ACL 2007 Workshop on Deep Linguistic Processing
June 28th, 2007
Prague, Czech Republic
Workshop Description
This workshop is aimed at bringing together the different
computational linguistic sub-communities which model language predominantly
by way of theoretical syntax, either in the form of a particular theory
(e.g. CCG, HPSG, LFG, LTAG+ or the Prague School) or a more general framework
which draws on theoretical and descriptive linguistics. We characterise this
style of computational linguistic research as deep linguistic processing
,
due to it aspiring to model the complexity of natural language in rich
linguistic representations.
Background
Deep linguistic processing has traditionally been concerned with grammar development. The linguistic precision and complexity of the grammars meant that they had to be manually developed and maintained, and were computationally expensive to run. With recent developments in computer hardware, parsing/generation algorithms and statistical learning theory, the way has been opened for deep linguistic processing to be successfully applied to an ever-growing range of languages, domains and applications.
This workshop aims to foster existing and new relationships between groups working on deep linguistic processing, highlighting the considerable linguistic, developmental and algorithmic commonalities shared by the various approaches.
Topics
Papers are invited on substantial, original, and unpublished research concerning deep linguistic processing. Possible topics include:- grammar engineering (e.g. frameworks for grammar evaluation, best practice in grammar engineering, cross-linguistic/formalism generalisations & comparisons, semantic representation)
- treebanking (e.g. frameworks for treebank evaluation/normalisation, grammar extraction/induction, the interface between grammar engineering and treebanking, treebanking methodologies, cross-linguistic/formalism generalisations & comparisons)
- system development (e.g. grammar profiling, system integration, preprocessing strategies, robustness enhancement)
- parser/generator development (e.g. algorithm development, grammar reversibility, efficiency, evaluation)
- machine learning for deep linguistic processing (e.g. parse selection/ranking, supertagging, deep lexical acquisition, grammar induction)
- applications of deep linguistic processing (e.g. information extraction, question answering, machine translation, dialogue systems, CALL)
Invited Speaker
Anette Frank (University of Heidelberg) is to give an invited talk at the workshop with the provisional title Across languages and grammar paradigms — New perspectives on resource acquisition, grammar engineering and applicationsSchedule
| 08:35—08:45 | Opening Remarks |
| Session 1: Parsing | |
| 08:45—09:15 | Multi-Component Tree Adjoining Grammars, Dependency Graph Models, and Linguistic Analyses |
| Joan Chen-Main and Aravind Joshi | |
| 09:15—09:45 | Perceptron Training for a Wide-Coverage Lexicalized-Grammar Parser |
| Stephen Clark and James Curran | |
| 09:45—10:15 | Filling Statistics with Linguistics — Property Design for the Disambiguation of German LFG Parses |
| Martin Forst | |
| 10:15—10:45 | Exploiting Semantic Information for HPSG Parse Selection |
| Sanae Fujita, Francis Bond, Stephan Oepen and Takaaki Tanaka | |
| 10:45—11:15 | COFFEE BREAK |
| Session 2: Applications of Deep Linguistic Processing | |
| 11:15—11:45 | Deep Grammars in a Tree Labeling Approach to Syntax-based Statistical Machine Translation |
| Mark Hopkins and Jonas Kuhn | |
| 11:45—12:15 | Question Answering based on Semantic Roles |
| Michael Kaisser and Bonnie Webber | |
| 12:15—13:45 | LUNCH |
| 13:45—14:45 | INVITED TALK |
| Across Languages and Grammar Paradigms — New Perspectives on Resource Acquisition, Grammar Engineering and Application | |
| Anette Frank | |
| Session 3: Posters | |
| 14:45—15:45 | Deep Linguistic Processing for Spoken Dialogue Systems |
