "The query\, not the answer\, is the answer"\n \n By analyzing queries to medical journal databases and patient records we may \n be able to obtain clues to questioners´ knowledge and interests. In this \n presentation we will briefly outline present and future research on analysis \n of questioner behaviour\, and how future research may help providing improved \n possibilities for knowledge sharing in the context of medical research and \n health care.\n \n *Carola*:\n \n I will give a short personal and professional presentation of myself. I will \n describe my general interests in the field of linguistics as well as a \n preliminary outline of my own research during the next 4-5 years. I will also \n present the graduate school to which I belong\, which deals with the language \n within the disciplines of the Swedish education.\n \n *Grégoire*:\n \n Presenting the release of PhraseDroid - the first GF/MOLTO application on \n Android.\n \n *Taraka*:\n \n "Automated classification of languages using lexical similarity"\n \n In this thesis\, I explore various string similarity measures for automatic \n language classification into families. The datasets for this study consist of \n Swadesh word lists collected by the ASJP group for 3730 languages of the \n world. We will discuss some related work. We will talk about some preliminary \n results and the directions for the future research.\n \n  \n \n  \n \n Also: Glögg [1]!\n \n \n [1] http://en.wikipedia.org/wiki/Mulled_wine#Gl.C3.B6gg END:VEVENT BEGIN:VEVENT UID:calendar:398:field_eventdate:0:55 SUMMARY:Seminar planning meeting DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110113T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110113T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-01-13/seminar-planning LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION: \n \n ------------------------------------------------------------------------------\n Planning event for the CLT seminar series - Spring semester 2011\n \n   END:VEVENT BEGIN:VEVENT UID:calendar:466:field_eventdate:0:56 SUMMARY:Grammar Technology Lab event DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110113T131500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110113T170000 URL;VALUE=URI:http://www.clt.gu.se/event/2011-01-13/grammar-event LOCATION:L307\, Lennart Torstenssonsgatan 8 DESCRIPTION:------------------------------------------------------------------------------\n A special grammar technology lab event is planned in connection with the \n visit by professor Laurette Pretorius [1] (University of South Africa).\n \n *Schedule*:\n \n 13:15 Laurette Pretorius: /Language technology for the languages of South \n Africa/\n 14:00 break\n 14:15 Thomas Hallgren & Aarne Ranta: /GF and MOLTO - an overview of recent \n and forthcoming developments/\n 14:30 Markos Kassa Gobena: /The Amharic resource grammar/\n 14:45 Krasimir Angelov: /GF in the wild - first steps/\n 15:00 Ramona Enache: /The MOLTO phrasebook/\n 15:15 Grégoire Détrez: /GF in Java and on Android phones/\n 15:30 coffee/tea\n 16:00 /What kind of interfaces to text corpora do we want and how do we get \n them?/\n           Elisabet Engdahl: /From the user's perspective/\n           Peter Ljunglöf & Olof Olsson: /KORP: Språkbankens new \n generation of corpus interfaces/\n \n ------------------------------------------------------------------------------\n For questions regarding this event\, please contact Aarne [2] or Elisabet [3].\n \n \n [1] http://osprey.unisa.ac.za/staff/staff.htm?qemail=PRETOL\n [2] mailto:aarne@chalmers.se\n [3] mailto:elisabet.engdahl@svenska.gu.se END:VEVENT BEGIN:VEVENT UID:calendar:392:field_eventdate:0:57 SUMMARY:CLT seminar: Angus Roberts - Clinical information extraction DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110120T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110120T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-01-20/clt-seminar-angus-roberts LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:------------------------------------------------------------------------------\n A talk on two related subjects:\n \n 1. /Information Extraction in the Clinical E-Science Framework/\n \n The Clinical E-Science Framework (CLEF) project built a system to extract \n clinically significant information from the textual component of medical \n records. Conventional clinical Information Extraction (IE) systems often use \n purpose built software\, and many involve some degree of knowledge engineering \n to encode clinical and linguistic knowledge. The CLEF IE system is built \n largely from off-the-shelf components\, and involves no additional knowledge \n engineering. Instead\, clinical knowledge is provided by human annotated \n examples\, which are used to learn statistical models of the text. This talk \n will describe the CLEF IE system\, and the building of a training data set and \n gold standard. The talk will give evaluations of system performance for both \n entity and for relation extraction\, with comparisons between training sets of \n different sizes and types. Finally\, the talk will illustrate the quantity of \n data that can be extracted\, by describing application of the system to a \n corpus of half a million clinical narratives and reports.\n \n 2. /Pattern grammar based clinical information extraction: an agile process \n for building practical systems/\n \n When developing any IE application\, both software and data must be \n considered. From a software engineering point of view\, the last decade has \n seen the emergence of re-useable NLP frameworks and tool-kits. The task of \n building an NLP application for processing medical records can thus move from \n de novo systems development to the adaptation of these tool-kits and \n frameworks. From a data point of view\, it must be considered that virtually \n all usable NLP techniques require significant volumes of manually prepared \n examples. One of the main stumbling blocks to developing medical NLP \n applications is the lack of such example data.\n \n This talk will describe the application of an increasingly popular software \n engineering technique - an agile methodology - to tackle both IE systems \n adaptation and annotation of example texts at the same time\, in a large \n hospital setting. Agile methodologies replace the linear \n requirements-design-implement approach to software engineering with early \n implementation and the iterative evolution of requirements. The approach \n taken maximises the involvement of clinician and medical researcher \n end-users\, at low cost to their time. We believe that this has a beneficial \n effect on requirements gathering\, and on final system quality. The talk will \n be illustrated with quantitative results from a working Proof-of-Concept \n application\, and will discuss ongoing work to develop further applications \n in the same institutional setting\, where we now have a successful and \n expanding production system.\n \n In contrast to CLEF\, the system implemented is based on hand-crafted pattern \n matching grammars. The talk will discuss this difference\, and make \n comparisons between the two approaches.\n \n Angus Roberts on the web: http://www.dcs.shef.ac.uk/~angus/ [1]\n \n  \n \n \n [1] http://www.dcs.shef.ac.uk/~angus/ END:VEVENT BEGIN:VEVENT UID:calendar:570:field_eventdate:0:58 SUMMARY:Guest talk: Laurette Pretorius - Building a GF resource grammar for Afrikaans DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110125T161500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110125T170000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-01-25/guest-talk-laurette-pretorius-building-gf-resource-grammar-afrikaans LOCATION:EDIT room\, EDIT building\, Chalmers DESCRIPTION:------------------------------------------------------------------------------\n The existence of the GF resource grammar for Dutch and the well-known \n linguistic similarity between Dutch and Afrikaans provide a basis for the \n accelerated development of a GF resource grammar for Afrikaans. The main \n purpose of this talk is to provide a status report on recent progress made in \n this regard\, with specific reference to the lexicon and certain word \n categories.\n \n --\n \n Professor Laurette Pretorius [1] (University of South Africa) is at a the \n moment visiting the Language Technology Group [2] at Computer Science and \n Engineering at GU/Chalmers.\n \n \n [1] http://osprey.unisa.ac.za/staff/staff.htm?qemail=PRETOL\n [2] http://www.cse.chalmers.se/research/group/Language-technology/ END:VEVENT BEGIN:VEVENT UID:calendar:397:field_eventdate:0:59 SUMMARY:CLT seminar: Anette Hulth - Web query-based infectious disease surveillance DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110210T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110210T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-02-10/clt-seminar-anette-hulth LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION: \n \n ------------------------------------------------------------------------------\n \n Anette Hulth (Swedish Institute for Communicable Disease Control)\n \n Traditional infectious diseases surveillance is based on reports from \n clinicians and laboratories. It thus requires that those who are ill seek \n care. As a complement to this traditional surveillance\, sources collected for \n purposes other than surveillance (for example sales of certain medicines or \n the number of telephone calls to a medical help line) can be exploited. At \n the Swedish Institute for Communicable Disease Control\, a system based on \n queries submitted to a Swedish medical web site – Vårdguiden.se – has \n been implemented and functions on a regular basis as an additional source of \n surveillance. In the work performed at the institute\, we have shown that web \n queries are an accurate\, cheap and labour extensive source\, which gives \n access to individuals who are not (yet) seeking care. In this presentation\, I \n will focus on the web query-based influenza and norovirus surveillance. I \n will also give a very short introduction to infectious disease surveillance \n in general. END:VEVENT BEGIN:VEVENT UID:calendar:1229:field_eventdate:0:60 SUMMARY:Seminar: Daniel Keim - Visual Document Analysis (particularly readability \n analysis and literature fingerprinting) DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110223T133000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110223T153000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-02-23/seminar-daniel-keim LOCATION:L307\, Lennart Torstenssonsgatan 8 DESCRIPTION: \n \n ------------------------------------------------------------------------------\n http://www.informatik.uni-konstanz.de/en/arbeitsgruppen/infovis/members/... \n [1]\n \n  \n \n \n [1] http://www.informatik.uni-konstanz.de/en/arbeitsgruppen/infovis/members/prof-dr-daniel-keim/ END:VEVENT BEGIN:VEVENT UID:calendar:399:field_eventdate:0:61 SUMMARY:CLT seminar: Daniel Keim - Visual Analytics: Challenges and Applications DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110224T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110224T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-02-24/clt-seminar-daniel-keim LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:------------------------------------------------------------------------------\n Never before in history have data been generated and collected at such high \n volumes as today. As the volumes of data available to business people\, \n scientists\, and the public increase\, their effective use becomes more \n challenging. Keeping up to date with the flood of data\, using standard tools \n for data analysis and exploration is fraught with difficulty. The field of \n visual analytics seeks to provide people with better and more effective ways \n to understand and analyze large datasets\, while enabling them to act upon \n their findings immediately. Visual analytics integrates the analytic \n capabilities of the computer and the abilities of the human analyst\, allowing \n novel discoveries and empowering individuals to take control of the \n analytical process. The talk presents the challenges of visual analytics and \n exemplifies them with several application examples\, illustrating the exiting \n potential of visual analysis techniques but also their limitations.\n \n http://www.informatik.uni-konstanz.de/en/arbeitsgruppen/infovis/members/prof-dr-daniel-keim/ \n [1]\n \n  \n \n \n [1] http://www.informatik.uni-konstanz.de/en/arbeitsgruppen/infovis/members/prof-dr-daniel-keim/ END:VEVENT BEGIN:VEVENT UID:calendar:553:field_eventdate:0:62 SUMMARY:CLT seminar: Taraka Rama - Explorations in using phoneme n-grams for \n automatic language classification DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110303T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110303T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-03-03/clt-seminar-taraka-rama LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:------------------------------------------------------------------------------\n In this talk\, we present some explorations in using phoneme n-grams for the \n task of automatic language classification using Swadesh word lists. In recent \n years\, historical linguistics has seen a huge amount of work in verifying \n hypotheses using computational methods. Some of the tasks which are addressed \n are estimation of language distance\, the relation between the diversity of \n phonological inventory and the time depth of language families and spatial \n spread. We observe that there is a good correlation between the phonological \n diversity and spatial spread and time depth of a language family. END:VEVENT BEGIN:VEVENT UID:calendar:396:field_eventdate:0:63 SUMMARY:CLT seminar: Anna Hjalmarsson - Towards Incremental Speech Generation in \n Dialogue Systems DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110310T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110310T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-03-10/clt-seminar-anna-hjalmarsson LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:------------------------------------------------------------------------------\n http://www.speech.kth.se/~annah/ [1]\n \n \n [1] http://www.speech.kth.se/~annah/ END:VEVENT BEGIN:VEVENT UID:calendar:1233:field_eventdate:0:64 SUMMARY:Monday seminar: Jeffrey Parrott - On case allomorphy and variation in Danish \n and (North) Germanic DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110314T151500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110314T170000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-03-14/monday-seminar-jeffrey-parrott-case-allomorphy-and-variation-danish-and-north-germanic LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:------------------------------------------------------------------------------\n \n Jeffrey Parrott (LANCHART\, Copenhagen)\n \n http://lanchart.hum.ku.dk/staff_scan/medarbejderdetaljer/?id=340729&f=1 [1]\n \n  \n \n \n [1] http://lanchart.hum.ku.dk/staff_scan/medarbejderdetaljer/?id=340729&\;f=1 END:VEVENT BEGIN:VEVENT UID:calendar:1234:field_eventdate:0:65 SUMMARY:Seminar: Jeffrey Parrott - (N'CLAV) Recent investigations DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110315T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110315T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-03-15/seminar-jeffrey-parrott-nclav-recent-investigations LOCATION:L307\, Lennart Torstenssonsgatan 8 DESCRIPTION:------------------------------------------------------------------------------\n \n Jeffrey Parrott (LANCHART\, Copenhagen)\n \n http://lanchart.hum.ku.dk/staff_scan/medarbejderdetaljer/?id=340729&f=1 [1]\n \n http://spraakbanken.gu.se/nclav [2]\n \n \n [1] http://lanchart.hum.ku.dk/staff_scan/medarbejderdetaljer/?id=340729&\;f=1\n [2] http://spraakbanken.gu.se/nclav END:VEVENT BEGIN:VEVENT UID:calendar:554:field_eventdate:0:66 SUMMARY:CLT seminar: Krasimir Angelov - Porting Penn Treebank to GF DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110317T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110317T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-03-17/clt-seminar-krasimir-angelov LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:------------------------------------------------------------------------------\n \n So far GF was used only for parsing small controlled languages but the \n improvements in the parsing performance in the last few years made it \n possible to dream about parsing open domain unrestricted text. In the \n resource libraries\, we already have wide coverage grammars for many languages \n but having a grammar is only part of the problem. Even if we improve more and \n more our grammars it will be always possible to find syntactic constructions \n which are not covered by the grammar. Another problem is that when we add \n more syntactic constructions in the grammar\, this usually makes it more \n ambiguous. We need a parser that is robust and is able to do statistical \n ranking when there are ambiguities in the grammar. I did some preliminary \n expreriments in robust parsing but the general conclusion is that we need a \n good treebank which we can use for statistical training. Since we don't want \n to build our own treebanks an attractive alternative is to try to convert \n some existing one.\n \n In this talk I will present the current state of the GF port of Penn \n Treebank. All parse trees in the treebank were converted to abstract syntax \n trees for the English Resource Grammar. When there are unknown syntactic \n constructions then we just leave placeholders in the abstract tree. Currently \n we have matched 69% of the constructions with the grammar. More is possible \n but takes time. END:VEVENT BEGIN:VEVENT UID:calendar:1232:field_eventdate:0:67 SUMMARY:Guest seminar: Simon Dobnik (Oxford University) - Grounding the meaning of \n spatial words in a mobile robot setting DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110318T141500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110318T151500 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-03-18/guest-seminar-simon-dobnik-oxford-university-grounding-meaning-spatial-words-mobile-robot-setting LOCATION:T116\, Gamla Hovrätten (FLoV) DESCRIPTION:Grounding the meaning of spatial words in a mobile robot setting\n \n Simon Dobnik\n \n Oxford University Computing Laboratory\n \n  \n \n We describe a system where the semantics of spatial referential\n expressions have been automatically learned by finding mappings between\n symbolic natural language descriptions of the environment and non-symbolic\n representations from the sensory data of a mobile robot used for\n localisation and map building (SLAM). Although the success of learning can\n be measured by examining classifier performance on held-out data\, this\n does not in itself guarantee that the descriptions or actions generated\n will be natural and informative for a human observer. To demonstrate that\n the robot has captured human-like spatial knowledge the classifiers were\n integrated with a system that drives a mobile robot and its performance in\n generating descriptions and answering questions was evaluated by human\n observers. END:VEVENT BEGIN:VEVENT UID:calendar:1235:field_eventdate:0:68 SUMMARY:Text Technology Lab: Three guest talks DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110321T133000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110321T155000 URL;VALUE=URI:http://www.clt.gu.se/event/2011-03-21/text-technology-lab-three-guest-talks LOCATION:K332\, Lennart Torstenssonsgatan 6 DESCRIPTION:------------------------------------------------------------------------------\n *CLT Text Technology Laboratory seminar with three guest talks*\n \n 13.30-14.10 Gerlof Bouma (University of Potsdam) \n \n 14.15-14.55 Simon Dobnik (Oxford University) \n \n 14.55-15.10 break\n \n 15.10-15.50 Richard Johansson (University of Trento)\n \n /Abstracts/:\n \n *Gerlof Bouma* (University of Potsdam): /Association measures assorted/\n \n In this talk I will discuss extensions of pointwise mutual information and \n (average) mutual information\, two classic association measures popular in the \n field of collocation extraction. The essence of both measures is the \n comparison of observed occurrence of a combination and its expected \n occurrence under the assumption that the parts in the combination are \n independently distributed. The first extension tries to reduce the effect of \n occurrence probability and to improve the interpretability of the association \n measures by normalization. In the second extension\, we relax the independence \n assumption in the calculation of expected occurrence\, in an attempt to filter \n out the effect of dependencies that are not directly relevant for collocation \n extraction. In each case we show the effect of the extension in an evaluation \n against an extraction gold standard.\n \n If time permits\, I will end by discussing some other uses of association \n measures in (computational) corpus linguistics\, in particular ongoing work on \n the effect of 'associatedness' on object fronting in Dutch.\n \n ------------------------------------------------------------------------------\n *Simon Dobnik* (Oxford University): /On learning the semantics of spatial \n words and generating extractive summaries of scientific papers/\n \n The talk describes two projects. In the first part of the talk I discuss how \n the semantics of spatial expressions can be grounded in the sensory data of a \n mobile robot. The mapping between the human conceptual domain and that used \n to represent space in physical sciences and mobile robotics is learned with \n machine learning. We discuss the performance classifiers to generate new \n contextually appropriate descriptions. In the second part of the talk I \n discuss my work with Maria Liakata (Aberystwyth University\, EBI: Cambridge) \n on generating and evaluating extractive summaries of chemistry scientific \n papers. The sentences chosen for the summary are tagged for core scientific \n concepts such as Background\, Hypothesis and Result which correspond to \n methods used in scientific investigation.\n \n ------------------------------------------------------------------------------\n *Richard Johansson* (University of Trento): /Semantic role extraction with \n applications/\n \n In this presentation\, we give an introduction to semantic role\n representations such as FrameNet\, VerbNet\, and PropBank\, and their\n potential benefits in practical NLP applications.  We describe how these\n representations can be automatically extracted from free text\, and examine\n how this task is interrelated with syntactic parsing. In addition\, we show\n how parallel corpora can mitigate the problem of limited data availability\n for development of semantic role extractors for small languages. END:VEVENT BEGIN:VEVENT UID:calendar:555:field_eventdate:0:69 SUMMARY:CLT seminar: Julia Pennlert & Sverker Lundin - Language technology as a \n research tool for literary studies DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110324T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110324T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-03-24/clt-seminar-julia-pennlert-sverker-lundin LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION: \n \n ------------------------------------------------------------------------------\n *Data and Methodology in the PhD-Project “Poetry Processes on the \n Internet”*\n \n Poetry communities are good examples of how Internet during the last decade \n has radically changed the condition for production and consumption of \n literary texts. On the Web\, everyone can publish\, read and discuss poetry. \n One of these new arenas is the Swedish community www.poeter.se [1]\, with \n about 26 000 registered members. This community is the main focus of Julia \n Pennlert’s PhD project “Poetry processes on the Internet”. The project \n deals with questions concerning the content of web-published poetry\, the \n comments on these poems and the valuation of these poems (i.e. processes of \n canonization) in this special web-based context.\n \n On poeter.se [2] thousands of users publish an enormous amount of poems\, \n comments\, discussions posts and value judgements. There are different ways of \n “using” this arena for anyone interested in poetry. Together they \n constitute a large research material with a complex structure. This kind of \n material also opens up for fascinating possibilities for using language \n technology as a tool in the task of answering new kinds of questions for the \n literary scholars\, as well as other web-based researches. The material and \n these questions also pose a challenge to language technology.\n \n In the seminar we talk about how we work together and present examples of how \n the material can be represented and explored in ways meaningful from the \n point of view of literary studies.