COLING 2008 Workshop
TextGraphs-3
Graph-based Algorithms for
Natural Language Processing

Manchester, UK, August 24, 2008


NEW: Registration is now open - Early registration till July 11, 2008
Go to COLING 2008 registration

In connection with COLING 2008, the 22nd International Conference on Computational Linguistics August 18-22


The workshop will be held on August 24, 2008, in conjunction with COLING in Manchester, UK.

Workshop Schedule

24th August 2008, Sunday

Time

Paper

Authors

9:20 - 9:25

Opening

Chairs

9:30 -10:30

Invited Talk: Lexical Centrality

Dragomir Radev

10:30-11:00

BREAK

Session I: Full Papers

11:00-11:30

Acquistion of the morphological structure of the lexicon based on lexical similarity and formal analogy

Nabil Hathout

11:30-12:00

Learning to Map Text to Graph-based Meaning Representations via Grammar Induction

Smaranda Muresan

12:00-12:30

How Is Meaning Grounded in Dictionary Definitions?

Alexandre Blondin Massé, Guillaume Chicoisne, Yassine Gargouri, Stevan Harnad, Odile Marcotte and Olivier Picard

12:30-14:00

LUNCH

Session II: Full Papers

14:00-14:30

Encoding Tree Pair-based Graphs in Learning Algorithms: the Textual Entailment Recognition Case

Alessandro Moschitti and Fabio Massimo Zanzotto

14:30-15:00

Graph-based Clustering for Semantic Classification of Onomatopoetic Words

Kenichi Ichioka and Fumiyo Fukumoto

15:00-15:30

Affinity Measures Based on the Graph Laplacian

Delip Rao, David Yarowsky and Chris Callison-Burch

15:30-16:00

BREAK

Session III: Short Papers

16:00-16:20

Semantic structure from Correspondence Analysis

Barbara McGillivray, Christer Johansson and Daniel Apollon

16:20-16:40

Concept-graph based Biomedical Automatic Summarization using Ontologies

Laura Plaza, Alberto Díaz and Pablo Gervás

16:40-17:00

Random Graph Model Simulations of Semantic Networks for Associative Concept Dictionaries

Hiroyuki Akama, Jaeyoung Jung, Terry Joyce and Maki Miyake

NEW!

Invited Talk

We are pleased to announce the invited talk by , Dragomir Radev, Associate Professor at School of Information, Department of Electrical Engineering and Computer Science, Department of Linguistics at the University of Michigan in Ann Arbor. The talks title and abstract will appear shortly.

NEW!

List of Accepted Papers

  • Graph-based Clustering for Semantic Classification of Onomatopoetic Words

    Fumiyo Fukumoto and Kenichi Ichioka

  • Concept-graph based Biomedical Automatic Summarization using Ontologies

    Laura Plaza, Alberto Díaz and Pablo Gervás

  • How Is Meaning Grounded in Dictionary Definitions?

    Alexandre Blondin Massé, Guillaume Chicoisne, Yassine Gargouri, Stevan Harnad, Odile Marcotte and Olivier Picard

  • Acquistion of the morphological structure of the lexicon based on lexical similarity and formal analogy

    Nabil Hathout

  • Random Graph Model Simulations of Semantic Networks for Associative Concept Dictionaries

    Hiroyuki Akama, Jaeyoung Jung, Terry Joyce and Maki Miyake

  • Learning to Map Text to Graph-based Meaning Representations via Grammar Induction

    Smaranda Muresan

  • Affinity Measures Based on the Graph Laplacian

    Delip Rao and David Yarowsky

  • Semantic structure from Correspondence Analysis

    Barbara McGillivray, Christer Johansson and Daniel Apollon

  • Encoding Tree Pair-based Graphs in Learning Algorithms: the Textual Entailment Recognition Case

    Alessandro Moschitti and Fabio Massimo Zanzotto


Call for Papers

Recent years have shown an increased interest in bringing the field of graph theory into natural language processing. Graph theory is a well studied discipline, and so is the field of natural language processing. Traditionally, these two areas of study have been perceived as distinct, with different algorithms, different applications, and different potential end-users. However, as recent research work has shown, the two disciplines are in fact intimately connected, with a large variety of natural language processing applications finding efficient solutions within graph-theoretical frameworks.

