LEARNING LANGUAGE IN LOGIC (LLL) WORKSHOP

Information & On-line workshop notes

30th June 1999
White Room, Grand Hotel Toplice
Bled, Slovenia
Supported by:
ILPNet2
ILP2 project end-user club
MLNET
CompulogNet


Introduction

Natural language learning based on statistical approaches (eg n-gram language modelling) has proved successful, but it is well known that such linguistically impoverished approaches have severe limitations. The flexibility and expressivity of logical representations make them highly suitable for natural language analysis, so there is consequently a growing interest in applying Inductive Logic Programming techniques to linguistic learning problems.

From the NLP point of view the promise of ILP is that it will be able to steer a mid-course between the two alternatives of large scale, but shallow levels of analysis, and small scale, but deep and precise analyses, and produce a better ratio between breadth of coverage and depth of analysis.

From the ILP point of view, NLP has recently been recognised as an ideal application area. The existence within NLP problems of hierarchically defined, structured data with large amounts of relevant logically defined background knowledge provides a perfect testbed for stretching ILP technology in a way that would also be beneficial in other application areas.

The workshop aims to provide a forum for discussion on all aspects of learning language in logic. Research topics include, but are not restricted to:
 



Invited speakers

Eric Brill
Natural Language Processing Lab, Center for Language and Speech Processing & Department of Computer Science,
Johns Hopkins University, USA
           http://www.cs.jhu.edu/~brill/home.html

        "Toward Linguistically Sophisticated Models of Language"


Ted Briscoe
        Natural Language Group,
        University of Cambridge Computer Laboratory
        http://www.cl.cam.ac.uk/users/ejb/

        "Bayesian Learning of (Stochastic) Grammars"
 


Ray Mooney
        Machine Learning Research Group,
        University of Texas at Austin
        http://www.cs.utexas.edu/users/mooney/

        "Learning for Semantic Interpretation: Scaling Up Without Dumbing Down"


Programme and on-line workshop notes

The workshop will take place at the:

White Room, Grand Hotel Toplice, Bled, Slovenia

All papers are in gzipped PostScript form. All papers will be available as bound hard-copy for those registered for the workshop. (Front matter for the workshop notes here.)


Session 1 : 0900-1040


Coffee break : 1040-1100


Session 2 : 1100-1240


Lunch : 1240-1400


Session 3 : 1400-1540


Coffee break : 1540-1600


Session 4 : 1600-1740


Programme committee

Christer Samuelsson (XRCE, Grenoble, France)
Claire Nédellec (Université Paris Sud (LRI), France)
Gosse Bouma (University of Groningen, Netherlands)
Gregor Erbach (DFKI, Saarbrücken, Germany)
Henrik Boström (University of Stockholm, Sweden)
James Cussens (University of York, UK)
Suresh Manandhar (University of York, UK)
Mark Steedman (University of Edinburgh, UK)
Stefan Wrobel (GMD, Germany)
Steve Pulman (University of Cambridge, UK)
Tomaz Erjavec (Jozef Stefan Institute, Slovenia)


Registration details

Registration for LLL is handled in the same way as registration for
the ICML-99 workshops. See the ICML-99 web page
http://www-ai.ijs.si/SasoDzeroski/ICML99/main.html for details.

The registration fee will be:

$45 if registered for ICML-99
$70 if not registered for ICML-99



 

Programme and local chair

PROGRAMME CHAIR
James Cussens
Department of Computer Science
University of York
Heslington, York YO10 5DD, UK
Email:  jc@cs.york.ac.uk
Phone:   +44 (0)1904 434732
Fax:     +44 (0)1904 432767

LOCAL CHAIR
Saso Dzeroski
Dept. of Intelligent Systems
Jozef Stefan Institute
Jamova 39
1000 Ljubljana, Slovenia
Email: Saso.Dzeroski@ijs.si
Phone: +386 61 177 3217
Fax: +386 61 125 1038


Acknowledgements

The organisers wish to acknowledge the help of the following in supporting this workshop: