HOME
Program
Submission
Registration
Directions
Accommodation
Video










Flier download (pdf)
Flier download (Word)
Flier download (A4,pdf)
Flier download (A4,Word)
Word file has better quality















Proteus ProjectProteus Project,
New York University

ODIE Project

National Science Foundation

NSF Sponsored Symposium

Semantic Knowledge Discovery, Organization and Use



November, 14 and 15, 2008
(Friday and Saturday)
Warren Weaver Hall, New York University


What's NEW

12/4: Slides and video are available The slides and video of most of the presenters are available. Please go to Program HP for downloading slides, and go to Video HP to see video of the presentations.
11/15: The symposium is over The symposium is successfuly over. Thank you very much for all invited speakers, presenters and participants!!
11/3: Registration closed As the number of participants exceeds the capacity of the room, we are very sorry to announce that we will close the registration. It is the law that we can't have more than the capacity people in the room. We will distribute the notes, slides and other materials. Please stay tuned!!
10/7: Program is now available. Click here. Thak you for 45 submissions for the general session. We assign 6 oral presentations, 31 poster presentations and 8 demos.
9/15: Registration/Accomodation page up. Registration should be done through Registration page by October 31, 2008. Also, the Accomodation page is now open.
9/12: Travel Support (CLOSED) Six people will receive free hotel stay. The selection will be made based on the peoposal. Many thanks to Dr. Bill Dolan, Dr. Marius Pasca and Dr. Dekang Lin for their support. (See Submission page)

Objectives

The focus of NLP research has been shifting towards semantic analysis from syntactic analysis. It has become evident that the methods employed for developing syntactic analyzers, i.e. supervised methods using small annotated corpora, are not the best methods for the semantic task. In order to handle semantics, we need large amounts of knowledge which may be best collected by semi/un- supervised methods from a huge unannotated corpus.

Many methods have already been proposed along this line of research, e.g. discovery of synonyms, hyponym-hypernyms, part-of relations, paraphrase, textual entailment, relations between things, relations between events and so on. The technical methods include distributional similarity, lexico-syntactic patterns, alignment and so on. Although such discovery techniques have been improving through numerous trials, there has been little discussion of the background and high-level picture of the field.

The objectives of this symposium will be to discuss the following topics:

  • High-level picture of the current technologies
  • The types of semantic knowledge we need
  • The relationships between knowledge and applications
  • The format for representing this knowledge and its re-use
  • Community effort for resource and platform
  • Possible directions for the technology in the future

The presentations will consist of invited talks by leaders in the field (shown below) and partially of general submissions. Presentations are expected to provide the speaker's high level position regarding the objectives described above.


Invited Speakers

Prof. Ido Dagan (Bar Ilan University, Israel)
Dr. Bill Dolan (Microsoft Research, USA)
Prof. Oren Etzioni (University of Washington, USA)
Prof. Christiane D. Fellbaum (Princeton University, USA)
Prof. Marti Hearst (UC Berkeley, USA)
Prof. Kentaro Inui (NAIST, Japan)
Dr. Dekang Lin (Google, USA)
Prof. Bernardo Magnini (FBK-irst, Italy)
Prof. Dan Moldovan (Lymba Corporation, USA)
Prof. Patrick Pantel (Yahoo! Labs, USA)
Dr. Marius Pasca (Google, USA)
Prof. Peter Turney (National Research Council, Canada)

PC Member

  • Prof. Ralph Grishman (New York University)
  • Prof. Sadao Kurohashi (Kyoto University)
  • Prof. Dan Roth (UIUC)
  • Prof. Deepak Ravichandran (Google)
  • Prof. Julio Gonzalo (UNED)
  • Prof. Dan Jurafsky (Stanford University)
  • Dr. Idan Szpektor (Bar-Ilan University)
  • Prof. Rada Mihalcea (University of North Texas)
  • Prof. Kentaro Torisawa (NICT)
  • Prof. Jun-ichi Tsujii (Tokyo University)
  • Dr. Roy Bar-Haim (Bar-Ilan University)
  • Prof. Eduard Hovy (ISI, USC)
  • Prof. Nicoletta Calzolari (ILC-CNR)
  • Prof. Mark Stevenson (University of Sheffield)
  • Prof. Diana McCarthy (University of Sussex)
  • Prof. Mirella Lapata (University of Edinburgh)
  • Prof. Chris Callison-Burch (Johns Hopkins University)
  • Dr. Peter Clark (Networked Systems Technology, The Boeing Company)
  • Prof. Martha Palmer (University of Colorado Boulder)
  • Prof. Dragomir R. Radev (University of Michigan)
  • Prof. Yorick Wilks (University of Shefield)
  • Prof. Adam Meyers (New York University)
  • Prof. Lillian Lee (Cornell University)
  • Prof. Ellen Riloff (University of Utah)

Organizer

Prof. Satoshi Sekine (New York University, USA)

General Submission

There are three categories for general submissions. Presentations are likely to be 15-20 minute talks. Posters or demos will be presented in a 2 hour slot. It is NOT necessary that the material be previously unpublished. We are expecting presentations giving an overview of the research at your site, or your top-level view of the topic, or a demo of a system related to the topic. Please send sekine cs.nyu.edu a one page description of your presentation (please indicate if it is a poster or demo). Deadline for submissions is September 30.

We are NOT planning to produce proceedings (i.e. you do NOT have to prepare papers), but we would like to produce a binder of the presentation slides for invited speakers and abstracts for general submissions, which may due about two weeks before the symposium.

Thank you very much for your interest; we are looking forward to see you at the symposium.


Sponsor

This symposium is sponsored by the National Science Foundation under Grant IIS-00325657, "On-Demand Information Extraction Project.