In many NLP applications entities can be naturally represented as nodes in a graph and relations between them can be represented as edges. Recent research has shown that graph-based representations of linguistic units as diverse as words, sentences and documents give rise to novel and efficient solutions in a variety of NLP tasks, ranging from part of speech tagging, word sense disambiguation and parsing to information extraction, semantic role assignment, summarisation, sentiment analysis and up to the study of the evolutionary dynamics of language.

The TextGraphs workshop addresses a broad spectrum of research areas and brings together researchers working on problems related to the use of graph-based algorithms for natural language processing as well as on the theory of graph-based methods. Different NLP applications use different graph-based approaches, and bringing together researchers from different fields fosters the exchange of ideas. Furthermore, looking at graph-based methods from the perspective of diverse applications facilitates a discussion about the theory of graph-based methods and about the theoretical justification of the empirical results within the NLP community. This discussion is vital because further progress in graph-based NLP applications is impossible without deeper understanding of basic theoretical principles.

Starting with TextGraphs-3 we would like to have one area of graph-based NLP research as the primary topic for discussion.We intend to set this year's focus on large scale lexical acquisition and representation. Efficient graph methods can help to alleviate the acquisition bottleneck for lexicon construction and resource building. They also provide smarter representation schemes for the lexicon that facilitate fast search and word retrieval.SIGLEX endorsed our workshop proposal for COLING-08.

We invite submissions of papers on graph-based methods applied to NLP problems. Especially, we encourage submissions regarding

  • Large-scale lexical acquisition using graph representations
  • Graph-based representation schemes of the mental lexicon

Other topics include, but are not limited to:
  • Graph representations for ontology learning
  • Graph labeling and edge labeling for semantic representations
  • Encoding semantic distances in graphs
  • Graph algorithms for word sense disambiguation
  • Graph methods for Information Retrieval, Information Extraction, Text Mining and Understanding
  • Random walk graph methods
  • Spectral graph clustering
  • Small world graphs in natural language processing
  • Semi-supervised graph-based methods
  • Statistical network methods and analysis
  • Dynamic graph representations for NLP


Organisation Committee

Irina Matveeva, Accenture Technology Labs, matveeva AT cs.uchicago.edu
Chris Biemann, Powerset, biem AT informatik.uni-leipzig.de
Monojit Choudhury, Microsoft Research, monojitc AT microsoft.com
Mona Diab,Columbia University, mdiab AT ccls.columbia.edu

Programme Committee

Eneko Agirre, University of the Basque Country
Edo Airoldi, Princeton University
Regina Barzilay, MIT
Fernando Diaz, Yahoo! Montreal
Güneş Erkan, Google
Michael Gamon, Microsoft Research
Andrew Goldberg, University of Wisconsin
Hany Hassan, IBM Egypt
Samer Hassan, University of North Texas
Gina Levow, University of Chicago
Rada Mihalcea, University of North Texas
Animesh Mukherjee, IIT Kharagpur
Dragomir Radev, University of Michigan
Uwe Quasthoff, University of Leipzig
Aitor Soroa, University of the Basque Country
Hans Friedrich Witschel, University of Leipzig
Fabio Massimo Zanzotto, University of Rome "Tor Vergata"
Thorsten Zesch, University of Darmstadt



Important Dates, note the Deadline Extension

Regular paper submissionsMay 12, 2008 (23:59 GMT -12:00)
Short paper submissions May 19, 2008
Notification of acceptance June 6, 2008
Camera-ready papers July 1, 2008
Workshop August 24, 2008



Author Instructions

Submissions will consist of regular full papers of max. 8 pages and short papers of max. 4 pages, formatted following the COLING 2008 formatting guidelines. Papers should be submitted using the Online Submission Form. The review process double blind, please anonymise your submission. For any questions, please contact one of the organisers.


Previous Workshops

The TextGraphs-3 workshop builds on the success of the first and second TextGraphs workshops:
TextGraphs-1 at HLT-NAACL 2006, for proceedings click here
TextGraphs-2 at HLT-NAACL 2007, for proceedings click here
There is the tutorial Graph-based Algorithms for Information Retrieval and Natural Language Processing from HLT-NAACL 2006 by Dragomir Radev and Rada Mihalcea.