\n \n Julia Pennlert [3] (Umeå university) and Sverker Lundin [4] (University of \n Gothenburg)\n \n  \n \n \n [1] http://www.poeter.se/\n [2] http://www.poeter.se/\n [3] http://www.kultmed.umu.se/om-institutionen/personal/visa-person/?uid=juapet04&\;guise=anst1\n [4] http://www.flov.gu.se/om/personal/sverker-lundin/ END:VEVENT BEGIN:VEVENT UID:calendar:557:field_eventdate:0:70 SUMMARY:CLT seminar: Staffan Larsson - Do Dialogues have Content? DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110407T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110407T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-04-07/clt-seminar-staffan-larsson LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:------------------------------------------------------------------------------\n The notion of "the content of a dialogue" is shown to be problematic in light \n of the phenomena of semantic coordination in dialogue\, and the associated \n notion of semantic plasticity - the ability of meanings to change as a result \n of language use. Specifically\, it appears that any notion of content in \n dialogue based on classical model-theoretical semantics will be insufficient \n for capturing semantic plasticity. An alternative formal semantics\, type \n theory with records (TTR) is briefly introduced and is shown to be better \n equipped to deal with semantic coordination and plasticity. However\, it is \n also argued that any account of content in dialogue which takes semantic \n coordination seriously will also need to consider the problems it raises for \n some concepts central to traditional notions of meaning\, namely inference and \n truth.\n \n   END:VEVENT BEGIN:VEVENT UID:calendar:558:field_eventdate:0:71 SUMMARY:CLT seminar: Moved to May 5th DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110414T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110414T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-04-14/clt-seminar-moved LOCATION:L308\, Lennart Torstenssonsgatan 8 END:VEVENT BEGIN:VEVENT UID:calendar:559:field_eventdate:0:72 SUMMARY:Workshop: Tense & Aspect DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110428T100000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110428T160000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-04-28/ws-tense-aspect LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:*Workshop on Tense & Aspect*\n \n Thursday 28 April\, 10:00 – 16:00 (possibly longer) in L308\n \n Preliminary program:\n \n 10:00 Introduction and welcome (Elisabet Engdahl)\n \n 10:15 Nancy Chang: Reframing aspectual composition: constructions\, event \n structure and simulation \n \n 11:15 Robin Cooper: Tense\, Frames and Record Types\n \n 12:15 Lunch\n \n 13:15 Verner Egerland (LU): ONCE. On the syntactic expression of aspect and \n logophoricity\n \n 14:15 Kristian Blensenius: Work on progress\n \n 15:00 coffee/tea\n \n 15:15 Ida Larsson: On possession\, particles and perfectives. Intended events \n with mixed results.\n \n wine reception\n \n   END:VEVENT BEGIN:VEVENT UID:calendar:560:field_eventdate:0:73 SUMMARY:CLT seminar: Preparing for NODALIDA DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110505T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110505T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-05-05/clt-seminar-preparing-NODALIDA LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:------------------------------------------------------------------------------\n Preparing for NODALIDA:\n \n *Markus Forsberg* - /Green resources in plain sight: opening up the SweFN++ \n project/\n \n /SweFN++/ is a project conducted at Språkbanken with two main objectives: \n the creation of a Swedish framenet covering at least 50\,000 lexical units \n built on the same principles as the English Berkeley FrameNet\; an integration \n of a number of existing free lexical resources\, constructed by harmonizing\, \n standardizing and merging these resources\, and thereby reusing the valuable \n grammatical and semantic information painstakingly collected in these \n resources.\n \n An important theme of the project is *openness*\, which manifests itself in \n many different ways: /openness from day one\; open content\; downloadable \n development versions updated daily\; the use of open standards and open \n licenses\; the use and production of open source tools\; and open web service \n APIs for both resources and tools./\n \n This is a philosophical stance – we believe that research should be carried \n out in the open to enable scrutinization and increased collaboration. It is\, \n from our point of view\, more valuable that anyone is allowed to download and \n inspect unfinished work today\, and\, at the same time\, run the risk that it is \n confused with something more mature\, rather than taking the safer\, but less \n productive\, road of publishing the "finished" product at the end of the \n project.\n \n The project is open for inspection at: http://spraakbanken.gu.se/eng/swefn/ \n [1]\n \n *Peter Ljunglöf* - /Editing Syntax Trees on the Surface/\n \n We describe a system for interactive modification of syntax trees by \n intuitive editing operations on the surface string. The system has a \n graphical interface\, where the user can move\, replace\, add\, and in other ways \n modify\, words or phrases. During editing\, the sentence is kept grammatical\, \n by automatically rearranging words and changing inflection\, if necessary. \n This is accomplished by combining constraints on syntax trees with a distance \n measure between trees.\n \n *Dimitrios Kokkinakis* - /Evaluating the Coverage of three Controlled Health \n Vocabularies with Focus on Findings\, Signs & Symptoms/\n \n The medical domain is blessed with a magnitude of terminological resources of \n various characteristics\, sizes\, structure\, depth and breadth of descriptive \n power\, granularity etc. In this domain a particularly interesting and \n difficult entity type are signs\, symptoms and findings which to a large \n extend are expressed in a periphrastic manner\, sometimes by the use of \n figurative or metaphorical language\, or contextualized using a wealth of \n vague variant expressions. We hypothesize therefore that no major official \n terminology source alone can accommodate for the variation and complexity \n present in real text data\, such as electronic medical records\, notes or \n health related documents. In this paper we evaluate the content of the three \n largest medical control vocabularies available for Swedish on extracted \n reference symptom lists and initiate a discussion on how we should proceed in \n order to accommodate for increased coverage on similar genres.\n \n  \n \n \n [1] http://spraakbanken.gu.se/eng/swefn/ END:VEVENT BEGIN:VEVENT UID:calendar:561:field_eventdate:0:74 SUMMARY:NODALIDA DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110511T000000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110513T000000 URL;VALUE=URI:http://www.clt.gu.se/event/2011-05-11/nodalida LOCATION:Riga DESCRIPTION:http://www.lumii.lv/nodalida2011/ [1]\n \n \n [1] http://www.lumii.lv/nodalida2011/ END:VEVENT BEGIN:VEVENT UID:calendar:562:field_eventdate:0:75 SUMMARY:CLT seminar: Torbjörn Lager - Ideas for a CLT Cloud DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110519T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110519T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-05-19/clt-seminar-torbjorn-lager LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:There are lots of reasons why setting up a CLT Cloud in the form of web-API:s \n to our language resources and processors is a good idea. In my talk\, I will \n begin by reviewing some such reasons. The question is\, how do we design and \n implement a cloud once we know we want one? A web-API request usually results \n in a chunk of XML or JSON being returned to the client\, and this can of \n course easily be hacked. But XML and JSON is just syntax\, and it seems we \n want to take a more pricipled and perhaps semantical approach to the API \n design problem. Markus Forsberg and myself have a project where we intend to \n explore such an approach. In most of my talk – or hey\, let's call it a \n working seminar! – I intend to take a first stab at this\, based on the \n concept of a graph. END:VEVENT BEGIN:VEVENT UID:calendar:1245:field_eventdate:0:76 SUMMARY:CLT seminar: Larry Horn (Yale University) - Uncrypting a cryptotype: Lexical \n semantics\, lexical pragmatics\, and the un-verb DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110519T131500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110519T150000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-05-19/clt-seminar-larry-horn LOCATION:L308 DESCRIPTION:------------------------------------------------------------------------------\n Since Whorf (1936)\, many linguists have tried their hand at corralling the \n restrictions on the formation of "reversative" /un/-verbs\; cf. e.g. Marchand \n (1969)\, Dowty (1979)\, Horn (1988)\, Clark et al. (1995). Why can you /unwrap/ \n a sandwich but not /unrecognize/ its contents or /unremember/ to toss it in \n the trash? Why can a snake /uncoil/ while a painting can't /unhang/?  If \n /unfreeze/ is the opposite of /freeze/\, why is /unthaw/ a synonym of /thaw/?\n \n The standard approach to the constraints on un-verb formation invokes Whorf's \n CRYPTOTYPE -- a covert category encompassing transitive verbs of covering and \n enclosing that rules out a wide range of possible bases and outputs of the \n relevant rule. Pullum (1999)\, for example\, reckons that there are "about a \n dozen verbs" that allow /un/-prefixation\, citing /undo/ (/a good deed/) and \n /unknow/ as examples of formations we know "intuitively" are impossible. \n Clark et al. (1995) exclude /unbury/ and /unbend/\, while Kemmerer & Wright \n (2002) rule out /unboil/ and /undecorate/. Yet many of the verbs depicted in \n the literature as impossible\, non-occurring\, or -- as in Whorf's label for \n /unsay/ and /unmake/ -- "semi-archaic" are readily attested\, even when the \n actions they denote may be physically irreversible.\n \n While the Pullum/Whorf view may be extended to predict the pleonastic \n interpretation of source-oriented reversatives in Swedish\, French\, and \n English (/unloosen/\, /unthaw/)\, it incorrectly limits the productivity of \n /un/-verb formation by conflating the SEMANTIC (aspectual) restrictions with \n the PRAGMATIC conditions on the way the world (normally) works\; verbs like \n /unsay/\, /unknow/\, /unboil/\, and /unhappen/ are motivated precisely by the \n need to describe those (typically counterfactual) situations in which the \n tape of reality is reversed.\n \n The pragmatic nature of the restrictions on /un/-verb formation is supported \n by a survey of contexts that favor the emergence of innovative /un/-verbs: \n advances in science and technology (as in the /unerase/ and /undelete/ \n commands\, the /unfriend/ or /unlike/ verbs of social networking\, or the \n /unfuck/ program to reverse software protection)\, science fiction (as in \n time-travel scenarios)\, advertising copy (as in KFC's current /unthink/ \n campaign)\, and the imagination of poets from Shakespeare (whose \n "/un-king’d/" Richard II is the unchallenged monarch of this realm) to pop \n songsters ("/How can I unlove you?/"\, "/Un-break my heart/")\, whether the \n implausibility of a given reversal is conceded\, mourned\, or overridden.\n \n Larry Horn [1] (Yale University)\n \n \n [1] http://www.ling.yale.edu/faculty/horn.html END:VEVENT BEGIN:VEVENT UID:calendar:1249:field_eventdate:0:77 SUMMARY:Seminar: Dale Miller - Programming in higher-order logic (Part I) DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110523T100000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110523T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-05-23/seminar-dale-miller-1 LOCATION:EDIT room (3320)\, Chalmers DESCRIPTION:Dale Miller (INRIA Saclay and LIX/Ecole Polytechnique\, Palaiseau\, France)\n will give a two-part seminar about Lambda Prolog (the language which \n influenced the development of the logical aspects in the latest version of \n GF).\n \n Abstract:\n \n Lambda Prolog is a logic programming language based on an intuitionistic \n fragment of Church's Simple Theory of Types.  Such a strong logical \n foundations provides lambda Prolog with logically supported notions of \n modular programming\, abstract datatypes\, higher-order programming\, and the \n lambda-tree syntax approach to the treatment of bound variables.\n \n This seminar will provide the sequent calculus foundations of logic \n programming\, an overview of higher-order unification\, and several examples of \n lambda Prolog programming tasks\, taken from topics such as theorem proving\, \n operational semantics\, program transformations\, and the pi-calculus.\n \n Web:\n \n Dale Miller: http://www.lix.polytechnique.fr/Labo/Dale.Miller/ [1]\n \n lambda prolog: http://www.lix.polytechnique.fr/~dale/lProlog/ [2]\n \n \n [1] http://www.lix.polytechnique.fr/Labo/Dale.Miller/\n [2] http://www.lix.polytechnique.fr/~dale/lProlog/ END:VEVENT BEGIN:VEVENT UID:calendar:1250:field_eventdate:0:78 SUMMARY:Seminar: Dale Miller - Programming in higher-order logic (Part II) DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110524T100000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110524T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-05-24/seminar-dale-miller-2 LOCATION:EDIT room (3320)\, Chalmers DESCRIPTION:Dale Miller (INRIA Saclay and LIX/Ecole Polytechnique\, Palaiseau\, France)\n will give a two-part seminar about Lambda Prolog (the language which \n influenced the development of the logical aspects in the latest version of \n GF).\n \n Abstract:\n \n Lambda Prolog is a logic programming language based on an intuitionistic \n fragment of Church's Simple Theory of Types.  Such a strong logical \n foundations provides lambda Prolog with logically supported notions of \n modular programming\, abstract datatypes\, higher-order programming\, and the \n lambda-tree syntax approach to the treatment of bound variables.\n \n This seminar will provide the sequent calculus foundations of logic \n programming\, an overview of higher-order unification\, and several examples of \n lambda Prolog programming tasks\, taken from topics such as theorem proving\, \n operational semantics\, program transformations\, and the pi-calculus.\n \n Web:\n \n Dale Miller: http://www.lix.polytechnique.fr/Labo/Dale.Miller/ [1]\n \n lambda prolog: http://www.lix.polytechnique.fr/~dale/lProlog/ [2]\n \n \n [1] http://www.lix.polytechnique.fr/Labo/Dale.Miller/\n [2] http://www.lix.polytechnique.fr/~dale/lProlog/ END:VEVENT BEGIN:VEVENT UID:calendar:563:field_eventdate:0:79 SUMMARY:CLT seminar: CANCELLED DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110526T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110526T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-05-26/clt-seminar-villing-lundholm-fors LOCATION:L308\, Lennart Torstenssonsgatan 8 END:VEVENT BEGIN:VEVENT UID:calendar:1267:field_eventdate:0:80 SUMMARY:Job offer: Research engineer/systems developer at FLoV DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110526T120000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110526T130000 URL;VALUE=URI:http://www.clt.gu.se/event/2011-05-26/job-offer-research-engineersystems-developer-flov DESCRIPTION:http://ledig-anstallning.adm.gu.se/detail.php?lt_id=6852&lang=eng&eng=true \n [1]\n \n \n [1] http://ledig-anstallning.adm.gu.se/detail.php?lt_id=6852&\;lang=eng&\;eng=true END:VEVENT BEGIN:VEVENT UID:calendar:564:field_eventdate:0:81 SUMMARY:Human reasoning seminar: Nancy Chang - Neurally plausible relational \n representations DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110609T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110609T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-06-09/clt-seminar-nancy-chang LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Nancy Chang [1] will give a guest talk at the Human reasoning seminar series \n [2]\n \n Abstract:\n \n In this talk I present an overview of how certain relational concepts can be \n captured using neurally plausible representations. The approach exploits \n structured connectionist models that are\, like traditional neural networks\, \n inspired by the general properties of human neural computation. But these \n models also recognize the highly structured nature of neural circuitry\, and \n the possibility of using local clusters of neurons to represent traditionally \n symbolic relational concepts. I will describe how several such concepts – \n including feature-value association\, variable binding and dynamic action \n control – can be modeled by neurally plausible systems.\n \n  \n \n \n [1] http://www.icsi.berkeley.edu/~nchang/\n [2] https://sites.google.com/site/humanreasoningseminarseries/ END:VEVENT BEGIN:VEVENT UID:calendar:1272:field_eventdate:0:82 SUMMARY:Seminar: Malin Ahlberg - A Wide-Coverage Grammar and Parser for Swedish - \n first Results and Perspectives DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110609T130000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110609T140000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-06-09/seminar-malin-ahlberg LOCATION:Room 6128 (ED house 6th floor)\, Chalmers DESCRIPTION:During a 7.5hp project I've been working on a pilot project\, evaluating and \n extending the existing GF grammar for Swedish. By using a large lexicon \n ported from SALDO and an extra GF module we hope to eventually implement a \n substantial parser for Swedish. This should be able to parse all of \n Talbanken\, a treebank consisting of over 6000 sentences.\n \n In the course of the project\, grammatical constructs have been added to the \n GF grammar\, and some improvements have been made to the lexicon. Also\, an \n interactive tool for lexicon acquisition of Swedish verbs has been \n implemented\, which builds GF lexicons by guessing the inflection paradigm for \n a given word. I will continue the work in my Master Project\, and at the \n presentation I will discuss my results and the directions for future work \n that have been identified. END:VEVENT BEGIN:VEVENT UID:calendar:1269:field_eventdate:0:83 SUMMARY:CLT seminar: Marina Santini - Computational Models for Automatic Web Genre \n Identification DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110616T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110616T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-06-16/clt-seminar-marina-santini LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Broadly speaking\, "genre" is a classification concept. A genre is a recurring \n and recognized pattern of communication that has a specific name. The web \n hosts many recognised genres\, such as FAQs\, press releases\, product \n descriptions\, instructions\, guides\, e-magazines\, blogs\, professional \n profiles\, how-tos\, web ads and reviews. Each of these genres serves a number \n of communicative and social purposes and carries additional contextual \n information that helps the reader interpret the content. Can web genres be \n identified and detected automatically? Which computational models have been \n tried out so far in automatic genre identification research? How well do they \n perform? In this talk\, I will present and discuss the latest findings in \n automatic genre identification and suggest viable future directions.\n \n Web:\n \n Academic: http://sites.google.com/site/marinasantiniacademicsite/ [1]\n \n Artificial Solutions: http://www.artificial-solutions.com/ [2]\n \n \n [1] http://sites.google.com/site/marinasantiniacademicsite/\n [2] http://www.artificial-solutions.com/ END:VEVENT BEGIN:VEVENT UID:calendar:1292:field_eventdate:0:84 SUMMARY:GF Summer School 2011 DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110815T170000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110826T130000 URL;VALUE=URI:http://www.clt.gu.se/event/2011-08-15/gf-summer-school-2011 LOCATION:Universitat Politècnica de Catalunya\, Barcelona DESCRIPTION:http://school.grammaticalframework.org/ [1]\n \n \n [1] http://school.grammaticalframework.org/ END:VEVENT BEGIN:VEVENT UID:calendar:1285:field_eventdate:0:85 SUMMARY:CLT seminar planning DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110825T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110825T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-08-25/clt-seminar-planning LOCATION:L308\, Lennart Torstenssonsgatan 8 END:VEVENT BEGIN:VEVENT UID:calendar:1294:field_eventdate:0:86 SUMMARY:CLT seminar: Hans Leiß - Resolving Plural Ambiguities by Type Reconstruction DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110908T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110908T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-09-08/clt-seminar-hans-lei%C3%9F LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Hans Leiß (LMU\, München)\n \n Abstract:\n A type reconstruction algorithm for a fragment of natural language is \n presented\, based on Hindley's algorithm for simple types plus structural \n subtyping and overloading of constants. We extend Montague's PTQ-fragment of \n English by plural noun phrases (which may have several types) and overloaded \n verbs to allow for distributed and non-distributed readings of noun phrases \n and verb arguments. We demonstrate how type reconstruction can select \n suitable meanings of subject noun phrases depending on the meaning of verb \n phrases\, thereby handling some violations of Frege's compositionality \n principle.\n \n http://www.cis.uni-muenchen.de/~leiss/ [1]\n \n \n [1] http://www.cis.uni-muenchen.de/~leiss/ END:VEVENT BEGIN:VEVENT UID:calendar:1295:field_eventdate:0:87 SUMMARY:CLT seminar: Preparing for SemDial DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110915T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110915T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-09-15/clt-seminar-preparing-semdial LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:A dry run of three talks to be presented at the SemDial workshop in Los \n Angeles.\n \n http://projects.ict.usc.edu/nld/semdial2011/ [1]\n \n 1. Kristina Lundholm Fors & Jessica Villing: "Reducing cognitive load in \n in-vehicle dialogue system interaction"\n \n Abstract:\n \n In-vehicle dialogue systems need to be able to adapt to the cognitive load of \n the user\, and\, when possible\, reduce cognitive load. To accomplish this\, we \n need to know how humans act while driving and talking to a passenger\, and \n find out if there are dialogue strategies that can be used to minimize \n cognitive load. In this study\, we have analyzed human-human in-vehicle \n dialogues\, focusing on pauses and adjacency pairs. Our results show that when \n the driver is experiencing high cognitive load\, the passenger’s median \n pause times increase. We also found that\, when switching to another domain \n and/or topic\, both driver and passenger try to avoid interrupting an \n adjacency pair. This suggests that a dialogue system could help lower the \n user’s cognitive load by in- creasing pause lengths within turns\, and plan \n system utterances in order to avoid switching task within an adjacency pair.\n \n 2. Ellen Breitholz & Robin Cooper: "Enthymemes as Rhetorical Resources"\n \n Abstract:\n \n In this paper we propose that Aristotelian enthymemes play a role in the \n resources available to dialogue participants.  We take as our point of \n departure the idea that every individual has a set of linguistic resources \n that are formed and reformed through interaction with other individuals and \n context.\n \n We regard enthymemes as dependent record types\, functions which map contexts \n modelled as records\, corresponding to the premises of the enthymeme\, to a \n record type which models a proposition corresponding to the conclusion of the \n enthymeme. The advantage of using record types is that they give us semantic \n objects corresponding to enthymemes (as opposed to textual objects such as \n inference rules) and a straightforward way of generalizing\, restricting and \n combining enthymemes thereby giving a theory of how agents can expand and \n reform their rhetorical resources on the basis of experience. In this paper \n we propose that Aristotelian enthymemes play a role in the resources \n available to dialogue participants.  We take as our point of departure the \n idea that every individual has a set of linguistic resources that are formed \n and reformed through interaction with other individuals and context.\n  \n We regard enthymemes as dependent record types\, functions which map contexts \n modelled as records\, corresponding to the premises of the enthymeme\, to a \n record type which models a proposition corresponding to the conclusion of the \n enthymeme. The advantage of using record types is that they give us semantic \n objects corresponding to enthymemes (as opposed to textual objects such as \n inference rules) and a straightforward way of generalizing\, restricting and \n combining enthymemes thereby giving a theory of how agents can expand and \n reform their rhetorical resources on the basis of experience.