[Corpora-List] Learning Language in Logic: 2nd CFP

James Cussens jc at cs.york.ac.uk
Thu Mar 24 17:16:00 CET 2005


The 4th International Workshop
Learning Language in Logic (LLL05)


Special Theme: Extracting Relations from Bio-medical Texts

Bonn, Germany
7 August 2005

Hosted by ICML 2005,
The 22nd International Conference on Machine Learning


This workshop is the fourth Learning Language in Logic (LLL) workshop
following previous LLL workshops in Bled (99), Lisbon (00) and
Strasbourg (01). The purpose of the workshop is to provide a focus for
work which applies machine learning using logical representations to
natural language. Much existing work in this area uses techniques from
inductive logic programming (ILP), and increasingly statistical
relational learning. More details on the LLL state-of-the-art are
available from the LLL home page ( http://www.cs.york.ac.uk/aig/lll/ )
which includes links to many LLL papers, including introductory ones.
We encourage submission of papers on any aspect of logic-based machine
learning for natural language learning.

One of the principal motivations for proposing such a workshop now is
the increasing interest in applying logic-based machine learning to
the problem of Information Extraction from bio-medical texts. For this
reason the workshop has an associated "challenge task" where the aim
is to learn IE rules which can be used to extract information on
gene/protein interactions. Information on the challenge task is given
below. Although, bio-medical IE is a special theme of the workshop, we
very much encourage good quality articles on any aspect of logic-based
machine learning for natural language learning. Here is a
non-exhaustive list of possible topics:

* Explanation-based learning
* Extracting predicate-argument structure
* Grammar learning
* Higher-order logic for LLL
* Handling very complex terms in ILP
* Information extraction
* Instance-based approaches in ILP
* Logical approaches to statistical NLP
* Learning subcategorisation frames
* Morphological analysis
* Multi-predicate learning
* POS and morphosyntactic tagging
* Semantic role labelling
* Shallow parsing
* Theory revision
* Using ILP with large data sets
* Word sense disambiguation



The LLL05 challenge task is to learn rules to extract gene/protein
interactions from bio-medical abstracts. The challenge task team at
MIG-INRA are providing training data containing the following information:

- annotation indicating agent and target of an interaction
- dictionary of named entities (including variants and synonyms)
- linguistic information: word segmentation, lemmatisation, syntactic
categories and syntactic dependencies

This data and full details of the LLL05
bio-medical challenge task are available from the challenge task
web page ( http://genome.jouy.inra.fr/texte/LLLchallenge ).

Results for this task are to be reported by the submission of a 4 page

Challenge task schedule:

11 February 2005 Initial version of the data made available
3 March 2005 Final version of the data made available
27 April 2005 Submission of papers with results
11 May 2005 Challenge task papers notification



Deadline: 1 April 2005 (23:59:59 GMT)

Submissions should not exceed 8 pages, and should be in line with the
ICML formatting guidelines given on their website (
http://icml2005.kdnet.org/papers/format/ ), except that there is no
need to make your submission anonymous. Simply email your submission
as PDF to James Cussens at the following (deliberately mangled) email
address: "firstname.lastname AT cs DOT york DOT ac DOT uk".

For challenge task submission instructions go to:



James Cussens, University of York, UK (workshop)
Claire Nédellec, INRA, France (challenge task)


* Zoltán Alexin, University of Szeged, Hungary
* Érick Alphonse, INRA, France
* Mary Elaine Califf, Illinois State University, USA
* Vincent Claveau, Université de Montréal, Canada
* James Cussens, University of York, UK
* Walter Daelemans, University of Antwerp, Belgium
* Saso Dzeroski, Jozef Stefan Institute, Slovenia
* Tomaz Erjavec, Jozef Stefan Institute, Slovenia
* Dimitar Kazakov, University of York, UK
* Suresh Manandhar, University of York, UK
* Claire Nédellec, INRA, France
* Lubos Popelínský, Masaryk University, Czech Republic
* Stephen Pulman, University of Oxford, UK
* Dan Roth, University of Illinois at Urbana-Champaign, USA
* Pascale Sébillot, IRISA, France
* Jude Shavlik, University of Wisconsin-Madison, USA



1 April 2005 Normal paper submission deadline
22 April 2005 Normal paper notification
27 April 2005 Challenge task paper submission deadline
11 May 2005 Challenge task paper notification
13 May 2005 Normal paper (final version) deadline
30 May 2005 Workshop notes available on-line


James Cussens jc at cs.york.ac.uk
Department of Computer Science Tel +44 (0)1904 434732
University of York Fax +44 (0)1904 432767
Heslington, York YO10 5DD, UK http://www.cs.york.ac.uk/~jc

More information about the Corpora-archive mailing list