MLJ special issue on Multi-relational Data Mining and Statistical Relational
Learning - deadlines 6/30 (abstract), 7/31 (paper)
See also http://www.cs.kuleuven.ac.be/~ml/mrdm-srl.html
****************************************
*** We apologize for multiple copies ***
****************************************
Machine Learning Journal Special Issue on
Multi-relational Data Mining and
Statistical Relational Learning
Call for Papers
AIM AND SCOPE:
==============
There has been an increasing interest in relational learning and relational
data mining during the last few years, as witnessed by, for instance, the
IJCAI-2003 Workshop on Learning Statistical Models from Relational Data and
the KDD-2003 Workshop on Multi-Relational Data Mining. Following the success
of these workshops, the Machine Learning Journal invites authors to submit
papers for a Special Issue on Multi-Relational Data Mining and Statistical
Relational Learning.
Authors are invited to submit papers presenting original results on all
aspects of relational learning, including but not limited to:
- Foundational aspects: types of problems (linked data vs. structured data),
semantics of probabilistic relational models, etc.
- Representation aspects: representational power of different formalisms
(PRMs, SLPs, BLPs, etc.) and data representations (relational databases,
multi-instance, graphs, logic, structured and semi-structured representations,
probabilistic representations), restricted relational pattern languages
enabling efficient mining
- Inference and learning tasks for relational data (e.g., attribute
prediction, link prediction, link analysis and discovery, consolidation,
entity detection, object identification and clustering)
- Methods for learning from relational or structured data: distance and kernel
based methods, ensemble methods, probabilistic and statistical approaches,
relational neural networks
- Algorithmic aspects: tractability, efficiency, scalability (e.g., of
learning in probabilistic relational representations)
- Methods of incorporating background knowledge
- Integration of data mining features into relational and other databases
- Inductive databases for mining structured and semi-structured data
- Propositionalization methods for transforming (multi-)relational data mining
problems to single-table data mining problems
- Integrating information from heterogeneous sources by means of
multi-relational data mining and statistical relational learning
- Evaluation and validation issues in multi-relational databases and
statistical relational learning
- Applications of relational learning: e.g. to computational chemistry,
bioinformatics, predictive toxicology and toxicogenomics, environmental
sciences, medical informatics, computational linguistics, relational text and
web mining, information retrieval, spatial data mining, time series data
mining, music data mining, social networks analysis, security and law
enforcement, etc.
This special issue will follow a special issue on Inductive Logic Programming.
As a rule, we encourage researchers whose contribution lies entirely within
the field of inductive logic programming to submit to that issue.
Contributors to SRL-2003 and MRDM-2003 are in particular invited to submit a
paper, but the special issue is open to everyone. Each submission will be
reviewed according to the standards of the Machine Learning Journal.
IMPORTANT DATES:
================
Titles and short abstracts due: June 30, 2004
Papers due: July 31, 2004
Author notification: November 30, 2004
Final versions of accepted papers due: January 31, 2005
Publication: Mid 2005
GUEST EDITORS:
==============
Hendrik Blockeel (Katholieke Universiteit Leuven, Belgium)
David Jensen (University of Massachusetts, USA)
Stefan Kramer (Technische Universitaet Muenchen, Germany)
PAPERS PREVIOUSLY PUBLISHED IN CONFERENCES/WORKSHOPS:
=====================================================
Authors of papers that have appeared previously in refereed conferences and
workshops are encouraged to submit extended versions of their papers. Such
extended papers must be significantly different from the conference version,
as well as accessible to the broad readership of the journal. They may expand
on the material that was included in the original paper, e.g., by providing
more details, giving a more in-depth discussion of the results and related
work, expanding upon the experimental results, or giving a more thorough and
scholarly treatment of the material than was possible in a conference paper.
Submissions must not have appeared in, nor be under consideration by,
other journals. Authors of papers whose previous versions appeared in
refereed conferences and workshops are requested to provide the previously
published version of their papers, as well as to include in their submission
a brief letter stating the differences between the prior published version
and this MLJ Special Issue submission.
SUBMISSION INFORMATION:
=======================
Only electronic submissions will be accepted. Instructions for submission
can be found at
http://www.kluweronline.com/issn/0885-6125
In the text of your electronic submission, please explicitly state that the
paper is for the special issue on Multi-Relational Data Mining and Statistical
Relational Learning. In addition to submitting the paper to
[EMAIL PROTECTED],
please also submit to the guest editors:
[EMAIL PROTECTED]
[EMAIL PROTECTED]
[EMAIL PROTECTED]
All inquiries regarding this special issue should be directed to the guest
editors.