\n \n 3. Robin Cooper and Jonathan Ginzburg: "Negation in Dialogue"\n \n Abstract:\n \n We consider the nature of negation in dialogue as revealed by semantic \n phenomena such as negative dialogue particles\, psycholinguistic \n experimentation\, and dialogue corpora. We examine alternative accounts of \n negation that can be used in TTR (Type Theory with Records)\, and conclude \n that an alternatives-based account which relates to the psychological notion \n of negation in simulation semantics is most appropriate. We show how this \n account relates to questions under discussion\, dialogical relevance\, and \n metalinguistic negation.\n \n \n [1] http://projects.ict.usc.edu/nld/semdial2011/ END:VEVENT BEGIN:VEVENT UID:calendar:1296:field_eventdate:0:88 SUMMARY:CLT seminar: Håkan Burden - Natural language generation from executable \n software models DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110922T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110922T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-09-22/clt-seminar-h%C3%A5kan-burden LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Abstract:\n \n Software models is one way of bridging the gap between the code and \n intentions behind a system. Over the years there has been a number of \n different modelling languages and frameworks for using them. One such \n language is Executable and Translatable UML (xtUML) which is one way of \n enabling a framework called Model-Driven Architecture (MDA). A core concept \n within MDA and xtUML is model transformations. This seminar will show how \n model transformations can be used for Natural Language Generation as a means \n for model validation. END:VEVENT BEGIN:VEVENT UID:calendar:1291:field_eventdate:0:89 SUMMARY:(Extended) Deadline TLT 2012 DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110926T130000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110926T235900 URL;VALUE=URI:http://www.clt.gu.se/event/2011-09-26/extended-deadline-tlt-2012 LOCATION:GERMANY DESCRIPTION:Deadline *TLT* [1]\n \n \n [1] http://tlt10.cl.uni-heidelberg.de/ END:VEVENT BEGIN:VEVENT UID:calendar:1297:field_eventdate:0:90 SUMMARY:CLT seminar: Staffan Larsson - The TTR perceptron DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110929T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20110929T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-09-29/clt-seminar-staffan-larsson LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:In this presentation\, a dynamic semantic approach to subsymbolic perceptual \n aspects of meaning is presented. We show how a simple classifier of spatial \n information based on the Perceptron can be cast in TTR (Type Theory with \n Records). Furthermore\, we show how subsymbolic aspects of meaning can be \n updated as a result of observing language use in interaction\, thereby \n enabling fine-grained semantic plasticity and semantic coordination. END:VEVENT BEGIN:VEVENT UID:calendar:1271:field_eventdate:0:91 SUMMARY:CLT annual workshop DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111003T090000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111005T150000 URL;VALUE=URI:http://www.clt.gu.se/event/2011-10-03/clt-annual-workshop LOCATION:Gullmarsstrand DESCRIPTION:The second annual workshop for CLT members will take place at Gullmarsstrand \n [1] 3-5 October. We will leave by bus from Humanisten (*update*\, see below) \n at 09.00 on Monday morning\, and expect to be back in Gothenburg by 3 pm on \n Wednesday. Below\, you can find a preliminary schedule and titles for the \n presentations. If you have any questions\, please contact Martin Kaså [2].\n \n *Update: *Behind the faculty of arts main building is \n Carlandersplatsen/Lundgrensgatan\, which is where we will get on the bus. See \n map: http://g.co/maps/ep9ve [3]\n \n ------------------------------------------------------------------------------\n *Schedule*\n \n *Monday* 09.00-10.45 Travelling 11.00-11.40 Introduction 11.40-12.00 \n Nicoletta Calzolari: The Language Library 12.00-12.30 Talk (1) 12.30-13.30 \n Lunch 13.30-15.00 Methods in verbal and visual data storing: transcriptions\, \n files and corpora 15.00-16.00 Talks (2) 16.00-16.30 Coffee 16.30-18.00 Talks \n (3) 19.30 Dinner *Tuesday* 09.00-10.00 Talks (2) 10.00-10.30 Coffee \n 10.30-12.30 Poster session 12.30-13.30 Lunch 13.30-15.00 Large scale Swedish \n grammar and parser 15.00-15.30 Coffee 15.30-15.50 Presentation by Rickard \n Domeij 15.50-17.00 Discussion: strategy for development of CLT 17.00-19.00 \n CLT board meets the advisors 19.30 Dinner *Wednesday* 09.00-09.30 \n Presentation by Johanna Moore 09.30-11.00 Talks (3) 11.00-11.30 Concluding \n discussion 11.30 Lunch 12.30-14.15 Travelling\n ------------------------------------------------------------------------------\n *Order and titles of talks*\n \n *Monday* *Speaker* *Title* 12.00-12.30 Torbjörn Lager & Markus Forsberg CLT \n Cloud 15.00-15.30 Daniela Oelke Visually Exploring Your Data 15.30-16.00 \n Ellen Breitholtz & Robin Cooper Enthymemes as Rhetorical Resources \n 16.30-17.00 Chiranjib Bhattacharyya Discovering facets and sentiments from \n product review corpora 17.00-17.30 Ylva Hård af Segerstad & Alexandra \n Weilenmann Mobile Communication Lab 17.30-18.00 Taraka Rama Phonological \n diversity\, word length\, and population sizes across languages *Tuesday* \n *Speaker* *Title* 09.00-09.30 Gerlof Bouma Object fronting and lexical \n cohesion in Spoken Dutch 09.30-10.00 Peter Ljunglöf & Magdalena Siverbo \n FraCaSBank\, a multilingual GF treebank for the FraCaS test suite *Wednesday* \n *Speaker* *Title* 09.30-10.00 Shafqat Virk Resource Grammars of South Asian \n Languages (Punjab\, Persian) 10.00-10.30 Jonas Lindh The comparison between \n human and machine judgments of voice similarity 10.30-11.00 Robin Cooper \n Counting with frame types\n ------------------------------------------------------------------------------\n *Poster and Demo session*\n \n Ann-Marie Eklund A Swedish Health Web Interaction Log Corpus Carola Ribeck \n Identifying lexical bundles in secondary school textbooks Dana Dannélls A \n small resource grammar for Hebrew (DEMO) Dimitrios Kokkinakis Semantic \n Parsing for Relation Extraction Karin Friberg Heppin & Maria Toporowska \n Gronostaj Adapting FrameNet to Swedish - Problems and solutions Katarina \n Mühlenbock Using lexical and corpus resources for augmenting the AAC-lexicon \n Kristina Lundholm Fors End-of-utterance detection Malin Ahlberg A wide \n coverage grammar and parser for Swedish Martha Dís Brandt & Leif-Jöran \n Olsson META-NORD Ramona Enache MOLTO - 1st year Richard Johansson Syntactic \n and Semantic Parsing for Opinion Expression Detection Sandra Derbring The \n SubTTS project - using speech synthesis to make foreign movies available to \n people with language difficulties Simon Dobnik From situated descriptions of \n spatial scenes to situated dialogue Sofie Johansson Kokkinakis Quantitative \n methods applied to first and second language student writing Staffan Larsson \n SIMSI - Safe In-vehicle Multimodal Spoken Interaction Thomas Hallgren GF \n online editor for simple multilingual grammars (DEMO) Torbjörn Lager & \n Markus Forsberg CLT Cloud (DEMO)\n  \n \n \n [1] http://gullmarsstrand.se/\n [2] mailto:martin.kasa@gu.se?subject=Gullmarsstrand%203-5%20Oct\n [3] http://g.co/maps/ep9ve END:VEVENT BEGIN:VEVENT UID:calendar:1298:field_eventdate:0:92 SUMMARY:CLT seminar: Peter Ljunglöf - How to edit syntax trees on the surface DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111013T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111013T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-10-13/clt-seminar-peter-ljungl%C3%B6f LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Abstract:\n I am developing a pedagogical tool for children with communication \n disabilities. The tool will consist of a computer system which gives support \n for language learning and language training. The system is graphical\, where \n the child can point to words or phrases to get explanations in the form of \n symbols\, written text\, or spoken words. The child can also reformulate the \n sentences\, by moving words to new places or change the inflection of words. \n The system will then automatically reformulate the sentence\, so that it will \n always be grammatically correct. The tool is intended to stimulate - in a \n playful way - the child's will to explore the possibilities of language.\n \n In this talk I will report on the current status of the tool\, and the \n underlying ideas. If you are lucky\, I might even show some \n pseudo-mathematical theories about tree similarity in type-theoretical \n grammars...\n \n If you are even luckier\, I might give you more awesome animations in the \n style of my legendary CLT-workshop-discussion-introduction! END:VEVENT BEGIN:VEVENT UID:calendar:1289:field_eventdate:0:93 SUMMARY:Deadline LREC 2012 DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111015T235900 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111015T235900 URL;VALUE=URI:http://www.clt.gu.se/event/2011-10-15/deadline-lrec-2012 DESCRIPTION:Deadline *LREC* [1]\n \n \n [1] http://www.lrec-conf.org/lrec2012/ END:VEVENT BEGIN:VEVENT UID:calendar:1293:field_eventdate:0:94 SUMMARY:Korp workshop DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111017T131500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111017T170000 URL;VALUE=URI:http://www.clt.gu.se/event/2011-10-17/korp-workshop LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Official release of Korp - Språkbanken's new corpus search interface.\n \n More info on workshop webpage: http://spraakbanken.gu.se/eng/korp-release [1]\n \n http://spraakbanken.gu.se/korp/ [2]\n \n \n [1] http://spraakbanken.gu.se/eng/korp-release\n [2] http://spraakbanken.gu.se/korp/ END:VEVENT BEGIN:VEVENT UID:calendar:1299:field_eventdate:0:95 SUMMARY:CLT seminar: Yvonne Adesam - Multilingual Trees - Investigating High-quality \n Cross-lingual Parallel Corpus Processing DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111020T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111020T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-10-20/clt-seminar-yvonne-adesam LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:The presentation is a summary of my thesis\, to be finished early in 2012. It \n explores parallel treebanks\, texts and their translation which have been \n syntactically annotated and where corresponding parts are linked through \n alignment. Building parallel treebanks is a time-consuming and \n labour-intensive task. Additionally\, for the treebank to be really useful\, it \n needs to be large\, containing millions of words. However\, tools and/or \n annotated data are generally only available for a number of the largest \n languages in the world\, and automatic annotation (as well as human \n annotation) induces errors\, and we still need to check and manually correct \n the annotation\, for the data to be useful.\n \n The main question explored in this thesis is how to create a good quality \n parallel treebank. This is done in two separate ways. First\, we describe how \n the Smultron parallel treebank was created. Second\, we explore augmentation \n of automatically created annotation\, in particular methods of projecting \n knowledge from multiple languages\, known as multilingual annotation \n projection. END:VEVENT BEGIN:VEVENT UID:calendar:1352:field_eventdate:0:96 SUMMARY:Workshop on comparing approaches to measuring linguistic differences DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111024T083000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111025T172500 URL;VALUE=URI:http://www.clt.gu.se/event/2011-10-24/workshop-comparing-approaches-measuring-linguistic-differences LOCATION:Lilla hörsalen\, Humanisten\, Renströmsgatan 6 DESCRIPTION:This workshop is a joint arrangement by the project partners (University of \n Gothenburg\, Uppsala University and the Max Planck Institute for Evolutionary \n Anthropology\, Leipzig). The funding is provided by the Swedish Research \n Council's research projects and by the The Royal Swedish Academy of Letters\, \n History and Antiquities.\n \n Organizing committee\n Lars Borin (University of Gothenburg)\n Anju Saxena (Uppsala University)\n K. Taraka Rama (University of Gothenburg)\n \n For more information\, see Språkbanken's web site.\n \n Workshop web page: http://spraakbanken.gu.se/eng/workshop2011 [1]\n \n Schedule: \n http://spraakbanken.gu.se/eng/research/digital-areal-linguistics/worksho... \n [2]\n \n \n [1] http://spraakbanken.gu.se/eng/workshop2011\n [2] http://spraakbanken.gu.se/eng/research/digital-areal-linguistics/workshop-october-2011/program/program END:VEVENT BEGIN:VEVENT UID:calendar:1300:field_eventdate:0:97 SUMMARY:CLT seminar: Torbjörn Lager - Web Science: A (computational) linguistics \n perspective on the Web DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111027T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111027T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-10-27/clt-seminar-torbj%C3%B6rn-lager LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Jenny Myrendal (Ph.D. student in linguistics) and I have received funding for \n the Web Science project "Ett (data-)lingvistiskt perspektiv på webben". \n Since Web Science (see http://webscience.org [1]) is an area where CLT wants \n to contribute (all according to the CLT Strategic Plan)\, I will give an \n overview of our project plans and also (with your help!) try to answer the \n question: What does linguistics and computational linguistics have to to with \n the Web?\n \n \n [1] http://webscience.org END:VEVENT BEGIN:VEVENT UID:calendar:1356:field_eventdate:0:98 SUMMARY:Extra seminar: Daniela Oelke - Visual Exploration of SALDO and Web Session \n Logs DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111027T131500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111027T150000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-10-27/extra-seminar-daniela-oelke-visual-exploration-saldo-and-web-session-logs LOCATION:L307\, Lennart torstenssonsgatan 8 DESCRIPTION:In the talk I am going to give you an example for an exploratory analysis of \n the lexical resource SALDO using the tool visone. I will furthermore show \n some first results of our visual analysis of Web Session Logs. END:VEVENT BEGIN:VEVENT UID:calendar:1301:field_eventdate:0:99 SUMMARY:CLT seminar: Markus Forsberg - A green and fuzzy Swedish wordnet DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111103T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111103T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-11-03/clt-seminar-markus-forsberg LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Swesaurus is a free Swedish wordnet that we are developing in the SweFN++ \n project. It is built using existing free lexical resources that gives \n Swesaurus some interesting properties such as fuzzy synsets.\n \n The meat of this talk is my experience of linking SALDO\, the backbone of \n Swesaurus\, to Princeton Core Wordnet\, but I will also give a general \n introduction and sketch our road map to Swesaurus 1.0\, which we are planning \n to release in the beginning of next year. END:VEVENT BEGIN:VEVENT UID:calendar:1302:field_eventdate:0:100 SUMMARY:CLT seminar: Simon Dobnik - Grounding the meaning of spatial descriptions DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111110T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111110T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-11-10/clt-seminar-simon-dobnik LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:In this talk I will discuss how the meaning of spatial descriptions such as \n "the chair is to the left of the table" is grounded in perception. I will \n outline my previous work with mobile robots where the mappings between words \n and perceptual representations were learned automatically from observations. \n I will discuss how spatial descriptions are also grounded in dialogue - \n through the alignment between the conversational partners. Finally\, I will \n present ideas towards a more complex model of grounding that also includes \n our knowledge about the interaction of objects in the world. END:VEVENT BEGIN:VEVENT UID:calendar:1303:field_eventdate:0:101 SUMMARY:CLT seminar: Pierre Nugues & Peter Exner- Semantic analysis and applications DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111117T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111117T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-11-17/clt-seminar-pierre-nugues LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:In this presentation\, we will give an overview of the tools and algorithms we \n are developing at LTH in the field of semantic processing. We will review the \n systems we have designed in the past two years and we will introduce our \n goals for the next two years. We will also describe in more detail two \n outcomes of our recent project Semantica: the construction of large \n proposition databases extracted from Wikipedia and the derivation a subset of \n events formatted in the LODE format. As LODE uses RDF triples\, users can \n query the event repository using the SPARQL language and extract information \n in a way similar to SQL databases. Finally\, we will outline our on-going \n effort to apply semantic parsing to Swedish. The presentation will include \n demonstrations. (http://nlp.cs.lth.se/demonstrations/ [1])\n \n http://cs.lth.se/pierre_nugues [2]\n \n \n [1] http://nlp.cs.lth.se/demonstrations/\n [2] http://cs.lth.se/pierre_nugues END:VEVENT BEGIN:VEVENT UID:calendar:1373:field_eventdate:0:102 SUMMARY:Linguistics seminar: Adam Kilgarriff - Ways of thinking about the Web DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111122T131500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111122T150000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-11-22/linguistics-seminar-adam-kilgarriff-ways-thinking-about-web LOCATION:Olof Wijksgatan 6 DESCRIPTION:The web is making all sorts of differences to how we live\, work\, shop\, study\, \n vote and chat. But it seems to change depending on how we look at it.  \n Sometimes it seems like a library\, at others like a shopping mall\, at others \n like a party.  How do companies like Google\, Facebook\, Microsoft and Apple \n fit into the picture?  Is every conversation I have on skype 'on the web'?  \n Is it something that used to be public\, but is now turning private?  The \n question I'll be battling with in the talk is how these different threads fit \n together.\n \n http://www.kilgarriff.co.uk/ [1]\n \n \n [1] http://www.kilgarriff.co.uk/ END:VEVENT BEGIN:VEVENT UID:calendar:1304:field_eventdate:0:103 SUMMARY:CLT seminar: Adam Kilgarriff - Corpus compilation and analysis with the \n Sketch Engine DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111124T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111124T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-11-24/clt-seminar-adam-kilgarriff LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:The Sketch Engine [1] is a corpus query tool and web service that allows \n users to build and explore corpora.  After an introduction to the core \n functionality of the query tool\, we shall see how corpora can be built from \n the web and also how a corpus you already have on your computer can be \n prepared for encoding in the Sketch Engine [2]\, uploaded\, and encoded\, and \n once it is there\, what the range of search options and reports are.\n \n http://www.kilgarriff.co.uk/ [3]\n \n \n [1] http://www.sketchengine.co.uk/\n [2] http://www.sketchengine.co.uk/\n [3] http://www.kilgarriff.co.uk/ END:VEVENT BEGIN:VEVENT UID:calendar:1305:field_eventdate:0:104 SUMMARY:CLT seminar: Taraka Rama - Summary of first year's work as a GSLT PhD student DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111201T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111201T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-12-01/clt-seminar-taraka-rama LOCATION:L308\, Lennart Torstenssonsgatan 8 END:VEVENT BEGIN:VEVENT UID:calendar:1306:field_eventdate:0:105 SUMMARY:CLT seminar: Maarten de Rijke - Web Log Analysis DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111208T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111208T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-12-08/clt-seminar-marten-de-rijke LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:The talk starts from the position that intelligence is about finding \n relationships in temporal phenomena and using that knowledge to make \n predictions and adapt to one's environment. As more and more of everyday life \n is being digitized\, the range of data accessible for mining purposes and the \n types of prediction that one can make grows rapidly.\n \n A large part of the talk will be centered around mining search engine logs: \n they form a very direct reflection of what it is that occupies people. In \n addition\, I will look at a number of scenarios in which we follow and \n interpret similar streaming signals and use them for prediction purposes. I \n will conclude with observations on new search paradigms that emerge with \n abundant human information signals.\n \n http://staff.science.uva.nl/~mdr/ [1]\n \n \n [1] http://staff.science.uva.nl/~mdr/ END:VEVENT BEGIN:VEVENT UID:calendar:1396:field_eventdate:0:106 SUMMARY:Grammatikseminariet: Ett svenskt konstruktikon DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111213T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111213T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-12-13/grammatikseminariet-ett-svenskt-konstruktikon LOCATION:L307\, Lennart Torstenssonsgatan 8 DESCRIPTION:I anslutning till projektet Svenskt frasnät (FN++) är vi på väg att \n utveckla en konstruktionsdatabas\, ett konstruktikon\, inspirerat av det \n engelskspråkiga constructicon som är under uppbyggnad vid UC Berkeley som \n ett tillägg till FrameNet. Tanken är att komplettera befintliga \n lexikografiska resurser med en uppsättning beskrivningar av strukturer som \n är för generella för att kunna knytas till enskilda uppslagsord (men för \n specifika för att betraktas som generella grammatiska regler): \n konstruktioner. Analyserna bör vara relativt enkla\, ungefär motsvarande \n ordboksdefinitioner\, för att möjliggöra beskrivningar i stor skala. \n Resursen ska helst vara till nytta för såväl språkteknologiska som \n lingvistiska och pedagogiska ändamål. Vid seminariet den 13/12 diskuteras \n databasens innehåll och format utifrån några preliminära analysskisser.\n \n Rekommenderad bakgrundsläsning\, för den som ev. vill förbereda sig: \n Fillmore\, Charles\, Russell Lee-Goldman & Russell Rhodes (to appear). The \n FrameNet Constructicon. To appear in Boas\, Hans and Ivan Sag (eds.)\, \n Sign-Based Construction Grammar. Stanford: CSLI. \n \n \n Välkomna!\n \n /Benjamin\n \n \n [1] http://www.icsi.berkeley.edu/pubs/ai/framenetconstructicon11.pdf END:VEVENT BEGIN:VEVENT UID:calendar:1307:field_eventdate:0:107 SUMMARY:CLT seminar: Detmar Meurers - Comparing Meaning in Context DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111215T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20111215T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2011-12-15/clt-seminar-meurers LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:How can meaning be compared and evaluated in realistic situations\, in which \n ill-formed language or differences in situative knowledge or world knowledge \n make a complete analysis difficult or impossible? In this talk I report on \n our ongoing work in the CoMiC project (SFB833-A4). We are exploring which \n linguistic representations are effective and robust in a \n computational-linguistic comparison of the meaning of sentences and text \n fragments. We focus on learner answers to reading comprehension questions\, \n for which we are collecting large authentic corpora. Using these \n task-specific corpora\, we study the properties of questions and texts which \n support the comparison of the meaning of answers at different levels of \n analysis from surface forms to deeper linguistic representations.