| James Allen, Myroslava Dzikovska, Mehdi Manshadi and Mary Swift | |
| Self- or Pre-Tuning? Deep Linguistic Processing of Language Variants | |
| AntónioBranco and Costa Francisco | |
| Pruning the Search Space of a Hand-Crafted Parsing System with a Probabilistic Parser | |
| Aoife Cahill, Tracy Holloway King and John T. Maxwell III | |
| Semantic Composition with (Robust) Minimal Recursion Semantics | |
| Ann Copestake | |
| A Task-based Comparison of Information Extraction Pattern Models | |
| Mark Greenwood and Mark Stevenson | |
| Creating a Systemic Functional Grammar Corpus from the Penn Treebank | |
| Matthew Honnibal and James R. Curran | |
| Verb Valency Semantic Representation for Deep Linguistic Processing | |
| Aleš Horák, Karel Pala, Marie Duží and Pavel Materna | |
| The Spanish Resource Grammar: Pre-processing Strategy and Lexical Acquisition | |
| Montserrat Marimon, Núria Bel, Sergio Espeja and Natalia Seghezzi | |
| Extracting a Verb Lexicon for Deep Parsing from FrameNet | |
| Mark McConville and Myroslava O. Dzikovska | |
| Fips, A "Deep" Linguistic Multilingual Parser | |
| Eric Wehrli | |
| Partial Parse Selection for Robust Deep Processing | |
| Yi Zhang, Valia Kordoni and Erin Fitzgerald | |
| 15:45—16:15 | COFFEE BREAK |
| Session 4: Grammar Engineering | |
| 16:15—16:45 | Validation and Regression Testing for a Cross-linguistic Grammar Resource |
| Emily M. Bender, Laurie Poulson, Scott Drellishak and Chris Evans | |
| 16:45—17:15 | Local Ambiguity Packing and Discontinuity in German |
| Berthold Crysmann | |
| 17:15—17:45 | The Corpus and the Lexicon: Standardising Deep Lexical Acquisition Evaluation |
| Yi Zhang, Timothy Baldwin and Valia Kordoni | |
| 17:45—18:15 | Discussion and Closing Remarks |
Important Dates
- Paper Submission Deadline: CLOSED
- Notification of Paper Acceptance: ALL NOTIFICATIONS SENT OUT
- Camera Ready Submission Deadline: ALL SUBMISSIONS RECEIVED
- Workshop date: June 28th, 2007
Workshop Organisers
- Timothy Baldwin
- University of Melbourne
- Mark Dras
- Macquarie University
- Julia Hockenmaier
- University of Pennsylvania
- Tracy Holloway King
- PARC
- Gertjan van Noord
- University of Groningen
Programme Committee
- Jason Baldridge (University of Texas at Austin)
- Emily Bender (University of Washington)
- Raffaella Bernardi (University of Bolzano)
- Francis Bond (NICT)
- Gosse Bouma (University of Groningen)
- Ted Briscoe (University of Cambridge)
- Miriam Butt (University of Konstanz)
- Aoife Cahill (Stuttgart University)
- David Chiang (ISI)
- Stephen Clark (Oxford University)
- Ann Copestake (University of Cambridge)
- James Curran (University of Sydney)
- Stefanie Dipper (Potsdam University)
- Katrin Erk (University of Texas at Austin)
- Dominique Estival (Appen Pty Ltd)
- Dan Flickinger (Stanford University)
- Anette Frank (University of Heidelberg)
- Josef van Genabith (Dublin City University)
- John Hale (Michigan State University)
- Ben Hutchinson (Google)
- Mark Johnson (Brown University)
- Aravind Joshi (University of Pennsylvania)
- Laura Kallmeyer (Tuebingen University)
- Ron Kaplan (PARC)
- Martin Kay (Stanford University/Saarland University)
- Valia Kordoni (Saarland University)
- Anna Korhonen (University of Cambridge)
- Jonas Kuhn (Potsdam University)
- Rob Malouf (San Diego State University)
- Ryan McDonald (Google)
- Yusuke Miyao (University of Tokyo)
- Diego Molla (Macquarie University)
- Stefan Müller (Bremen University)
- Joakim Nivre (Växjö University)
- Stephan Oepen (University of Oslo and Stanford University)
- Anoop Sarkar (Simon Fraser University)
- David Schlangen (Potsdam University)
- Mark Steedman (University of Edinburgh)
- Beata Trawinski (Tuebingen University)
- Aline Villavicencio (Federal University of Rio Grande do Sul)
- Tom Wasow (Stanford University)
- Michael White (Ohio State University)
- Shuly Wintner (University of Haifa)
- Fei Xia (University of Washington)