\n \n  \n \n Professor Detmar Meurers on the web: http://www.sfs.uni-tuebingen.de/~dm/ [1]\n \n \n [1] http://www.sfs.uni-tuebingen.de/~dm/ END:VEVENT BEGIN:VEVENT UID:calendar:1420:field_eventdate:0:108 SUMMARY:Masters seminar: Malin Ahlberg - Towards a wide-coverage grammar for Swedish DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120109T100000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120109T110000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-01-09/masters-seminar-malin-ahlberg LOCATION:The EDIT room at Chalmers END:VEVENT BEGIN:VEVENT UID:calendar:1419:field_eventdate:0:109 SUMMARY:Members' meeting: The future of CLT DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120117T120000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120117T133000 URL;VALUE=URI:http://www.clt.gu.se/event/2012-01-17/members-meeting-future-clt LOCATION:K333\, Lennart Torstenssonsgatan 6 END:VEVENT BEGIN:VEVENT UID:calendar:1418:field_eventdate:0:110 SUMMARY:CLT seminar planning DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120119T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120119T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-01-19/clt-seminar-planning LOCATION:L308\, Lennart torstenssonsgatan 8 DESCRIPTION:Planning event for the CLT seminar series Spring semester END:VEVENT BEGIN:VEVENT UID:calendar:1455:field_eventdate:0:111 SUMMARY:CANCELLED DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120126T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120126T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-01-26/cancelled LOCATION:L308\, Lennart torstenssonsgatan 8 END:VEVENT BEGIN:VEVENT UID:calendar:1476:field_eventdate:0:112 SUMMARY:Final PhD thesis seminar: Dana Dannélls DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120127T090000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120127T110000 URL;VALUE=URI:http://www.clt.gu.se/event/2012-01-27/final-phd-thesis-seminar-dana-dann%C3%A9lls LOCATION:L308\, Lennart torstenssonsgatan 8 DESCRIPTION:Dana Dannélls at the department of Swedish and Språkbanken will present her \n upcoming PhD thesis "Multilingual text generation from structured formal \n representations".\n \n Opponent: Richard Power\, The Open University\, UK.\n \n The thesis draft can be downloaded here: \n http://spraakdata.gu.se/svedd/finalseminar-120127.pdf [1].\n \n \n [1] http://spraakdata.gu.se/svedd/finalseminar-120127.pdf END:VEVENT BEGIN:VEVENT UID:calendar:1454:field_eventdate:0:113 SUMMARY:Licentiate thesis defense: Shafqat Virk DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120127T131500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120127T151500 URL;VALUE=URI:http://www.clt.gu.se/event/2012-01-27/licentiate-thesis-defense-shafqat-virk LOCATION:Steve Jobs\, plan 2 Patricia\, Forskningsgången 6\, Lindholmen DESCRIPTION:Shafqat Virk (graduate student at Applied IT) will defend his thesis for the \n licentiate degree: "Computational Grammar Resources for Indo-Iranian \n Languages"\n \n Discussion leader: Hans Leiß\n \n Abstract:\n \n A lot of research is being carried out on different aspects of natural \n language processing (NLP). As a result\, there exist state of the art machine \n translation systems such as Google translate. These systems are based on data \n driven (i.e. statistical) approaches\, which provide huge coverage\, but at the \n cost of limited accuracy. On the other hand knowledge intensive (i.e. grammar \n based) approaches provide high quality translations\, but their coverage is \n limited. Generally speaking\, one can say that a combined solution with \n human-like capabilities for accuracy and coverage is yet too far to be \n reached. One major reason of this limitation is the fact that natural \n languages are complex and ambiguous. This makes it a challenging task to \n develop a computational grammar of a natural language. Developing a grammar \n of a natural language requires at least comprehensive knowledge of the \n language\, expertise to describe it\, and also practical skills with a grammar \n formalism tool. This thesis is devoted to the development of computational \n grammars of four Indo-Iranian languages: Nepali\, Persian\, Punjabi\, and Urdu. \n We explore different lexical and syntactical features of these languages and \n develop their resource grammars according to the requirements of Grammatical \n Framework (GF) resource grammar API.\n \n Grammatical Framework (GF) is a grammar formalism tool\, which has been used \n to develop grammars of a number of natural languages. So far most of these \n languages belong to the Germanic\, Romance\, or Slavic branches of the Indo- \n European family of languages. On the other hand\, Indo-Iranian is the largest \n branch of the Indo-European family. With 310 living languages\, this branch \n comprises 70% of the total languages in its family. Most of these \n Indo-Iranian languages have either very limited computational resources or no \n resources at all. This is one reason of developing resource grammars of these \n resource-limited languages.\n \n Furthermore\, Indo-Iranian languages have some distinctive features such as \n the Ezafe construction and the partial ergativigy. Previously\, none of these \n features has been implemented in GF. Another reason of this study is to \n explore this dimension\, and demonstrate implementation of these features in \n GF END:VEVENT BEGIN:VEVENT UID:calendar:1475:field_eventdate:0:114 SUMMARY:CLT seminar: Leif-Jöran Olsson - Standardised metadata for discoverability \n and exchange of LT resources within the open infrastructure META-NORD DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120202T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120202T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-02-02/clt-seminar-leif-j%C3%B6ran-olsson-standardised-metadata-discoverability-and-exchange-lt-resources-within LOCATION:L308\, Lennart torstenssonsgatan 8 DESCRIPTION:META-NORD is covering the establishment of the Baltic and NORDic parts of an \n open resource exchange infrastructure within the European Open Linguistic \n Infrastructure META-SHARE.\n \n The focus of the project is to assemble\, link across languages\, and make \n widely available language resources of different types for use in academia \n and industry to specific products and applications. Among the core resources \n are treebanks\, wordnets and multilingual terminologies. This seminar focuses \n on the discoverability part and will hopefully give an interactive view on \n the current metadata model supporting this goal.\n \n The work within the project META-NORD has received funding from the ICT \n Policy Support Programme as part of the Competitiveness and Innovation \n Framework Programme\, Grant agreement no 270899 END:VEVENT BEGIN:VEVENT UID:calendar:1472:field_eventdate:0:115 SUMMARY:Seminar at CSE: Ann Copestake - Formal semantics and dependency structures DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120207T151500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120207T170000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-02-07/seminar-cse-ann-copestake LOCATION:The EDIT room at Chalmers DESCRIPTION:Ann Copestake (University of Cambridge Computer Laboratory).\n \n Abstract:\n \n Logical representations and dependency structures are both used to describe \n aspects of the meaning of natural language sentences\, but are formally very \n different. In this talk\, I will show that one widely used form of logical \n representation can be transformed into graph structures comparable to \n dependency representations without loss of information. This has some \n significant practical advantages for language processing\, which I will \n discuss in the context of the DELPH-IN (Deep Linguistic Processing with HPSG) \n collaboration.\n \n Professor Copestake on the Web: http://www.cl.cam.ac.uk/~aac10/ [1]\n \n \n [1] http://www.cl.cam.ac.uk/~aac10/ END:VEVENT BEGIN:VEVENT UID:calendar:1450:field_eventdate:0:116 SUMMARY:Thesis defense: Krasimir Angelov DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120208T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120208T130000 URL;VALUE=URI:http://www.clt.gu.se/event/2012-02-08/thesis-defense-krasimir-angelov LOCATION:EDIT building\, Chalmers DESCRIPTION:Krasimir Angelov at CSE defends his PhD thesis "The Mechanics of the \n Grammatical Framework".\n \n Faculty Opponent: Mark Steedman\n \n More info: \n http://publications.lib.chalmers.se/cpl/record/index.xsql?pubid=149979 [1]\n \n Full text: http://www.cse.chalmers.se/%7Ekrasimir/phd-thesis.pdf [2]\n \n \n [1] http://publications.lib.chalmers.se/cpl/record/index.xsql?pubid=149979\n [2] http://www.cse.chalmers.se/%7Ekrasimir/phd-thesis.pdf END:VEVENT BEGIN:VEVENT UID:calendar:1453:field_eventdate:0:117 SUMMARY:CLT seminar: Mark Steedman - The Statistical Problem of Language Acquisition DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120209T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120209T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-02-09/clt-seminar-mark-steedman LOCATION:L308\, Lennart torstenssonsgatan 8 DESCRIPTION:The talk reports recent work with Tom Kwiatkowski\, Sharon Goldwater\, and Luke \n Zettlemoyer on semantic parser induction by machine from a number of corpora \n pairing sentences with logical forms\, including GeoQuery and a corpus \n consisting of real child-directed utterance from the CHILDES corpus.\n \n The problem of semantic parser induction and child language acquisition are \n both similar to the problem of inducing a grammar and a parsing model from a \n treebank such as the Penn treebank\, except that the trees are unordered \n logical forms\, in which the preterminals are not aligned with words in the \n target language\, and there may be noise and spurious distracting logical \n forms supported by the context but irrelevant to the utterance.\n \n The talk shows that this class of problem can be solved if the child or \n machine initially parses with the entire space of possibilities that \n universal grammar allows under the assumptions of the Combinatory Categorial \n theory of grammar (CCG)\, and learns a statistical parsing model for that \n space using EM-related methods such as Variational Bayes learning.\n \n This can be done without all-or-none "parameter-setting" or attendant \n "triggers"\, and without invoking any "subset principle" of the kind proposed \n in linguistic theory\, provided the system is presented with a representative \n sample of reasonably short string-meaning pairs from the target language.\n \n Mark Steedman on the Web [1]\n \n \n [1] http://homepages.inf.ed.ac.uk/steedman/ END:VEVENT BEGIN:VEVENT UID:calendar:1457:field_eventdate:0:118 SUMMARY:CLT seminar: Krasimir Angelov\, John Camilleri and Gerardo Schneider - A \n Framework for Conflict Analysis of Normative Texts Written in Controlled \n Natural Language DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120216T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120216T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-02-16/clt-seminar-angelov-camilleri-schneider LOCATION:EDIT room\, EDIT building\, Chalmers DESCRIPTION:Our aim is to detect whether texts written in natural language contain \n normative conflicts (i.e.\, whether there are conflicting obligations\, \n permissions and prohibitions). In this talk we present AnaCon\, a framework \n where such texts are written in Controlled Natural Language (CNL) and \n automatically translated into the formal language CL using the Grammatical \n Framework (GF). In AnaCon such CL expressions are analyzed for normative \n conflicts by the tool CLAN which gives a counter-example in case a conflict \n is found. Our framework uses GF  to give a CNL version of the \n counter-example helping the user to identify the conflicts in the original \n text.\n \n The talk will be divided in two parts. In the first part we will introduce \n AnaCon as well as relevant background on CL and CLAN. In the second part we \n will we show the usability of AnaCon on two case studies. END:VEVENT BEGIN:VEVENT UID:calendar:1458:field_eventdate:0:119 SUMMARY:CLT seminar: Karin Friberg Heppin and Ann-marie Eklund - Highlights from \n PROMISE winter school 2012 DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120223T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120223T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-02-23/clt-seminar-karin-friberg-heppin-and-ann-marie-eklund LOCATION:L308\, Lennart torstenssonsgatan 8 DESCRIPTION:In January 2012 Karin Friberg Heppin and Ann-Marie Eklund participated in \n PROMISE winter school on Information Retrieval meets Information \n Visualization. In this presentation we will briefly describe the two main \n topics of the winter school\, Information Retrieval and Information \n Visualization. We will also present a few highlights from the lectures.\n \n http://www.promise-noe.eu/events/winter-school-2012 [1]\n \n \n [1] http://www.promise-noe.eu/events/winter-school-2012 END:VEVENT BEGIN:VEVENT UID:calendar:1452:field_eventdate:0:120 SUMMARY:Licentiate thesis defense: Ramona Enache DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120229T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120229T120000 URL;VALUE=URI:http://www.clt.gu.se/event/2012-02-29/licentiate-thesis-defense-ramona-enache LOCATION:Room EF\, EDIT building\, Chalmers DESCRIPTION:Ramona Enache (graduate student at CSE) will defend her thesis for the \n licentiate degree: "Automating the development of multilingual grammars"\n \n Opponent: Gordon Pace\, University of Malta\n \n Abstract:\n \n The thesis aims at exploring alternative ways of automating the development \n of multilingual GF (Grammatical Framework) grammars. The goal is to achieve \n semantics-preserving machine translation within a limited or semi-limited \n domain.\n \n First\, there is an experiment that investigates the relation between language \n skills\, programming skills and the effort to develop a grammar for natural \n language. Along with this\, we present a prototype for an example-based system \n aimed at automating and simplifying the task of GF grammar writing. This is \n done by partially alleviating the burden of GF programming and facilitating \n the integration of SMT (statistical machine translation) tools and feedback \n from native informants into the development of a GF grammar.\n \n Secondly\, there is work on ontology representation in GF. The goal is to \n automatically build a robust language-independent semantic interlingua for \n the multilingual grammars\, by using the projection of the ontology as a GF \n grammar. In this way the ontology can be verbalized with little effort into a \n number of languages. The resulting GF grammar is a controlled language \n describing ontological concepts and their relations as defined in the initial \n ontology.\n \n Another approach towards automating and enhancing GF grammars is the \n grammar-based approach to a rule-based SMT hybrid translation system. In \n order to increase the coverage of a grammar\, we enriched it with a bilingual \n lexicon built on the fly\, with the aid of an SMT system specialised on the \n corpus. The goal of the hybrid system is to parse English patent claims from \n the biomedical domain and translate them to French. The work represents the \n first large-scale experiment to use GF for parsing arbitrary unstructured \n text.\n \n A final direction considered is the development of a general language \n resource for Romanian. It can be used to build domain-specific resources \n which will use it as a library for handling syntactic phenomena. In this way \n more multilingual GF grammars can be ported to Romanian\, without the need to \n re-implement the linguistic technicalities of the language every time. END:VEVENT BEGIN:VEVENT UID:calendar:1459:field_eventdate:0:121 SUMMARY:CLT seminar: Elena Volodina & Sofie Johansson Kokkinakis - Introducing the \n Swedish Kelly-list\, a new lexical e-resource for Swedish DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120301T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120301T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-03-01/clt-seminar-volodina-johansson-kokkinakis LOCATION:L308\, Lennart torstenssonsgatan 8 DESCRIPTION:Frequency lists and frequency-based lexical resources contain information \n about the words and their statistics. They tend to find their "readers" among \n language learners\, language teachers\, linguists and lexicographers. Making \n them available in electronic format helps to expand the target group to cover \n language engineers\, computer programmers and other specialists working in \n such areas as information retrieval\, spam filtering\, text readability \n analysis\, test generation etc.\n \n In this presentation we describe a new freely available electronic frequency \n list of modern Swedish which was created in the EU project KELLY. We provide \n a short description of the KELLY project\; examine the methodological approach \n and provide some details on the compiling of the web-corpus SweWAC from which \n the list has been derived. Further\, we discuss the type of information the \n list contains\, describe the steps for list generation\; provide information on \n the coverage and some other statistics over the items in the list. Finally\, \n we suggest some potential application areas\; and mention future plans. We \n hope that with some publicity we can help this list find its users. END:VEVENT BEGIN:VEVENT UID:calendar:1460:field_eventdate:0:122 SUMMARY:CLT seminar: Tobias Berg\, Björn Klockljung Johansson and Svetoslav Marinov - \n Designing and implementing search solutions - The Findwise way DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120308T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120308T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-03-08/clt-seminar-berg-johansson-and-svetoslav-marinov LOCATION:L308\, Lennart torstenssonsgatan 8 DESCRIPTION:In this talk we will tell you more about Findwise as a company and how we \n design search solutions. We will focus on the collection and processing of \n the data as well as  its presentation to the end user. Our focus is on bring \n most value to the customers give their structured and unstructured data. END:VEVENT BEGIN:VEVENT UID:calendar:1473:field_eventdate:0:123 SUMMARY:CLT seminar: Sandra Derbring & Peter Ljunglöf - A Sample GF Grammar for the \n Concept Coding Framework and Swedish DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120315T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120315T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-03-15/clt-seminar-derbring-ljunglof LOCATION:L308\, Lennart torstenssonsgatan 8 DESCRIPTION:The Concept Coding Framework (CCF) is a collection of ontologies about the \n meaning of words and how to represent them in a multimodal and multilingual \n context. This technology is currently applied in the European AEGIS project \n to provide graphical symbol support for users of Augmentative and Alternative \n Communication (AAC).\n \n In this project we have developed a sample GF grammar for translating between \n CCF/Bliss and Swedish. This grammar has the potential to be used for \n applications like:\n \n * Symbol supported text editing and reading within the free\n OpenOffice/LibreOffice Writer application\n * A symbol and speech supported AAC application on mobile phone devices for\n direct and indirect communication.\n \n Sandra Derbring (DART) and Peter Ljunglöf (CSE) END:VEVENT BEGIN:VEVENT UID:calendar:1461:field_eventdate:0:124 SUMMARY:CANCELLED! CLT seminar: Stefan Gries - The corpus-based behavioral-profile \n approach to lexical semantics DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120322T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120322T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-03-22/clt-seminar-stefan-gries-corpus-based-behavioral-profile-approach-lexical-semantics LOCATION:L308\, Lennart torstenssonsgatan 8 DESCRIPTION:For more information\, see the following publication: \n http://www.linguistics.ucsb.edu/faculty/stgries/research/2010_STG_BehavProf_TheMentalLexicon.pdf \n [1]\n \n \n [1] http://www.linguistics.ucsb.edu/faculty/stgries/research/2010_STG_BehavProf_TheMentalLexicon.pdf END:VEVENT BEGIN:VEVENT UID:calendar:2979:field_eventdate:0:125 SUMMARY:IR seminar: Peter Ingwersen - Basic conceptual models for Information \n Retrieval research DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120323T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120323T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-03-23/ir-seminar-peter-ingwersen LOCATION:C363 DESCRIPTION:After defining Information Retrieval (IR) the presentation discusses the \n mainstream laboratory research framework based on the so-called Cranfield \n Model as well as selected interactive and task-based and contextual IR \n models.\n \n http://pure.iva.dk/en/persons/peter-ingwersen(06732ae6-b7ec-4427-8049-6b087f1e5f6d).html \n [1]\n \n \n [1] http://pure.iva.dk/en/persons/peter-ingwersen(06732ae6-b7ec-4427-8049-6b087f1e5f6d).html END:VEVENT BEGIN:VEVENT UID:calendar:1462:field_eventdate:0:126 SUMMARY:CLT seminar: Steven Krauwer - Towards a Schengen Area for Language Resources \n and Technology DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120329T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120329T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-03-29/clt-seminar-steven-krauwer LOCATION:L308\, Lennart torstenssonsgatan 8 DESCRIPTION:In my talk I will briefly describe what CLARIN wants to achieve and what \n needs to be done in the participating countries to transform the present \n fragmented European landscape to a Schengen area for access to language \n resources and technology for the humanities and social sciences research \n community END:VEVENT BEGIN:VEVENT UID:calendar:1463:field_eventdate:0:127 SUMMARY:CLT seminar: Kenneth Wilhelmsson - Swedish Adverbials and Question \n Correspondences DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120405T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120405T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-04-05/clt-seminar-kenneth-wilhelmsson LOCATION:L308\, Lennart torstenssonsgatan 8 DESCRIPTION:Whereas nominals in Swedish running text correspond to a (virtually) small \n group of wh-counterparts in questions\, it is clear from the beginning that \n adverbials as a group are diverse in comparison.\n \n From a computational perspective\, the task of determining question \n counterparts presupposes identification of full adverbials\, their syntactic \n heads and often heads of complements. But also information about modifiers\, \n the full sentence (its main verb) or even the placement of the adverbial can \n be essential for correct question generation. This is due to phenomena that \n could perhaps be more clearly explained through theory of information \n structure. END:VEVENT BEGIN:VEVENT UID:calendar:1464:field_eventdate:0:128 SUMMARY:CLT seminar: Richard Johansson - Coping with data scarcity in the Swedish \n FrameNet DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120412T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120412T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-04-12/clt-seminar-richard-johansson LOCATION:L308\, Lennart torstenssonsgatan 8 DESCRIPTION:Semantic role classification accuracy for most languages other than English \n is constrained by the small amount of annotated data. In this talk\, we \n describe how the frame-to-frame relations described in the FrameNet ontology \n can be used to improve the performance of a FrameNet-based semantic role \n classifier for Swedish\, a low-resource language. In order to make use of the \n FrameNet relations\, we cast the semantic role classification task as a \n non-atomic label prediction task. The experiments show that the cross-frame \n generalization methods lead to a significant reduction in the number of \n errors made by the classifier. This reduction is especially notable when the \n classifier is evaluated on previously unseen frames. END:VEVENT BEGIN:VEVENT UID:calendar:2980:field_eventdate:0:129 SUMMARY:IR seminar: Anni Järvelin - Different approaches to translation in \n cross-language information retrieval: what to translate and how? DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120413T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120413T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-04-13/ir-seminar-anni-jarvelin LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Cross-language information retrieval is information retrieval where some form \n of translation is involved. To reach an adequate level of translation \n quality\, issues related to inflection\, compounds and phrases\, among others\, \n also need to be solved. I will start the seminar by giving a short \n introduction to information retrieval and cross-language information \n retrieval (CLIR). The rest of the seminar will give a superficial \n introduction to the central issues of natural language processing in CLIR: \n What to translate\, how to translate\, what are the biggest problems and the \n suggested solutions? END:VEVENT BEGIN:VEVENT UID:calendar:1465:field_eventdate:0:130 SUMMARY:CLT seminar: Grégoire Détrez - Smart Paradigms and the Predictability and \n Complexity of Inflectional Morphology DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120419T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120419T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-04-19/clt-seminar-gr%C3%A9goire-d%C3%A9trez LOCATION:EDIT room\, EDIT building\, Chalmers DESCRIPTION:Morphological lexica are often implemented on top of morphological paradigms\, \n corresponding to different ways of building the full inflection table of a \n word. Computationally precise lexica may use hundreds of paradigms\, and it \n can be hard for a lexicographer to choose among them.\n \n To automate this task\, this paper introduces the notion of a smart paradigm. \n It is a meta-paradigm\, which inspects the base form and tries to infer which \n low-level paradigm applies. If the result is uncertain\, more forms can be \n given for discrimination. The number of forms needed in average is a measure \n of predictability of an inflection system.The overall complexity of the \n system also has to take into account the code size of the paradigms \n definition itself.\n \n This paper investigates inspects the smart paradigms as implemented in the \n open-source GF Resource Grammar Library. Predictability and complexity are \n estimated for four different languages: English\, French\, Swedish\, and \n Finnish. The main result is that predictability does not decrease when the \n complexity of morphology grows\, which means that smart paradigms provide an \n efficient tool for the manually construction and/or automatically \n bootstrapping of lexica.\n \n (This is joint work with Aarne Ranta.) END:VEVENT BEGIN:VEVENT UID:calendar:2983:field_eventdate:0:131 SUMMARY:IR seminar: Anni Järvelin & Karin Friberg Heppin - Information retrieval \n evaluation DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120420T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120420T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-04-20/ir-seminar-jarvelin-heppin LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Systematic and quantifiable evaluation of system components is the hallmark \n of information retrieval research. The laboratory evaluation methodology  \n developed within the field since the 1960s has paved the way for great \n research output and for successful systems in place and in use today. \n Evaluation of the human task performance (instead of system performance) in \n information access is a more difficult - yet no less important -  issue and \n no single accepted evaluation framework for such evaluation exists.\n \n In this seminar\, the laboratory evaluation framework will be first presented \n and discussed through an example test collection (MedEval). Then a short \n introduction to user-oriented information retrieval evaluation is given. END:VEVENT BEGIN:VEVENT UID:calendar:1466:field_eventdate:0:132 SUMMARY:CLT seminar: Shalom Lappin - A Probabilistic Semantics for Natural Language DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120426T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120426T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-04-26/clt-seminar-prel-shalom-lappin LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Probabilistic and stochastic methods have been fruitfully applied to a wide \n variety of problems in grammar induction\, natural language processing\, and \n cognitive modeling. In this paper we explore the possibility of developing a \n class of combinatorial semantic representations for natural languages that \n compute the semantic value of a (declarative) sentence as a probability value \n which expresses the likelihood of competent speakers of the language \n accepting the sentence as true in a given model\, relative to a specification \n of the world. Such an approach to semantic representation treats the \n pervasive gradience of semantic properties as intrinsic to speakers' \n linguistic knowledge\, rather than the result of the interference of \n performance factors in processing and interpretation. In order for this \n research program to succeed\, it must solve three central problems. First\, it \n needs to formulate a type system that computes the probability value of a \n sentence from the semantic values of its syntactic constituents. Second\, it \n must incorporate a viable probabilistic logic into the representation of \n semantic knowledge in order to model meaning entailment. Finally\, it must \n show how the specified class of semantic representations can be efficiently \n learned. We construct a probabilistic semantic fragment and consider how the \n approach that the fragment instantiates addresses each of these three issues.\n \n (Joint work with Jan van Eijck\, CWI\, Amsterdam) END:VEVENT BEGIN:VEVENT UID:calendar:1467:field_eventdate:0:133 SUMMARY:CLT seminar: Frédérique Segond - From electronic atoms of knowledge to \n Bounty: how to make sense of any information pieces to improve quality of \n care and understanding of customers: the ALADIN and GALATEAS examples DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120503T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120503T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-05-03/clt-seminar-fr%C3%A9d%C3%A9rique-segond LOCATION:L308\, Lennart torstenssonsgatan 8 DESCRIPTION:Frédérique Segond\, Research and Development Manager\, Viseo\, France\n \n The focus of this talk is to show how natural language technologies can \n support the process of understanding electronic information at your fingers \n tips. We will present two projects using NLP technologies for this purpose in \n two completely different domains: ALADIN in the domain of healthcare and \n GALATEAS in business domain of content providers.\n \n Acknowledgements:\n ALADIN was supported by The National French Research Agency (Agence nationale \n de recherche - ANR) as part of the TecSan 2008 Framework Programme \n (Technologies pour la Santé et l'Autonomie GALATEAS is supported by the \n European Union’s ICT Policy Support Programme as part of the \n Competitiveness and Innovation Framework Programme\, CIP ICT-PSP under grant \n agreement nr 250430 END:VEVENT BEGIN:VEVENT UID:calendar:2981:field_eventdate:0:134 SUMMARY:IR seminar: Jussi Karlgren - Genres on the web – what is a genre anyway? DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120503T131500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120503T150000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-05-03/ir-seminar-jussi-karlgren LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Genres can be understood in many different ways. They are often perceived as \n a primarily sociological construction\, or\, alternatively\, as a \n stylostatistically or philologically observable objective characteristic of \n texts. The latter view is more common in the research field of information \n and language technology. These two views can be quite compatible and can \n inform each other\; this talk will discusses how one might look at texts from \n the point of their observable features on the one hand and on their usage\, by \n observing reader and author behaviour on the other. The latter view\, if taken \n to its conclusion\, will have effects on the choice of features and their \n aggregation for language technologists. END:VEVENT BEGIN:VEVENT UID:calendar:3025:field_eventdate:0:135 SUMMARY:Grammatikfestivalen/Språkdagen 2012 DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120504T100000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120504T170000 URL;VALUE=URI:http://www.clt.gu.se/event/2012-05-04/grammatikfestivalenspr%C3%A5kdagen-2012 LOCATION:T302\, Olof Wijksgatan 6 DESCRIPTION:For more information\, see the dedicated web page:\n \n http://www.svenska.gu.se/samverkan/grammatikfestivalen/ [1]\n \n \n [1] http://www.svenska.gu.se/samverkan/grammatikfestivalen/ END:VEVENT BEGIN:VEVENT UID:calendar:1468:field_eventdate:0:136 SUMMARY:CLT seminar: Kalervo Järvelin - Light Statistical Morphology for IR DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120510T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120510T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-05-10/clt-seminar-prel-kalervo-j%C3%A4rvelin LOCATION:L308\, Lennart torstenssonsgatan 8 DESCRIPTION:Traditional morphological tools seek to treat the morphological variation of \n a language comprehensively. While the results tend to be good\, at least \n linguistically\, the down side is complexity of construction\, maintenance and \n use of such tools. During the past few years\, several statistical methods for \n morphological processing have been proposed for use in Information Retrieval \n (IR). These methods may be characterized as unsupervised\, semi-supervised\, \n light-weight\, and/or (partially) language-independent. Generally\, they are \n easy to set up for a new language or collection and provide competitive \n results for treating morphological variation in IR. The talk introduces some \n generative and reductive light statistical methods for use in IR and \n discusses their effectiveness and limitations.\n \n Kalervo Järvelin (Tampere)\n \n http://www.uta.fi/~likaja/ [1]\n \n \n [1] http://www.uta.fi/~likaja/ END:VEVENT BEGIN:VEVENT UID:calendar:2982:field_eventdate:0:137 SUMMARY:IR seminar: Kalervo Järvelin - Managing Morphologically Complex Languages in \n Information Retrieval DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120511T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120511T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-05-11/ir-seminar-kalervo-jarvelin DESCRIPTION:Morphologically complex languages are rich in inflectional and derivational \n morphology and compound formation. In contrast to languages like English\, \n Hindi or Chinese\, complex languages may offer tens or even hundreds of \n inflectional forms for nouns\, for instance. Such variation is a challenge to \n Information Retrieval (IR) methods that are based on matching keywords to \n text indexes. The talk discusses reductive (such as stemming and \n lemmatization) and generative (inflectional stem and full word form \n generation) methods for IR in several languages\, covering both index \n construction and query processing. Emphasis is given to IR effectiveness and \n the contribution of morphological processing in this.\n \n Kalrvo Järvelin (Tampere)\n \n http://www.uta.fi/~likaja/ [1]\n \n \n [1] http://www.uta.fi/~likaja/ END:VEVENT BEGIN:VEVENT UID:calendar:3035:field_eventdate:0:138 SUMMARY:CLT seminar: Raquel Fernandez - Towards a Flexible Semantics: Colour Terms in \n Collaborative Reference Tasks DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120529T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120529T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-05-29/clt-seminar-raquel-fernandez LOCATION:T340\, Olof Wijksgatan 6 DESCRIPTION:Speakers  do not always share identical semantic representations nor \n identical lexicons. However\, regardless of these differences\, dialogue \n participants are able to communicate successfully most of the time.\n \n In this talk I will present ongoing joint work with Bert Baumgaertner (UC \n Davis) and Matthew Stone (Rutgers) on the development of agents that can \n implicitly coordinate with their partners in referential tasks\, taking as a \n case study colour terms. I will describe our algorithms for generation and \n resolution of colour descriptions and report results of experiments on how \n humans use colour terms for reference in production and comprehension.\n \n http://staff.science.uva.nl/~raquel/ [1]\n \n \n [1] http://staff.science.uva.nl/~raquel/ END:VEVENT BEGIN:VEVENT UID:calendar:1470:field_eventdate:0:139 SUMMARY:CLT seminar: Anders Søgaard - Perceptron learning under sample bias DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120531T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120531T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-05-31/clt-seminar-anders-s%C3%B8gaard LOCATION:L308\, Lennart torstenssonsgatan 8 DESCRIPTION:Most machine learning methods we use in NLP are methods for learning from \n unbiased labeled data. However\, in NLP we always learn from biased data. When \n we train a parser on a treebank\, for example\, and apply it to emails\, legal \n text\, or university websites\, our training data is biased in terms of genre\, \n style\, recency\, possibly dialect\, etc. In this talk we present learning \n algorithms for automatically correcting bias - or algorithms for learning \n under sample bias.\n \n The first part of the talk focuses on large-margin perceptron learning \n algorithms for learning from weighted data. We discuss sampling vs. weighting \n and different weight functions. In the second part of the talk we consider \n the more challenging scenario where the target data cannot be assumed to form \n a single\, coherent distribution\, but where instead we need to adapt our model \n to every new data point on the fly.\n \n Anders Søgaard at Center For Sprogteknologi\, Copenhagen: \n http://cst.dk/anders/main.html [1]\n \n \n [1] http://cst.dk/anders/main.html END:VEVENT BEGIN:VEVENT UID:calendar:3050:field_eventdate:0:140 SUMMARY:Master thesis seminar: Liza Zimina - GF Japanese Resource Grammar DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120604T131500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120604T143000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-06-04/master-thesis-seminar-liza-zimina LOCATION:T116\, Olof Wijksgatan 6 DESCRIPTION:Liza Zimina in the MLT programme will defend her thesis "GF Japanese Resource \n Grammar". Opponent: Linqiqige Zhuo\n \n ABSTRACT\n \n This thesis describes the implementation of the open-source Japanese resource \n grammar as a part of the GF Resource Grammar Library (RGL). GF (Grammatical \n Framework) is a grammar formalism for multilingual grammars and their \n applications intended at performing various natural language processing \n tasks. The RGL is a set of natural language grammars with a shared abstract \n syntax and different concrete parts implemented in GF.\n \n The Japanese resource grammar covers all the categories and rules of the GF \n abstract syntax\, thus providing the full correlation with the resource \n grammars of other languages in the RGL. Due to some peculiarities of the \n Japanese language\, the process of grammar development was complicated by a \n number of challenges that had not been observed before in the implementation \n of other GF grammars. One of the most important peculiarities is stylistic \n stratification in Japanese\, which deeply affects its morphology and syntax. \n Moreover\, a number of syntactic constructions predetermined in the abstract \n syntax are hardly possible in the natural Japanese language (e.g. some types \n of complex subordinate clauses). Being typologically distant from the \n European languages\, Japanese brings up new issues in the discussion on the \n universal properties of languages and disputes the generality of some rules \n in the GF abstract syntax.\n \n Full text (preliminary version): \n http://www.ling.gu.se/~lager/MLT/zimina_thesis_draft.pdf [1]\n \n \n [1] http://www.ling.gu.se/~lager/MLT/zimina_thesis_draft.pdf END:VEVENT BEGIN:VEVENT UID:calendar:1471:field_eventdate:0:141 SUMMARY:CLT seminar: Reut Tsarfaty - Statistical Parsing in the Face of Diversity DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120607T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120607T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-06-07/clt-seminar-reut-tsarfaty LOCATION:L308\, Lennart torstenssonsgatan 8 DESCRIPTION:Statistical parsers aim to automatically discover a set of \n language-independent relations between elements such as a Subject\, a \n Predicate or an Object\, based on their realization patterns in the data of \n different languages. A Subject in English\, for example\, is realized in syntax \n using word order\, while in German it is realized in morphology\, using word \n affixes. The cross-linguistic diversity in the realization of grammatical \n relations has dramatic effects on parsing accuracy — existing statistical \n parsing models demonstrate excellent performance on English\, but when trained \n on data from other languages they often fail to yield comparable results. A \n research question thus emerges\, namely\, what kind of models are suitable for \n parsing different languages?\n \n In this talk I motivate\, develop and demonstrate the application of a \n Relational-Realizational (RR) parsing model which is designed to cope with \n cross-linguistic diversity by mapping grammatical relations to \n morphosyntactic realization in a non-rigid\, language-independent\, fashion. \n The model is defined over a formal grammar that inter-relates function\, \n syntax and morphology. The model parameters encode complex interactions\, \n which\, for particular languages\, are estimated based on corpus statistics. I \n demonstrated the application of the model to parsing Hebrew and Swedish\, \n showing significant improvement without paying any computational costs.\n \n ------------------------------------------------------------------------------\n Reut Tsarfaty is Post-Doctoral Researcher at the Computational Linguistics \n lab at Uppsala University in Sweden\, focusing on technologies and evaluation \n methods for cross-linguistic and cross-framework statistical parsing. She \n received her Ph.D. and MSc. from the Institute for Logic\, Language and \n Computation (ILLC) at the University of Amsterdam\, and her BSc. from the \n Computer Science department at the Technion. Reut is an expert in \n cross-linguistic processing and is interested in particular in modeling rich \n morphosyntactic and morphosemantic interactions. Reut is a recipient of the \n Dutch Science Foundation's prestigious MOSAIC award and she is now writing a \n book on "Parsing Morphologically Rich Languages (PMRL)" to be published by \n Morgan and Claypool in the summer of 2013.\n \n http://stp.lingfil.uu.se/~tsarfaty/ [1]\n \n \n [1] http://stp.lingfil.uu.se/~tsarfaty/ END:VEVENT BEGIN:VEVENT UID:calendar:3053:field_eventdate:0:142 SUMMARY:Guest lecture: Martin Kay - The new machine translation - Getting blood from \n a stone DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120614T093000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120614T103000 URL;VALUE=URI:http://www.clt.gu.se/event/2012-06-14/guest-lecture-martin-kay LOCATION:room EE\, EDIT building\, Rännv 6B\, Chalmers DESCRIPTION:Professor Martin Kay (Stanford U.\, Saarland U. and dr h.c. at Gothenburg U.) \n is invited speaker at the third international workshop on Free/Open-Source \n Rule-based Machine Translation (http://www.chalmers.se/hosted/freerbmt12-en/ \n [1]).\n \n More information here: \n http://www.chalmers.se/cse/EN/news/calendar-events/new-machine-translation \n [2]\n \n \n [1] http://www.chalmers.se/hosted/freerbmt12-en/\n [2] http://www.chalmers.se/cse/EN/news/calendar-events/new-machine-translation END:VEVENT BEGIN:VEVENT UID:calendar:3048:field_eventdate:0:143 SUMMARY:Licentiate Seminar: Håkan Burden - Three Studies on Model Transformations - \n Parsing\, Generation and Ease of Use DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120618T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120618T130000 URL;VALUE=URI:http://www.clt.gu.se/event/2012-06-18/licentiate-seminar-burden LOCATION:room EE\, EDIT building\, Rännv 6B\, Chalmers DESCRIPTION:Håkan Burden\, GSLT PhD student at CSE\, will defend his licentiate thesis \n "Three Studies on Model Transformations - Parsing\, Generation and Ease of \n Use".\n \n Opponent/discussion leader: Leon Moonen\, Simula Research Laboratory\, Oslo\, \n Norway\n \n Abstract:\n \n Transformations play an important part in both software development and the \n automatic processing of natural languages. We present three publications \n rooted in the multi-disciplinary research of Language Technology and Software \n Engineering and relate their contribution to the literature on syntactical \n transformations.\n \n /Parsing Linear Context-Free Rewriting Systems/\n The first publication describes four different parsing algorithms for the \n mildly context-sensitive grammar formalism Linear Context-Free Rewriting \n Systems. The algorithms automatically transform a text into a chart. As a \n result the parse chart contains the (possibly partial) analysis of the text \n according to a grammar with a lower level of abstraction than the original \n text. The uni-directional and endogenous transformations are described within \n the framework of parsing as deduction.\n \n /Natural Language Generation from Class Diagrams/\n Using the framework of Model-Driven Architecture we generate natural language \n from class diagrams. The transformation is done in two steps. In the first \n step we transform the class diagram\, defined by Executable and Translatable \n UML\, to grammars specified by the Grammatical Framework. The grammars are \n then used to generate the desired text. Overall\, the transformation is \n uni-directional\, automatic and an example of a reverse engineering \n translation.\n \n /Executable and Translatable UML - How Difficult Can it Be?/\n Within Model-Driven Architecture there has been substantial research on the \n transformation from Platform-Independent Models (PIM) into Platform-Specifc \n Models\, less so on the transformation from Computationally Independent Models \n (CIM) into PIMs. This publication reflects on the outcomes of letting novice \n software developers transform CIMs specified by UML into PIMs defined in \n Executable and Translatable UML.\n \n /Conclusion/\n The three publications show how model transformations can be used within both \n Language Technology and Software Engineering to tackle the challenges of \n natural language processing and software development.\n \n Full text: http://www.cse.chalmers.se/%7Eburden/pdfs/BurdenLic.pdf [1]\n \n \n [1] http://www.cse.chalmers.se/%7Eburden/pdfs/BurdenLic.pdf END:VEVENT BEGIN:VEVENT UID:calendar:3052:field_eventdate:0:144 SUMMARY:Ten years of BNF Converter DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120620T151500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120620T170000 URL;VALUE=URI:http://www.clt.gu.se/event/2012-06-20/ten-years-bnf-converter LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:A workshop organized by Markus Forsberg and Aarne Ranta\n \n Ten years have gone by since we started working on BNF Converter. We like to \n celebrate this by inviting you to an informal event with some talks and wine. \n Welcome! (No registration necessary.)\n \n Programme:\n \n 15:15-15:20 Welcome (Markus Forsberg and Aarne Ranta)\n 15:20-15:50 Ten years of BNF Converter (Markus Forsberg)\n 15:50-16:20 Implementing Programming Languages - A New Book Using BNF \n Converter (Aarne Ranta)\n 16:20-16:40 BNFC-meta (Jonas Duregård)\n 16:40-17:00 Discussion: BNF Converter from a user's perspective (Bengt \n Nordström)\n 17:00-           Wine+snacks\n \n http://bnfc.digitalgrammars.com/ [1]\n \n http://www.digitalgrammars.com/ipl-book/ [2]\n \n  \n \n \n [1] http://bnfc.digitalgrammars.com/\n [2] http://www.digitalgrammars.com/ipl-book/ END:VEVENT BEGIN:VEVENT UID:calendar:3067:field_eventdate:0:145 SUMMARY:CLT seminar planning DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120830T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120830T111500 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-08-30/clt-seminar-planning LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Planning event for the CLT seminar series Autumn semester. END:VEVENT BEGIN:VEVENT UID:calendar:3096:field_eventdate:0:146 SUMMARY:CLT seminar: Taraka Rama - Reticulation: Quantifying the untreelikeness in \n linguistic phylogenies DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120906T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120906T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-09-06/clt-seminar-taraka-rama LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Since the beginning of historical linguistics\, there has been a strong \n adherence to tree like evolution of languages. Recent experiments show that \n not all linguistic features evolve tree-like. In fact\, networks or mesh like \n diagrams are used to represent relations between languages. Given this \n context\, can we quantify the amount of evolution in a family which deviates \n away from the standard tree structure? The answer is yes. END:VEVENT BEGIN:VEVENT UID:calendar:3093:field_eventdate:0:147 SUMMARY:Final PhD thesis seminar: Katarina Mühlenbock DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120910T150000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120910T180000 URL;VALUE=URI:http://www.clt.gu.se/event/2012-09-10/final-phd-thesis-seminar-katarina-m%C3%BChlenbock LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Katarina Heimann Mühlenbock at the department of Swedish and Språkbanken \n will present her upcoming PhD thesis: "I see what you mean - Assessing \n readability for specific target groups".\n \n Web page [1] from the department calendar.\n \n \n [1] http://www.svenska.gu.se/ViewPage.action?siteNodeId=515670&\;languageId=100000&\;contentId=-1&\;eventId=1778209588 END:VEVENT BEGIN:VEVENT UID:calendar:3097:field_eventdate:0:148 SUMMARY:CLT seminar: Peter Ljunglöf - Practical Parsing of Parallel Multiple \n Context-Free Grammars DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120913T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120913T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-09-13/clt-seminar-peter-ljunglof LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:To be presented at TAG+11: The 11th International Workshop on Tree Adjoining \n Grammars and Related Formalisms\, Paris 26-28 september\n \n Abstract:\n We discuss four previously published parsing algorithms for parallell \n multiple context-free grammar (PMCFG)\, and argue that they are similar to \n each other\, all implementing an Earley-style top-down algorithm. Starting \n from one of these algorithms\, we derive three modifications -- one bottom-up \n and two variants using a left corner filter. An evaluation shows that \n substantial improvements can be made by using the algorithm that performs \n best on a given grammar. The algorithms are implemented in Python and \n released under an open-source licence.\n \n Paper: \n http://www.cse.chalmers.se/~peb/pubs/Ljunglof2012a%20-%20Practical%20Par... \n [1]\n \n \n [1] http://www.cse.chalmers.se/~peb/pubs/Ljunglof2012a%20-%20Practical%20Parsing%20of%20Parallel%20Multiple.pdf END:VEVENT BEGIN:VEVENT UID:calendar:3098:field_eventdate:0:149 SUMMARY:CLT seminar: TBA - TBA DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120927T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20120927T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-09-27/clt-seminar-tba-tba LOCATION:L308\, Lennart Torstenssonsgatan 8 END:VEVENT BEGIN:VEVENT UID:calendar:3099:field_eventdate:0:150 SUMMARY:CLT seminar: TBA - TBA DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121004T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121004T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-10-04/clt-seminar-tba-tba LOCATION:L308\, Lennart Torstenssonsgatan 8 END:VEVENT BEGIN:VEVENT UID:calendar:3057:field_eventdate:0:151 SUMMARY:CLT Workshop 2012 DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121010T090000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121012T150000 URL;VALUE=URI:http://www.clt.gu.se/event/2012-10-10/clt-workshop-2012 LOCATION:Gullmarsstrand DESCRIPTION:The third annual workshop for CLT members will take place at Gullmarsstrand \n [1] 10-12 October. We will leave by bus from Carlandersplatsen/Lundgrensgatan \n (map [2]) at 09.00 on Wednesday morning\, and expect to be back in Gothenburg \n by 3 pm on Friday. Below\, you can find a preliminary schedule and titles for \n the presentations. If you have any questions\, please contact Martin Kaså \n [3].\n \n *Preliminary schedule*\n \n *Wednesday*   09.00-10.45 Travelling 11.00-11.45 General report on CLT \n 12.00-13.00 Lunch 13.00-13.30 Talk: Søren Wichmann 13.30-14.00 Talk: Harald \n Hammarström 14.00-14.30 Talk: Kimmo Koskenniemi 14.30-16.30 Speed dating + \n Coffee (at 15.30) 16.30-17.30 Panel: Conceptual and distributional \n computational semantics 19.30- Dinner     *Thursday*   09.00-09.30 Talk: \n Magnus Merkel 09.30-10.00 Talk: Hrafn Loftsson 10.00-10.30 Talk: Nicoletta \n Calzolari 10.30-11.00 Coffee 11.00-12.00 Future CLT flagships 12.00-13.00 \n Lunch 13.00-14.45 Poster session 1 14.45-15.15 Coffee 15.15-17.00 Poster \n session 2 17.15-19.00 Board meets advisors 19.30- Dinner     *Friday*   \n 09.00-09.30 Talk: Giampiero Salvi 09.30-10.00 Talk: Rickard Domeij \n 10.00-10.30 Talk: Johanna Moore 10.30-11.00 Coffee 11.00-12.00 Panel: LT \n teaching 12.00-13.00 Lunch 13.00-15.00 Travelling\n *Poster session 1*\n \n Ann-Marie Eklund How query annotations may help in providing accurate public \n health\n information Aarne Ranta Demo: Some Techniques for Multilingual and Multimodal \n Query Systems Benjamin Lyngfelt & Markus Forsberg SweCxn - towards a \n comprehensive Swedish constructicon Cajsa Ottesjö & Stina Ericsson LekBot \n Active Play Dana Dannélls Demo: Generating coherent multilingual painting \n descriptions from Web ontologies Elena Volodina\, Lars Borin Lärka - an \n emerging ICALL platform for Swedish Elena Volodina\, Richard Johansson\, Sofie \n Johansson Kokkinakis Ranking corpus examples from "best" down Grégoire \n Détrez Smart Paradigms and the Predictability and Complexity of Inflectional \n Morphology Johan Roxendal Demo: Korp v2.0 Preview John J. Camilleri A \n computational morphology for Maltese\, and using it to build a full-form \n lexicon Karin Friberg Heppin & Maria Toporowska Gronostaj The Rocky Road \n towards a Swedish FrameNet – Creating SweFN Magdalena Siverbo FraCaSBank - \n a bilingual treebank for the FraCaS test suite Sandra Derbring Pratmakaren - \n speaking movie subtitles Staffan Larsson\, Robin Cooper\, Simon Dobnik Toward \n perceptually grounded formal semantics David Junger\, Torbjörn Lager\, Johan \n Roxendal Poster och demo: Preview of the Dialog Web Lab\n *Poster session 2*\n \n Dimitrios Kokkinakis Medication Event Extraction Using Frame Semantics ... a \n pilot investigation Dimitrios Kokkinakis\, Markus Forsberg\, Frida Smith\, \n Joakim Öhlén Analysis of written information material to cancer patients \n Elizabeth Coppock Scalarity in Discourse Gerlof Bouma\, Malin Ahlberg\, Yvonne \n Adesam /bokstaffua\, bokstaffwa\, bokstafwa\, bokstaua\, bokstawa/... Towards \n lexical link-up for a corpus of Old Swedish Judy Ribeck & Sofie Johansson \n Kokkinakis A Swedish Academic Word List Karin Cavallin Automatic extraction \n of potential examples of semantic change using lexical sets Karin Friberg \n Heppin & Anni Järvelin Towards Improving Search Results for Medical Experts \n and Laypersons Krasimir Angelov Demo: The Penn Treebank and Statistical\n and Robust Parsing in GF Peter Ljunglöf Editing syntax trees on the surface \n Richard Johansson Semantic role labeling with the Swedish FrameNet Stina \n Ericsson Information structural leaks - a CLT small project Taraka Rama \n Typological distances and vector similarity measures Thomas Hallgren Demo: \n cloud.grammaticalframework.org\n  \n \n \n [1] http://gullmarsstrand.se/o.o.i.s/250\n [2] http://g.co/maps/ep9ve\n [3] mailto:martin.kasa@gu.se?subject=CLT%20Gullmarsstrand END:VEVENT BEGIN:VEVENT UID:calendar:3094:field_eventdate:0:152 SUMMARY:Språkbanken workshop DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121017T130000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121017T170000 URL;VALUE=URI:http://www.clt.gu.se/event/2012-10-17/spraakbanken-workshop LOCATION:L100\, Lennart Torstenssonsgatan 8 DESCRIPTION:http://spraakbanken.gu.se/swe/Om%20oss/hoestworkshop [1]\n \n \n [1] http://spraakbanken.gu.se/swe/Om%20oss/hoestworkshop END:VEVENT BEGIN:VEVENT UID:calendar:3100:field_eventdate:0:153 SUMMARY:CLT seminar: Gunnar Eriksson - It takes two to tango — simple features for \n profiling chat forum participants DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121018T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121018T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-10-18/clt-seminar-gunnar-eriksson LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:My presentation describes experiments that were performed in the \n participation of  “Uncovering Plagiarism\, Authorship and Social Software \n Misuse” Lab (PAN) of the 2012 CLEF conference. One of the tracks in the \n recurring PAN workshop is different aspects of authorship attribution. This \n year\, one of the tasks was to identify one specific class of authors in \n written interactive on-line chat forum logs\, namely authors who attempt to \n convince other participants to provide sexual favours.\n \n We find that features which model interaction and conversational behaviour \n contribute well to identifying sexual grooming behaviour in chat and forum \n text. Together with the obviously useful lexical features — which we find \n are more valuable if separated by who generates them — we achieve \n successful results in identifying behavioural patterns which may characterise \n sexual grooming.\n \n Gunnar Eriksson SICS [1]\n \n \n [1] http://www.sics.se/people/guer END:VEVENT BEGIN:VEVENT UID:calendar:3065:field_eventdate:0:154 SUMMARY:Workshop on Language\, Action and Perception DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121025T000000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121025T000000 URL;VALUE=URI:http://www.clt.gu.se/event/2012-10-25/workshop-language-action-and-perception LOCATION:Lund University DESCRIPTION:Simon Dobnik is organizing a workshop during this year's SLTC in Lund.\n \n Web page: http://clt.gu.se/dialogue-technology-lab/sltc2012-apl [1]\n \n SLTC: http://nlp.cs.lth.se/events/sltc_2012/ [2]\n \n \n [1] http://clt.gu.se/dialogue-technology-lab/sltc2012-apl\n [2] http://nlp.cs.lth.se/events/sltc_2012/ END:VEVENT BEGIN:VEVENT UID:calendar:3118:field_eventdate:0:155 SUMMARY:Workshop on NLP for computer-assisted language learning DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121025T000000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121025T000000 URL;VALUE=URI:http://www.clt.gu.se/event/2012-10-25/workshop-nlp-computer-assisted-language-learning LOCATION:Lund University DESCRIPTION:Lars Borin and Elena Volodina are organizing a workshop during this year's \n SLTC in Lund.\n \n Web page: NLP for computer assisted language learning [1]\n \n SLTC: http://nlp.cs.lth.se/events/sltc_2012/ [2]\n \n \n [1] http://spraakbanken.gu.se/eng/Research/A%20System%20Architecture%20for%20ICALL/SLTC%20workshop%202012/nlp-computer-assisted-language-learning\n [2] http://nlp.cs.lth.se/events/sltc_2012/ END:VEVENT BEGIN:VEVENT UID:calendar:3109:field_eventdate:0:156 SUMMARY:Fakultetsseminarium: Digital humaniora - metodutveckling och möjligheter DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121031T100000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121031T151500 URL;VALUE=URI:http://www.clt.gu.se/event/2012-10-31/fakultetsseminarium-digital-humaniora-metodutveckling-och-m%C3%B6jligheter LOCATION:Lilla hörsalen\, Renströmsgatan 6 DESCRIPTION:The Text Technology Lab will arrange a Faculty seminar on "Digital \n Humanities" togehther wit the department of literature\, history of ideas and \n religion.\n \n More information here: (in Swedish). END:VEVENT BEGIN:VEVENT UID:calendar:3078:field_eventdate:0:157 SUMMARY:CLT seminar: John Nerbonne - A Structuralist's Dream (Realized) DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121101T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121101T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-11-01/clt-seminar-john-nerbonne LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:John Nerbonne\, reporting on joint work with Martijn Wieling and Eliza \n Margaretha\n CLCG\, Groningen & FRIAS\, Freiburg\n \n Structuralists famously observed that language is "un systême oû tout se \n tient" (Meillet\, 1903:407)\, emphasizing that the system of relations of \n linguistic units was more important than their concrete content. The modern \n study of speech has not vindicated the structuralists on this point\, as the \n contemporary emphasis lies on instrumental studies of pronunciation and \n auditory perception.  Instrumental phonetics and laboratory phonology are \n leaders in this direction.\n \n But modern computational linguistics and corpus linguistics emphasizes that \n there is a wealth of information in distributions.  This study attempts to \n derive content from relations\, in particular phonetic (acoustic) content from \n the distribution of alternative pronunciations used in different linguistic \n varieties. It proceeds from data documenting language variation\, examining \n six dialect atlases each containing the phonetic transcriptions of the same \n sets of words at hundreds of sites. We collect the correspondences via an \n alignment procedure\, and then apply an information-theoretic measure\, \n pointwise mutual information\, assigning smaller segment distances to segments \n which frequently correspond. We iterate alignment and information-theoretic \n distance assignment until both stabilize and we evaluate the quality of the \n phonetic distances obtained by comparing them to acoustic vowel distances. \n For all dialect datasets (Dutch\, German\, Gabon Bantu\, U.S. English\, Tuscan \n and Bulgarian) we find strong signicant correlations between the induced \n phonetic distances and the acoustic distances\, illustrating the usefulness of \n the method in deriving valid phonetic distances from dialectal \n pronunciations.\n \n Martijn Wieling\, Eliza Margarethe and John Nerbonne (2012) Inducing a Measure \n of Phonetic Similarity from Pronunciation Variation. Journal of Phonetics \n 40(2): 307-314. DOI: http://dx.doi.org/10.1016/j.wocn.2011.12.004 [1]\n \n \n [1] http://dx.doi.org/10.1016/j.wocn.2011.12.004 END:VEVENT BEGIN:VEVENT UID:calendar:3101:field_eventdate:0:158 SUMMARY:CLT seminar: Bolette Pedersen - Exploiting conventional dictionaries in \n language technology resources DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121108T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121108T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-11-08/clt-seminar-bolette-pedersen LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:The talk will address the compilation of computational lexical-semantic \n resources from conventional dictionaries and thesauri and vice versa. Seen in \n a historical perspective\, the lack of synergy between these two kinds of \n resources is surprising.\n \n Containing an enormous amount of lexical and semantic knowledge\, dictionaries \n are a likely source of information for use in computational semantic lexicons \n and semantic knowledge bases. Where several studies have concluded that the \n results of reuse experiments are disappointing mainly due to inconsistent \n dictionaries\, other recent experiments are\, however\, much more promising. \n This new interest in reuse partly stems from the fact that dictionaries (be \n they printed or electronic) are changing and improving rapidly together with \n modern corpus and compilation facilities and are therefore becoming more \n attractive as background resources for computational use.\n \n I will discuss this topic on the basis on our own experiments with the two \n Danish dictionaries Den Danske Ordbog and the Danish Thesaurus\, of which the \n latter is currently under construction. I will show how we – by keeping \n track of the sense inventory across all three resources (DDO\, DT and the \n Danish wordnet\, DanNet) – can transfer information from one resource to \n another\, and how we are currently extending DanNet with thematic and \n syntagmatic information from Dansk the Danish Thesaurus.\n \n Bolette Sandford Pedersen [1] (Center for Sprogteknologi\, Copenhagen)\n \n \n [1] http://cst.dk/bolette/my.html END:VEVENT BEGIN:VEVENT UID:calendar:3102:field_eventdate:0:159 SUMMARY:CLT seminar: Markus Forsberg - Tracking lexical change in Swedish DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121115T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121115T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-11-15/clt-seminar-markus-forsberg-tracking-lexical-change-swedish LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:In this seminar we will discuss the task of tracking lexical change in \n written Swedish. The focus will be on two related research questions: (1) how \n do we track the changes of a lexical unit in Swedish texts during a specified \n time period\; and (2) how do we detect interesting lexical changes in Swedish \n texts\, again\, during a specified time period. For both questions we are \n interested in what may be considered as satisfying proofs for statements \n about lexical change. Question (1) will be related to what is known as \n Culturomics\, and (2) to the Swedish Language Council's work on compiling \n Swedish new words lists.\n   END:VEVENT BEGIN:VEVENT UID:calendar:3103:field_eventdate:0:160 SUMMARY:CLT seminar: Victoria Rosén and Paul Meurer - INESS: an infrastructure for \n treebanking DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121122T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121122T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-11-22/clt-seminar-victoria-ros%C3%A9n-and-paul-meurer-iness-infrastructure-treebanking LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:The INESS project is building an infrastructure for treebanks\, a virtual \n laboratory for treebanking with rich functionality which can be used via an \n ordinary web browser interface.  The infrastructure is in particular geared \n towards dynamic parsing\, disambiguation and advanced search of LFG treebanks\, \n and an important task for INESS is the construction of a large LFG treebank \n for Norwegian. INESS will also provide for online search and processing of \n dependency and phrase structure treebanks created by others. In this talk we \n will give a basic overview of the project and present some of the newer \n developments we have been working on recently.\n \n \n Victoria Rosén\, University of Bergen [1] and Uni Research [2]\n Paul Meurer\,\n Uni Research\n \n \n [1] http://www.uib.no/persons/Victoria.Rosen/\n [2] http://digital.uni.no/folk/victoria-rosen END:VEVENT BEGIN:VEVENT UID:calendar:3138:field_eventdate:0:161 SUMMARY:Language in the digital age DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121123T090000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121123T153000 URL;VALUE=URI:http://www.clt.gu.se/event/2012-11-23/language-digital-age LOCATION:Stacken\, David Bagares gata 17\, Stockholm\, Sweden DESCRIPTION: \n \n Welcome to the workshop "Language in the digital age"!\n \n When: Friday 23rd November 2012\, ca. 9:00–15:30\n Where: Stacken\, David Bagares gata 17\, Stockholm\, Sweden\n \n Registration: By 15th November\, via this webform:  \n www.spraakbanken.gu.se/eng/meta-nord-seminarium [1]\n \n Other: Participation is free\, but registration is necessary since coffee and \n lunch will be served.  NB the workshop will mostly be in Swedish..\n \n Contact: Martha D. Brandt (martha.brandt@svenska.gu.se [2]) \n \n  \n \n Preliminary program:\n \n * The Swedish Language in the Digital Age - one of META-NET's White Paper\n Series about the status of   each European language with respect to\n language technology\, Lars Borin and Martha D. Brandt\, Språkbanken (The\n Language Bank) in Gothenburg.  Read more about the white paper here: \n http://www.meta-net.eu/whitepapers/volumes/swedish [3] /\n http://spraakbanken.gu.se/eng [4] (Språkbanken in English)\n * Media for all – accessibility to television and the internet\, Mia\n Ahlgren\, Handikappförbunden (The Handicap Association).  Read more in a\n Nordic report that Mia co-authored (in Swedish):\n www.nordicwelfare.org/News/Medier-for-alla/ [5]\n * Infrastructure for languages in Sweden - proposal for a national\n infrastructure for the digital community\, Rickard Domeij\, Språkrådet\n (The Language Council).  Read more about the proposal (in Swedish):\n www.sprakradet.se/nationell-spr%C3%A5kdatabank [6]\n * Open data creates new opportunities for innovation\, Erik Borälv\,\n Vinnova.  Read more about open data (in Swedish): www.vinnova.se/opendata\n [7] / http://www.vinnova.se/en/ [8] (Vinnova in English)\n * Thoughts of an entrepreneur in the field of Language Technology on\n language resources and language infrastructure\, Jussi Karlgren\, Gavagai. \n Read more about Gavagai (in Swedish):\n http://internetworld.idg.se/2.1006/1.463642/arets-entreprenorer-plats-2-gavagai\n [9] / http://www.gavagai.se/index.php [10] (Gavagai in English)\n \n  \n \n This workshop is held as part of the strategic effort of the European \n research network META-NET to promote the development of a multilingual \n European information society.   The workshop's focus is on the situation in \n Sweden and the description and study by META-NET in the Language White Paper \n "The Swedish Language in the Digital Age" (\n ). The purpose of the workshop is to create a venue to highlight and discuss \n the situation from different perspectives.  This workshop is organized \n jointly by Språkbanken (The Language Bank) in Gothenburg and Språkrådet \n (The Language Council) and is made possible with economic support from the EU \n commission to the META-NORD project (\n [11]\; the ICT PSP program\, contract no. 270899). \n \n \n [1] http://www.spraakbanken.gu.se/eng/meta-nord-seminarium\n [2] mailto:martha.brandt@svenska.gu.se\n [3] http://www.meta-net.eu/whitepapers/volumes/swedish\n [4] http://spraakbanken.gu.se/eng\n [5] http://www.nordicwelfare.org/News/Medier-for-alla/\n [6] http://www.sprakradet.se/nationell-spr%C3%A5kdatabank\n [7] http://www.vinnova.se/opendata\n [8] http://www.vinnova.se/en/\n [9] http://internetworld.idg.se/2.1006/1.463642/arets-entreprenorer-plats-2-gavagai\n [10] http://www.gavagai.se/index.php\n [11] http://www.meta-net.eu/whitepapers/volumes/swedish END:VEVENT BEGIN:VEVENT UID:calendar:3104:field_eventdate:0:162 SUMMARY:CLT seminar: Preparing for COLING DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121129T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121129T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-11-29/clt-seminar-preparing-coling LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:1. A best-first anagram hashing filter for approximate string matching with \n generalized edit distance\, by Malin Ahlberg and Gerlof Bouma\n \n 2. Same grammar\, diverging lexicons: the case of Hindi/Urdu\, by K.V. S. \n Prasad and Shafqat Virk\n \n 3. Grammatical Framework: Formalizing the Grammars of the World\, by Aarne \n Ranta (presented by K. V. S. Prasad) END:VEVENT BEGIN:VEVENT UID:calendar:3105:field_eventdate:0:163 SUMMARY:CLT seminar: Lars Hellan - A cluster of digital applications based on a typed \n feature structure architecture DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121206T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121206T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-12-06/clt-seminar-lars-hellan-cluster-digital-applications-based-typed-feature-structure-architecture LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:The talk will present a small cluster of digital applications held together \n with a concise linguistic architecture based on a typed feature structure \n architecture (essentially: HPSG). The applications are:\n \n 1) two computational grammars based on the LKB platform\, one for Norwegian\n (called 'Norsource'\,\n http://typecraft.org/tc2wiki/Norwegian_HPSG_grammar_NorSource [1] ) and\n one being in a sense 'universal'\, called 'TypeGram'\;\n 2) an e-learning tool based on Norsource\, viz. a 'grammar sparrer'\n (http://typecraft.org/tc2wiki/A_Norwegian_Grammar_Sparrer [2])\;\n 3) a demo of a valence database for Norwegian\, derived from Norsource\n (http://regdili.idi.ntnu.no:8080/vpbwebdemo/parse [3])\;\n 4) a demo of 'Grammar induction from flat annotation'\, based on the\n annotation tool TypeCraft (http://typecraft.org/tc2wiki/Main_Page [4])\n and the 'universal' grammar mentioned.\n \n The talk will describe the applications\, and the linguistic architecture \n behind them\, in particular its encoding of valence information and situation \n types. The talk will then present some ideas of extending these applications\, \n for instance in the direction of a cross-linguistic European valence \n repository.\n \n \n [1] http://typecraft.org/tc2wiki/Norwegian_HPSG_grammar_NorSource\n [2] http://typecraft.org/tc2wiki/A_Norwegian_Grammar_Sparrer\n [3] http://regdili.idi.ntnu.no:8080/vpbwebdemo/parse\n [4] http://typecraft.org/tc2wiki/Main_Page END:VEVENT BEGIN:VEVENT UID:calendar:3161:field_eventdate:0:164 SUMMARY:Talk by Alessandro Moschitti DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121207T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121207T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2012-12-07/talk-alessandro-moschitti LOCATION:L307\, Lennart Torstenssonsgatan 8 DESCRIPTION:Bonus CLT talk by Alessandro Moschitti: Shallow Semantic Models for \n Automatically Answering Complex Jeopardy! Questions END:VEVENT BEGIN:VEVENT UID:calendar:3140:field_eventdate:0:165 SUMMARY:Master Thesis Defence: Christian Ramstedt -- A Swedish FraCas DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121210T090000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121210T100000 URL;VALUE=URI:http://www.clt.gu.se/event/2012-12-10/master-thesis-defence-christian-ramstedt-swedish-fracas LOCATION:T340\, Olof Wijksgatan 6 DESCRIPTION:Defence of Christian Ramsted's master thesis for the Master Programme in \n Language Technology:\n \n A Swedish FraCas -- A Swedish-English translation of quantifiers from the \n FraCas test suite from a semantical perspective\n \n Supervisor: Robin Cooper\n Opponent: Parna Seyedin\n Examinater: Torbjörn Lager\n \n Abstract:\n \n The purpose of this thesis is to give a brief summary of the FraCas test \n suite and to make a translation of its Quantifier chapter from English to \n Swedish and to do this as accurate as possible. To do this\, the semantical \n differences between English and Swedish quantifiers will be noted and \n compared. The structure of the FraCas test suite will be discussed and also \n the structure of the xml-version of the test suite. The main goal of the \n thesis is to make a semantical translation of English and Swedish quantifiers \n as accurate as possible in the case of a future translation of the FraCas \n test suite. END:VEVENT BEGIN:VEVENT UID:calendar:3141:field_eventdate:0:166 SUMMARY:Master Thesis Defence: Tam Luong -- Edutainment Games DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121210T100000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20121210T110000 URL;VALUE=URI:http://www.clt.gu.se/event/2012-12-10/master-thesis-defence-tam-luong-edutainment-games LOCATION:T340\, Olof Wijksgatan 6 DESCRIPTION:Defence of Tam Loung's one year master thesis for the Master Programme in \n Language Technology:\n \n Edutainment Games -- Using Speech Recognition in a self-learning game for \n primary school students and an entertaining game for people at all ages.\n \n Supervisor: Torbjörn Lager\n Opponent: Bamshad Lotfabadi\n Examinater: Robin Cooper\n \n Abstract:\n \n The purpose of this study was to estimate how Voice Technology worked\n in educational and entertaining (edutaining) games. In order to find\n the answer\, the author designed two edutaining games using Speech\n Technology and game development. One educational game was for primary\n school students with the range of age from 7 to 12 while the other\n entertaining game was for people at all ages.\n \n The qualitative research methodology was applied in this study. The\n research data consisted of participant observation\, semi-structured\n interviews and structured interviews. Participants of the first game\n were 37 students from 4 primary schools in Gothenburg City\, Sweden.\n Participants of the other game were 20 adults from different cities in\n Sweden and other countries.\n \n The result of the evaluation revealed that Speech Technology worked\n well in both educational and entertaining games. The findings\n indicated the technology had implications for 3 main matters: the\n content of the games\; teaching methods and hands- free and touch tone\n games. It was discovered that the technology stimulated participants’\n interest in playing the voice games which were different from common\n video games. 100% of the participants and students would consider\n playing both games again in the future. Apparently the voice games\n helped increase engagement of students when they explored a new\n potential technology in learning. The result also showed that the\n technology worked better when participants used telephone or mobile\n phone than computer via Skype. Less technical errors occurred when\n they played the games with phones. When asked to compare the two\n games\, the students thought both games provided benefits in variant\n ways. The educational game was helpful in the self-learning of\n different school subjects while the entertaining game had a better\n quality in aspects of fun element\, sound techniques and language\n clarity.\n \n The author recommends voice game programmers to implement higher\n advanced technology such as Browse-Accessible\, an application that\n recognizes speech and save to the database. It is also recommended\n that programmers collaborate with teachers for pedagogical advices in\n the process of collecting appropriate material for educational games.\n Experiences and tips derived from the evaluation and implementation of\n this study are also shared with programmers to facilitate the work of\n their similar applications in the future. END:VEVENT BEGIN:VEVENT UID:calendar:3209:field_eventdate:0:167 SUMMARY:CLT seminar: Marco Kuhlmann - The Divergence of Mildly Context-Sensitive \n Grammar Formalisms DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130117T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130117T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-01-17/clt-seminar-marco-kuhlmann-divergence-mildly-context-sensitive-grammar-formalisms LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Marco Kuhlmann [1] (Uppsala University) will speak on "The Divergence of \n Mildly Context-Sensitive Grammar Formalisms".\n \n Abstract:\n \n One of the key results in the literature on mathematical models of natural \n language is the weak equivalence between Tree Adjoining Grammar (TAG) and \n Combinatory Categorial Grammar (CCG) [1]. However\, the version of CCG for \n which this equivalence was established differs significantly from the \n formalism that is being used today. In particular\, old-style CCG allows to \n restrict and even ban the application of combinatory rules on a per grammar \n basis\, while modern CCG assumes a universal set of rules\, isolating all \n cross-linguistic variation in the lexicon. In this talk I will discuss the \n relevance of grammar-specific rule restrictions for the expressive power of \n CCG from a number of different perspectives\, focusing on CCG’s ability to \n give a lexicalized account of word order. This discussion will culminate in a \n formal result stating that without grammar-specific rule restrictions\, the \n equivalence between TAG and CCG breaks down. This raises important questions \n about the descriptive adequacy of grammar formalisms.\n \n [1] K. Vijay-Shanker and David J. Weir. The Equivalence of Four Extensions of \n Context-Free Grammars [2]. Mathematical Systems Theory\, 27(6):511–546\, \n 1994.\n \n  \n \n \n [1] http://stp.ling.uu.se/~kuhlmann/\n [2] http://www.springerlink.com/index/M8210866375PP50V.pdf END:VEVENT BEGIN:VEVENT UID:calendar:3223:field_eventdate:0:168 SUMMARY:Web Science seminar: Hans Akkermans - On the Globalization of the World Wide \n Web DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130122T131500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130122T150000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-01-22/web-science-seminar-hans-akkermans LOCATION:T340\, Olof Wijksgatan 6 DESCRIPTION:The World Wide Web is the biggest global social network in human history to \n date. It is an important empowering instrument for information exchange and \n knowledge sharing. Yet\, two out of three people on this planet still do not \n have access to the Web. A truly global Web requires mechanisms able to cater \n for the need for multi-lingual and multi-modal access to the Web (not only \n text\, but also speech\,) as well as its coupling to other mass communication \n media such as mobile phones and radio. In this regard\, I will discuss Web \n Science and Technology initiatives and experience\, based on our Web projects \n in rural Africa that we are currently undertaking such as EU-FP7 VOICES and \n IPI-Foroba Blon.\n \n ------------------------------------------------------------------------------\n Prof. Dr. J.M. (Hans) Akkermans is a full professor of Business Informatics \n at VU University Amsterdam (VUA). He is the founder\, first Director \n (2007-2011) and now Chair of the various boards of The Network Institute \n (www.thenetworkinstitute.eu [1])\, a new interdisciplinary and multi-faculty \n research institute established at VU Amsterdam and supported by its Executive \n Board (CvB) as of initio 2008.\n \n The institute researches the networked world in all its facets \n (technological\, social\, economical)\, and hosts 250 plus researchers from \n different disciplines\, including informatics and computer science\, \n mathematics\, economics and business administration (marketing\, MIS\, knowledge \n management)\, linguistics and social sciences (communication science\, \n organization science\, cultural studies of management).\n \n He holds cum laude Master and PhD degrees in theoretical physics from the \n University of Groningen. Recently\, as part of the Web Science initiatives at \n VUA\, he initiated and is leading a new international collaboration between \n VUA\, Tim Berners-Lee's World Wide Web Foundation and partners in Africa\, \n called W4RA\, the Web alliance for Regreening in Africa (see \n www.webfoundation.org [2] \, www.w4ra.org [3]\, w4ra.few.vu.nl [4]). The aim is \n to expand global access to the Web and create new mechanisms for open \n information and knowledge sharing.\n \n \n [1] http://www.thenetworkinstitute.eu\n [2] http://www.webfoundation.org\n [3] http://www.w4ra.org\n [4] http://w4ra.few.vu.nl END:VEVENT BEGIN:VEVENT UID:calendar:3228:field_eventdate:0:169 SUMMARY:CLT seminar planning DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130124T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130124T111500 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-01-24/clt-seminar-planning LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Planning event for the regular CLT seminar series END:VEVENT BEGIN:VEVENT UID:calendar:3210:field_eventdate:0:170 SUMMARY:CLT seminar: Nina Tahmasebi - On automatic language change detection for \n finding and interpreting content in long-term archives DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130131T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130131T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-01-31/clt-seminar-nina-tahmasebi LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:High impact events\, political changes and new technologies are reflected in \n our language and lead to constant change of terms\, expressions and names. Not \n knowing about these changes can severely limit our possibilities to find and \n interpret information from the past.\n \n In this talk we will present work undertaken in the LiWA and ARCOMEM projects \n on classifying and automatically finding language changes.  Our \n classification is based on our motivation to *find* and *interpret* documents \n found in long-term archives. We will highlight the characteristics of each \n class\, relate the classes and present application scenarios.\n \n For two classes\, namely word sense changes and named entity changes\; we go in \n depth and propose unsupervised algorithms as well as present results. We \n showcase our algorithms on the New York Times Annotated Corpus (1986-2007) as \n well as The Times Archive (1785-1985) and show that the proposed algorithms \n are able to capture named entity changes and move towards automatic detection \n of word sense changes.\n \n Nina Tahmasebi [1] is at the L3S Research Center [2]\, Leibniz Universität \n Hannover\n \n \n [1] http://www.tahmasebi.se/\n [2] http://www.l3s.de/ END:VEVENT BEGIN:VEVENT UID:calendar:3165:field_eventdate:0:171 SUMMARY:Thesis defence: Dana Dannélls - Multilingual text generation from structured \n formal representations DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130205T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130205T130000 URL;VALUE=URI:http://www.clt.gu.se/event/2013-02-05/thesis-defence-dana-dannells LOCATION:Lilla hörsalen\, Humanisten\, Renströmsgatan 6 DESCRIPTION:Dana Dannélls\, GSLT PhD student in natural language processing at the \n Department of Swedish/Språkbanken\, is defending her thesis "Multilingual \n text generation from structured formal representations".\n \n Opponent: Professor Michael Elhadad\, Ben Gurion University of the Negev\n \n Link to the thesis in GUPEA: hdl.handle.net/2077/31856 [1]\n \n \n [1] http://hdl.handle.net/2077/31856 END:VEVENT BEGIN:VEVENT UID:calendar:3229:field_eventdate:0:172 SUMMARY:CLT seminar: He Tan - Semantic Role Labeling for Biomedical Events DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130207T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130207T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-02-07/clt-seminar-he-tan LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Semantic Role Labeling (SRL) plays a key role in many text mining (TM) \n applications. The development of SRL systems for the biomedical domain is \n frustrated by the lack of large domain specific corpora. In the project we \n proposed a method for building corpus that are labeled with semantic roles \n for the domain of biomedicine. The method is based on the theory of frame \n semantics\, and uses domain knowledge provided by ontologies. By using the \n method\, we have built a corpus for transport events strictly following the \n domain knowledge provided by Gene Ontology. This demonstrates that \n ontologies\, as a formal representation of domain knowledge\, can guide us and \n ease all the tasks in building this kind of corpus. Furthermore\, ontological \n domain knowledge leads to well-defined semantics exposed on the corpus\, which \n will be valuable in TM applications. We have experimented with a \n word-chunking approach for identifying semantic roles of biomedical \n predicates describing transport events using the corpus. We trained a \n first-order conditional random field for chunking applications with the \n traditional role labeling features and also domain-specific features. The \n results show that the system performance varies between different roles and \n the performance was not improved for all roles by introducing domain specific \n features.\n \n He Tan [1] is a postdoctoral fellow (forskarassistent) at the Division for \n Databases and Information Techniques (ADIT) and a member of the Laboratory \n for Intelligent Information Systems [2]\, Linköping university.\n \n \n [1] http://www.ida.liu.se/~hetan/\n [2] http://www.ida.liu.se/labs/iislab END:VEVENT BEGIN:VEVENT UID:calendar:3239:field_eventdate:0:173 SUMMARY:[CANCELLED] CLT seminar: Stina Ericsson and Cajsa Ottesjö - LekBot part 2: \n robot development and genuine play DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130214T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130214T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-02-14/clt-seminar-ericsson-ottesjo LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:The LekBot project\, part 1 [1]\, was a collaboration between DART [2]\, \n Talkamatic [3] and GU in 2010-2011. The project developed a talking and \n playing robot for children with communicative disabilities\, with the aim of \n providing a toy that is easy and fun to use\, and that provides opportunities \n for genuine play in the sense of play that is spontaneous\, independent\, on \n equal terms\, etc. Three test groups participated in the project\, with each \n test group consisting of a child with cerebral palsy\, a peer and pre-school \n staff. All groups were recorded in interactions with various versions of the \n system. The LekBot project\, part 2\, started in 2012\, and is a collaboration \n between GU and DART. The focus now is on the analysis of the recorded \n interactions. In the talk we will present current results of the analysis\, \n and discuss implications for further development of the LekBot system.\n \n \n [1] http://www.clt.gu.se/finishedproject/lekbot\n [2] http://www.dart-gbg.org/index2.html\n [3] http://www.talkamatic.se/ END:VEVENT BEGIN:VEVENT UID:calendar:3240:field_eventdate:0:174 SUMMARY:CLT seminar: Shalom Lappin - Developing an Enriched Statistical Language \n Model of Grammaticality DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130221T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130221T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-02-21/clt-seminar-shalom-lappin LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:The question of whether it is possible to characterise grammatical knowledge \n in probabilistic terms is central to determining the relationship of \n linguistic representation to other cognitive domains. We present a \n statistical model of grammaticality which maps the probabilities of a \n statistical model for sentences in parts of the British National Corpus (BNC) \n into grammaticality scores\, using various functions of the parameters of the \n model. We test this approach with a classifier on test sets containing \n different levels of syntactic infelicity. With appropriate tuning\, the \n classifiers achieve encouraging levels of accuracy. These experiments suggest \n that it may be possible to characterise grammaticality judgements in \n probabilistic terms using an enriched language model.\n \n Shalom Lappin [1]\, King's College London\n \n (Joint work with Alexander Clark and Gianluca Giorgolo)\n \n \n [1] https://www.kcl.ac.uk/artshums/depts/philosophy/people/staff/academic/lappin/index.aspx END:VEVENT BEGIN:VEVENT UID:calendar:3241:field_eventdate:0:175 SUMMARY:CLT seminar: John McCrae - Modelling the ontology-lexicon interface with \n lemon DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130228T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130228T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-02-28/clt-seminar-john-mccrae LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Ontologies have become a popular method for modelling knowledge about \n specific domains such as bio-medicine as well as general domain such as in \n the DBpedia ontology. However\, current models for ontologies contain little \n information about the usage of such labels in natural language and are \n difficult to apply in a multilingual setting. In order to cover this gap\, we \n propose a model called /lemon/ (Lexicon Model for Ontologies)\, which enables \n existing ontologies to be extended with an independent lexical layer.\n \n /lemon/ allows ontologies to be connected with lexical\, syntactic\, \n terminological and morphological description of words and terms and in \n particular it describes how lexical entries are mapped to ontological \n predicates. As /lemon/ is based on the linked data model RDF\, /lemon/ is also \n ideally suited to representing machine-readable dictionaries on the Web and \n linking them with existing semantic and lexical resources. I will describe \n the usage of /lemon/ in leveraging existing resources as part of the Lexical \n Linked Open Data Cloud and the challenges presented\, as well as recent \n developments in the model\, that have occurred under the W3C Community Group \n on Ontology Lexica.\n \n Finally\, I will briefly describe the applications of /lemon/ for NLP\, in \n particular in a question answering system developed in Bielefeld that uses \n /lemon/ alongside the Grammatical Framework (GF) to answer question by means \n of queries over the linked data cloud\n \n More about John McCrae [1] (post-doctoral researcher at the University of \n Bielefeld [2])\n \n More about /lemon/ [3]\n \n \n [1] http://john.mccr.ae/\n [2] http://www.uni-bielefeld.de/\n [3] http://lemon-model.net/ END:VEVENT BEGIN:VEVENT UID:calendar:3283:field_eventdate:0:176 SUMMARY:CLT seminar: Valia Kordoni - Robust Automated Grammar Engineering DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130305T133000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130305T150000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-03-05/clt-seminar-valia-kordoni LOCATION:TBA DESCRIPTION:In this talk I deal with automated acquisition of linguistic knowledge as a \n means of enhancing robustness of lexicalised grammars for real life \n applications.\n \n I focus on Multiword Expressions (henceforward MWEs). Specifically\, in the \n first part of the talk I am taking a closer look at the linguistic properties \n of MWEs\, in particular\, their lexical\, syntactic\, as well as semantic \n characteristics.\n \n With the observations about the linguistic properties of MWEs at hand\, I turn \n in the second part of the talk to methods for the automated acquisition of \n these properties for robust grammar engineering and parsing. To this effect\, \n I first investigate the hypothesis that MWEs can be detected by the distinct \n statistical properties of their component words\, regardless of their type\, \n comparing various statistical measures\, a procedure which leads to extremely \n interesting conclusions. I then investigate the influence of the size and \n quality of different corpora\, using the BNC and the Web search engines Google \n and Yahoo. I conclude that\, in terms of language usage\, web generated corpora \n are fairly similar to more carefully built corpora\, like the BNC\, indicating \n that the lack of control and balance of these corpora are probably \n compensated by their size.\n \n Then\, I show a qualitative evaluation of the results of automatically adding \n extracted MWEs to existing linguistic resources. To this effect\, I first \n discuss two main approaches commonly employed in NLP for treating MWEs: the \n words-with-spaces approach which models an MWE as a single lexical entry and \n it can adequately capture fixed MWEs like "by and large"\, and compositional \n approaches which treat MWEs by general and compositional methods of \n linguistic analysis. On this basis\, I argue that the process of the automatic \n addition of extracted MWEs to existing linguistic resources improves \n qualitatively\, if a more compositional approach to grammar/lexicon automated \n extension is adopted.\n \n Finally\, I propose that the methods developed for the acquisition of \n linguistic knowledge in the case of English MWEs can be tuned to enhance \n robustness of parsing with lexicalised grammars for languages with richer \n morphology and freer word order\, as is the case of German.\n \n Valia Kordoni is at Humboldt University\, Berlin and at Saarland University END:VEVENT BEGIN:VEVENT UID:calendar:3243:field_eventdate:0:177 SUMMARY:CLT seminar: Elizabeth Coppock - Computing Implicatures with the Question \n Under Discussion DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130314T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130314T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-03-14/clt-seminar-elizabeth-coppock LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:In this talk\, I will present two versions of a system for computing Gricean \n implicatures based on the Question Under Discussion\, first in classical \n possible worlds semantics (sufficient for most purposes) and then in \n inquisitive semantics (for fancier tricks).\n \n With the classical version\, we can compute that\, for example\, "She speaks \n English" implies "She does not speak Swedish" in a context where the question \n is what languages she speaks and Swedish is one of the relevant languages (a \n 'scalar' or 'exhaustivity' implicature). The system is based on Stalnaker's \n (1978) notion of common ground and Groenendijk and Stokhof's (1984) semantics \n and pragmatics of questions.\n \n In the second part of the talk\, I will combine these ideas with inquisitive \n semantics (Groenendijk & Roelofsen 2009)\, which makes it possible to \n distinguish between "She speaks English"\, which can imply that she does not \n speak Swedish\, and "She speaks at least English"\, which does not. END:VEVENT BEGIN:VEVENT UID:calendar:3296:field_eventdate:0:178 SUMMARY:Flerordsworkshop DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130319T131500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130319T173000 URL;VALUE=URI:http://www.clt.gu.se/event/2013-03-19/flerordsworkshop LOCATION:K332\, Lennart Torstenssonsgatan 8 DESCRIPTION:Flerordsenheter i språkteknologi\, språkvetenskap och språkinlärning.\n \n Anmälan: (ej bindande) till yvonne.adesam (a) gu.se senast torsdag 14 mars. \n Meddela ev. specialkost för fika (vegan\, laktos\, gluten etc.)\n \n Mer information: http://spraakbanken.gu.se/swe/flerordsworkshop [1]\n \n \n [1] http://spraakbanken.gu.se/swe/flerordsworkshop END:VEVENT BEGIN:VEVENT UID:calendar:3297:field_eventdate:0:179 SUMMARY:Guest talk: Maciej Piasecki - A linguistically motivated WordNet model DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130320T151500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130320T170000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-03-20/guest-talk-maciej-piasecki LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Wordnets are built of synsets\, not of words. A synsets consists of words. \n Synonymy is a relation between words. Words go into a synset because they are \n synonyms. Later\, a wordnet treats words as synonymous because they belong in \n the same synset. . . Such circularity\, which is a well-known problem\, poses a \n practical difficulty in wordnet construction\, notably when it comes to \n maintaining consistency.\n \n plWordNet – a very large Polish wordnet – is a net of lexical units. We \n will discuss our assumptions and present their implementation in a steadily \n growing Polish wordnet. A small set of constitutive relations allows us to \n construct synsets automatically out of groups of lexical units of the same \n connectivity.\n \n plWordNet system of relations will be presented and compared to systems of \n relations in several influential wordnets. Additional synset-forming \n mechanisms such as stylistic registers and verb aspect will be also \n discussed. The rich morphology of Polish pertains to the important role of \n lexico-semantic relations that are derivationally motivated.\n \n The work was co-funded by the European Union Innovative Economy Programme \n (Project POIG.01.01.02-14-013/09) and the Polish Ministry of Science and \n Higher Education (Project N N516 068637).\n \n www.nlp.pwr.wroc.pl [1]\n \n www.plwordnet.pwr.wroc.pl [2]\n \n \n [1] http://www.nlp.pwr.wroc.pl\n [2] http://www.plwordnet.pwr.wroc.pl END:VEVENT BEGIN:VEVENT UID:calendar:3244:field_eventdate:0:180 SUMMARY:CLT seminar: Maciej Piasecki - Automated corpus-based WordNet expansion DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130321T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130321T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-03-21/clt-seminar-maciej-piasecki LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:A wordnet is a kind of lexical semantic network that describes word meanings \n in terms of lexical semantic relations which they enter. This is a limited \n kind of description\, but it has allowed to build large wordnets (e.g. the \n seminal Princeton WordNet\, plWordNet)\, which are useful in various Natural \n Language Processing applications. A wordnet must be large enough to provide \n practical support for such applications. Still\, its construction process is \n laborious which is a serious limitation. However\, the indispensable manual \n work can be effectively facilitated by language knowledge extracted from \n large corpora. An example are pairs of words linked by various \n lexico-semantic relations that are processed by the WordnetWeaver system to \n produce suggestions for wordnet expansion.\n \n /Pattern-based methods/ exploit occurrences of word pairs in search for \n lexico-syntactic constructions that can be markers of particular \n lexico-semantic relations. /Distributional Semantics methods/ are based on \n the analysis of statistically significant similarities among different word \n uses in order to identify those words that are semantically related. Results \n produced by methods of both types are complementary to some extent: \n pattern-based methods extract pairs of words that seem to be associated by \n particular lexico-semantic relations\, while Distributional Semantics produces \n measures of semantic relatedness between words. Advantages\, disadvantages and \n limitations of both paradigms will be discussed on the basis of rich \n practical experience in their utilisation.\n \n The WordnetWeaver system can utilise the results of a number of different \n extraction methods and suggest the likely location of a new word in the \n wordnet. Each suggestion defines a potential sense of a new word. The \n suggestions are presented visually on the relation network graph. Linguists \n can browse suggestions\, modify and freely edit the wordnet structure.\n \n The complete process of data processing and relation extraction will be \n discussed from the perspective of our experience of wordnet building. A \n corpus-based lexicographic process supported by the WordnetWeaver system will \n be presented. Possibilities and limitations of the semi-automated wordnet \n expansion will be discussed on the basis of examples collected during \n plWordNet expansion\n \n The work was co-funded by the European Union Innovative Economy Programme \n (Project POIG.01.01.02-14-013/09) and the Polish Ministry of Science and \n Higher Education (Project N N516 068637).\n \n Maciej Piasecki [1] (Wroclaw)\n \n www.nlp.pwr.wroc.pl [2]\n \n www.plwordnet.pwr.wroc.pl [3]\n \n \n [1] http://www.ii.pwr.wroc.pl/~piasecki/index-en.html\n [2] http://www.nlp.pwr.wroc.pl\n [3] http://www.plwordnet.pwr.wroc.pl END:VEVENT BEGIN:VEVENT UID:calendar:3211:field_eventdate:0:181 SUMMARY:Grammatikfestivalen DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130322T000000 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130322T000000 URL;VALUE=URI:http://www.clt.gu.se/event/2013-03-22/grammatikfestivalen LOCATION:Lilla hörsalen\, Humanisten\, Renströmsgatan 6 DESCRIPTION:More information here: http://www.svenska.gu.se/samverkan/grammatikfestivalen \n [1]\n \n Observe that deadline for submission is 15 February.\n \n If you have any questions\, please contact Elisabet Engdahl [2].\n \n \n [1] http://www.svenska.gu.se/samverkan/grammatikfestivalen\n [2] mailto:elisabet.engdahl@svenska.gu.se END:VEVENT BEGIN:VEVENT UID:calendar:3298:field_eventdate:0:182 SUMMARY:Guest talk: Maciej Piasecki - Corpus-based development process of plWordNet DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130322T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130322T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-03-22/guest-talk-maciej LOCATION:L307\, Lennart Torstenssonsgatan 8 DESCRIPTION:Wordnets are mostly constructed either on the basis of the transfer method \n applied to Princeton WordNet or on the basis of knowledge extraction from \n monolingual dictionaries. Neither of the methods could be applied in the \n construction of plWordNet. There were no publicly available bilingual \n Polish-English dictionaries nor monolingual Polish lexical resources. \n Moreover\, we wanted plWordNet to be a faithful description of the Polish \n lexicalsystem.\n \n Thus\, from the very beginning plWordNet development process was based on the \n exploration of a huge Polish corpus. Language tools were employed in \n plWordNet development on every possible step: from data gathering through \n data analysis to data presentation. A set of language tools for advanced \n corpus browsing\, as well as for the extraction of lexical semantic knowledge \n was developed and applied. The extracted knowledge was the input to the \n WordnetWeaver system which suggested nodes in the wordnet structure as \n potential attachment places for new synsets. The suggestions are presented \n visually on the relation network graph. Linguists can browse suggestions\, \n modify and edit the wordnet structure. Automatically discovered senses are \n also described by automatically identified usage examples.\n \n During the seminar\, we will discuss the complete plWordNet development cycle: \n corpus gathering and preprocessing\, lemma and lexico-semantic relation \n extraction\, visual wordnet editing supported by the extracted knowledge and \n coordination supported by a system for monitoring the work of a team of \n linguists. Expansion of derivationally motivated lexico-semantic relations is \n facilitated by tools for example-based relation learning and corpus-based \n discovering of new relation instances. Next\, we will present plWordNet to \n Princeton WordNet mapping process and tools facilitating it.\n \n The most recent size of the corpus is 1.8 billion words. The complete process \n of data processing and relation extraction will be discussed from the \n perspective of our experience of wordnet building. A corpus-based \n lexicographic process supported by the Wordnet Weaver system will be \n presented. Possibilities and limitations of the semi-automated wordnet \n expansion will be discussed on the basis of examples collected during \n plWordNet expansion.\n \n The work was co-funded by the European Union Innovative Economy Programme \n (Project POIG.01.01.02-14-013/09) and the Polish Ministry of Science and \n Higher Education (Project N N516 068637).\n \n www.nlp.pwr.wroc.pl [1]\n \n www.plwordnet.pwr.wroc.pl [2]\n \n \n [1] http://www.nlp.pwr.wroc.pl\n [2] http://www.plwordnet.pwr.wroc.pl END:VEVENT BEGIN:VEVENT UID:calendar:3245:field_eventdate:0:183 SUMMARY:CLT seminar: Victor Sánchez-Cartagena - Learning resources for \n shallow-transfer machine translation DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130404T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130404T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-04-04/clt-seminar-victor-sanchez-cartagena LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Rule-based machine translation systems need explicit linguistic resources \n which are usually coded by human experts in a highly time-consuming task. \n When experts are not available for a given language pair\, automatic or \n semi-automamic acquisition of these resources can be carried out.\n \n In this talk\, a method to build monolingual dictionaries from the \n contributions of non-expert users is presented. A strategy to learn \n shallow-transfer rules from small parallel corpora is also described. \n Finally\, the integration of shallow-transfer rules in statistical machine \n translation is addressed.\n \n Victor Sánchez-Cartagena [1] (Alacant)\n \n \n [1] http://www.dlsi.ua.es/~vmsanchez/ END:VEVENT BEGIN:VEVENT UID:calendar:3301:field_eventdate:0:184 SUMMARY:Guest talk: Chris Koniaris - Dealing With Spoken Language Communication DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130405T131500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130405T150000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-04-05/guest-talk-chris-koniaris LOCATION:T346\, Olof Wijksgatan 6 DESCRIPTION:Research in Spoken language technology (SLT) has done tremendous steps during \n the last 50 years. From the offline isolated word recognition in late \n 60's/early 70's to nowadays real-time multimodal (and multipurpose) spoken \n dialogue systems\, the initially limited scientific area has been expanded\, \n and its research outcome has become broadly available. As a result\, new \n fields of application have been developed with an increasing demand for high \n quality products. Still\, the available technology has not yet reached the \n performance of human communication.\n \n A recent trend includes biologically-inspired hypotheses and \n perceptually-relevant assumptions in order to find a path for a new \n breakthrough in human-machine interaction by incorporating different \n scientific areas in a joint\, multidisciplinary research effort. In this talk\, \n we examine the importance of human perception in SLT and give examples of \n relevant applications.\n \n Web page at KTH [1]\n \n \n [1] https://www.kth.se/profile/147757/ END:VEVENT BEGIN:VEVENT UID:calendar:3246:field_eventdate:0:185 SUMMARY:CLT seminar: Dry run of talks for the FrameNet Workshop in Berkeley DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130411T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130411T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-04-11/clt-seminar-dry-run-talks-framenet-workshop-berkeley LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Four short talks in preparation for the FrameNet workshop in Berkeley [1].\n  \n \n * Richard Johansson: Automatic Lexicon Expansion in the Swedish FrameNet\n * Dimitrios Kokkinakis: Medical Event Extraction using the Swedish FrameNet\n * Leif-Jöran Olsson: The lexical editing system of Karp\n * Rudolf Rydstedt: Globally defined semantic roles in the Swedish\n Constructicon\n \n \n [1] http://spraakbanken.gu.se/eng/Forskning/SweFN/FrameNetWS2013 END:VEVENT BEGIN:VEVENT UID:calendar:3248:field_eventdate:0:186 SUMMARY:CLT seminar: Eleni Miltsakaki - Do machine learning and NLP improve \n traditional readability formulas? DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130425T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130425T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-04-25/clt-seminar-eleni-miltsakaki LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Readability formulas are methods used to match texts with the readers' \n reading level. Several methodological paradigms have previously been \n investigated in the field. The most popular paradigm dates several decades \n back and gave rise to well known readability formulas such as the Flesch \n formula (among several others).\n \n In this talk\, I will present the results of a study I did in collaboration \n with Thomas Francois from the UCLouvain. We compare traditional readability \n formula approaches (henceforth "classic") with an emerging paradigm which \n uses sophisticated NLP-enabled features and machine learning techniques.\n \n Our experiments\, carried on a corpus of texts for French as a foreign \n language\, yield four main results: (1) the new readability formula performed \n better than the "classic" formula\; (2) "non-classic" features were slightly \n more informative than "classic" features\; (3) modern machine learning \n algorithms did not improve the explanatory power of our readability model\, \n but allowed to better classify new observations\; and (4) combining "classic" \n and "non-classic" features resulted in a significant gain in performance.\n \n https://sites.google.com/site/elenimi2/home [1]\n \n \n [1] https://sites.google.com/site/elenimi2/home END:VEVENT BEGIN:VEVENT UID:calendar:3254:field_eventdate:0:187 SUMMARY:Thesis defence: Katarina Heimann Mühlenbock - I see what you mean DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130426T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130426T130000 URL;VALUE=URI:http://www.clt.gu.se/event/2013-04-26/thesis-defence-katarina-heimann-m%C3%BChlenbock LOCATION:Stora hörsalen\, Humanisten\, Renströmsgatan 6 DESCRIPTION:Katarina Heimann Mühlenbock [1] will defend her PhD thesis: I see what you \n mean - Assessing readability for specific target groups\n \n Opponent: Eleni Miltsakaki [2] (University of Pennsylvania)\n \n Abstract:\n \n This thesis aims to identify linguistic factors that affect readability and \n text comprehension\, viewed as a function of text complexity. Features at \n various linguistic levels suggested in existing literature are evaluated\, \n including the Swedish readability formula LIX. Natural language processing \n methods and resources are employed to investigate characteristics that go \n beyond traditional superficial measures. A comparable corpus of eay-to-read \n and ordinary texts from three genres is investigated\, and it is shown how \n features present at various levels of representation differ quantitatively \n across text types and genres. The findings are confirmed in significance \n tests as well as principal component analysis. Three machine learning \n algorithms are employed and evaluated in order to build a statistical model \n for text classification. The results demonstrate that a proposed language \n model for Swedish (SVIT)\, utilizing a combination of linguistic features\, \n actually predicts text complexity and genre with a higher accuracy than LIX. \n It is suggested that the SVIT language model should be adopted to assess \n surface language properties\, vocabulary load\, sentence structure\, idea \n density levels as well as the personal interests of different texts. Specific \n target groups of readers may then be provided with materials tailored to \n their level of proficiency.\n \n Full text at GUPEA [3].\n \n \n [1] http://spraakbanken.gu.se/swe/personal/katarina\n [2] https://sites.google.com/site/elenimi2/home\n [3] https://gupea.ub.gu.se/handle/2077/32472 END:VEVENT BEGIN:VEVENT UID:calendar:3249:field_eventdate:0:188 SUMMARY:CLT seminar: Elena Volodina and Ildikó Pilán - Developing Lärka\, an ICALL \n platform for studying Swedish as a Second Language DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130502T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130502T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-05-02/clt-seminar-volodina-and-pilan LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:The ICALL* platform /Lärka/ [1]  is an open-source web-based application \n that uses principles of Service-Oriented Architecture  and reuses Korp [2] \n and Karp [3] for exercise generation.\n \n The platform in aimed primarily at learners of Swedish as a Second/Foreign \n language. It is divided into several modules: an exercise generator with \n activities for university students of linguistics and second/foreign language \n learners including multiple-choice and spelling exercises\; and modules \n facilitating different aspects of development and research. These at the \n moment consist of an experimental sentence readability module for a \n level-wise selection of appropriate dictionary examples / exercise items and \n an editor for learner-oriented corpora.\n \n The platform is under active development\, and in this talk we will describe \n its current state\, including the exercise generation and the principles \n behind them\, as well as the two projects relevant for Lärka's development: \n the project on the collection of a corpus of course book texts used in \n CEFR**-based language teaching and the sentence readability project.\n We expect our talk to be interesting for computational linguists\, language \n teachers\, lexicographers and linguists in general.\n \n *ICALL = Intelligent Computer-Assisted Language Learning\n **CEFR = Common European Framework of Reference for Languages\n \n \n [1] http://spraakbanken.gu.se/eng/research/infrastructure/l%C3%A4rka\n [2] http://spraakbanken.gu.se/korp/\n [3] http://spraakbanken.gu.se/karp END:VEVENT BEGIN:VEVENT UID:calendar:3250:field_eventdate:0:189 SUMMARY:CLT seminar: Dry run of talks for Nodalida 2013 in Oslo DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130516T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130516T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-05-16/clt-seminar-dry-run-nodalida LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Dry run of talks for Nodalida 2013 [1] in Oslo\n \n Yvonne Adesam & Gerlof Bouma: Experiments on sentence segmentation in Old \n Swedish editions\n \n Malin Ahlberg & Peter Andersson: Towards automatic tracking of lexical \n change: linking historical lexical resources\n \n \n [1] http://www.hf.uio.no/iln/english/research/news-and-events/events/conferences/2013/nodalida/index.html END:VEVENT BEGIN:VEVENT UID:calendar:3311:field_eventdate:0:190 SUMMARY:Final PhD seminar: Karin Cavallin - Detecting Lexical change via Semantic \n Distribution DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130528T131500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130528T150000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-05-28/final-phd-seminar-cavallin LOCATION:T340\, Olof Wijksgatan 6 DESCRIPTION:Karin Cavallin (PhD student in GSLT [1]) will present her forthcoming PhD \n thesis:\n \n Detecting Lexical change via Semantic Distribution - Investigating meaning \n change in and through Lexical Sets\n \n Opponent: Richard Johansson [2]\, Språkbanken\n \n \n [1] http://www.gslt.hum.gu.se/people-research/phd-students/\n [2] http://spraakbanken.gu.se/eng/personal/richard END:VEVENT BEGIN:VEVENT UID:calendar:3317:field_eventdate:0:191 SUMMARY:Guest talk: Graeme Hirst - Who decides what a text means? DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130529T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130529T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-05-29/guest-talk-graeme-hirst LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Writer-based and reader-based views of text-meaning are reflected by the \n respective questions "What is the author trying to tell me?" and "What does \n this text mean to me personally?" Contemporary computational linguistics\, \n however\, generally takes neither view\; applications do not attempt to answer \n either question. Instead\, a text is regarded as an object that is independent \n of\, or detached from\, its author or provenance\, and as an object that has the \n same meaning for all readers. This is not adequate\, however\, for the further \n development of sophisticated NLP applications for intelligence gathering and \n question answering.\n \n I will discuss different views of text-meaning from the perspective of the \n needs of computational text analysis\, and then extend the analysis to include \n discourse as well – in particular\, the collaborative construction of \n meaning and the collaborative repair of misunderstanding.\n \n Graeme Hirst (Toronto) [1]\n \n \n [1] http://www.cs.toronto.edu/~gh/ END:VEVENT BEGIN:VEVENT UID:calendar:3251:field_eventdate:0:192 SUMMARY:CLT seminar: Graeme Hirst - Natural language processing methods for the \n detection of symptoms of Alzheimer's disease in writing DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130530T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130530T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-05-30/clt-seminar-graeme-hirst LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Studies of language in Alzheimer's disease have concluded that\, along with a \n general cognitive decline\, linguistic features are also negatively affected. \n Studies of the language of healthy elders also observe a linguistic decline\, \n but one which\, in contrast\, is markedly less severe than that induced by \n dementia.  We examine whether the disease can be detected from the \n diachronic changes in written texts and\, more importantly\, whether it can be \n clearly distinguished from normal aging.\n \n Lexical and syntactic analyses were conducted on 51 novels by three prolific \n literary authors: Iris Murdoch\, P.D. James\, and Agatha Christie. Murdoch was \n diagnosed with Alzheimer's disease shortly after finishing her last novel\; \n James\, at 89 years of age\, continues to publish critically-acclaimed works\; \n Christie\, whose last few novels are deemed strikingly subpar compared to her \n previous works\, presents an interesting case study of possible dementia.\n \n The lexical analysis reveals significant patterns of decline in Murdoch's and \n Christie's later novels\, while James's rates remain relatively consistent \n throughout her career. The syntactic measures\, though unveiling fewer \n significant linear trends\, discover a cubic model of change in Murdoch's \n novels\, with a deep decline around her 50s. Our findings provide support for \n the hypothesis that dementia\, which manifests clearly in lexical features\, \n can be detected in writing.\n \n Graeme Hirst (Toronto) [1]\n \n (Joint work with Xuan Le\, Ian Lancashire\, and Regina Jokel)\n \n \n [1] http://www.cs.toronto.edu/~gh/ END:VEVENT BEGIN:VEVENT UID:calendar:3252:field_eventdate:0:193 SUMMARY:(PRELIMINARY) CLT seminar: Jyrki Nummenmaa - TBA DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130613T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130613T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-06-13/clt-seminar-jyrki-nummenmaa LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:Jyrki Nummenmaa [1] (Tampere)\n \n \n [1] http://www.cs.uta.fi/~jyrki/ END:VEVENT BEGIN:VEVENT UID:calendar:3253:field_eventdate:0:194 SUMMARY:CLT seminar: Martin Volk - TBA DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130829T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130829T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-08-29/clt-seminar-martin-volk-tba LOCATION:L308\, Lennart Torstenssonsgatan 8 END:VEVENT BEGIN:VEVENT UID:calendar:3292:field_eventdate:0:195 SUMMARY:CLT seminar: Stina Ericsson and Cajsa Ottesjö - LekBot part 2: robot \n development and genuine play DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130905T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130905T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-09-05/clt-seminar-ericsson-and-ottesjo LOCATION:L308\, Lennart Torstenssonsgatan 8 DESCRIPTION:The LekBot project\, part 1 [1]\, was a collaboration between DART [2]\, \n Talkamatic [3] and GU in 2010-2011. The project developed a talking and \n playing robot for children with communicative disabilities\, with the aim of \n providing a toy that is easy and fun to use\, and that provides opportunities \n for genuine play in the sense of play that is spontaneous\, independent\, on \n equal terms\, etc. Three test groups participated in the project\, with each \n test group consisting of a child with cerebral palsy\, a peer and pre-school \n staff. All groups were recorded in interactions with various versions of the \n system.\n \n The LekBot project\, part 2\, started in 2012\, and is a collaboration between \n GU and DART. The focus now is on the analysis of the recorded interactions. \n In the talk we will present current results of the analysis\, and discuss \n implications for further development of the LekBot system.\n \n \n [1] http://www.clt.gu.se/finishedproject/lekbot\n [2] http://www.dart-gbg.org/index2.html\n [3] http://www.talkamatic.se/ END:VEVENT BEGIN:VEVENT UID:calendar:3295:field_eventdate:0:196 SUMMARY:CLT seminar: TBA - TBA DTSTAMP;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130520T004530 DTSTART;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130926T101500 DTEND;TZID=Europe/Stockholm;VALUE=DATE-TIME:20130926T120000 URL;VALUE=URI:http://www.clt.gu.se/seminar/2013-09-26/clt-seminar-tba-tba LOCATION:L308\, Lennart Torstenssonsgatan 8 END:VEVENT END:VCALENDAR