CFP

"Theoretical Advances in Data Clustering" Special Issue of Machine Learning journal

Guest Editors:

Nina Mishra, HP Labs/Stanford University
Rajeev Motwani, Stanford University

Data clustering, the problem of grouping similar objects together, has attracted significant practical attention due to modern applications like web document clustering, click stream analysis, multimedia applications, and telecommunication records. The field has experienced a corresponding theoretical surge due to recent algorithmic discoveries and novel methods of modeling the clustering problem. In this special issue, we seek papers that document theoretical advances related to data clustering. Topics of interest include, but are not limited to:

• approximation/randomized algorithms
• graph-based clustering
• spectral clustering
• dimension reduction techniques
• dense region identification
• Bayesian clustering
• statistical/machine-learning theory
• testing clusterability
• clustering large datasets
• clustering dynamic datasets, like data streams
• mixture modeling
• conceptual clustering Rigorous papers supported with empirical case studies are also welcome.

Submissions must not have appeared in, nor be under consideration by other journals. Exceptional surveys will be considered.

Except in extraordinary circumstances, submissions should not exceed 35 pages (journal-formatted pages), excluding abstract/references.

Submission Process

Only electronic submissions will be accepted. Instructions for submission can be found at http://www.cs.ualberta.ca/~holte/mlj/initialsubmission.pdf In the text of your electronic submission, please explicitly state that the paper is for this special issue. In addition to submitting the paper to jml@wkap.com, please also submit to nmishra@hpl.hp.com.
Important Dates

Dec 6, 2002 Abstract submission.

Dec 13, 2002 Full submissions should be received.

May 27, 2003 Decisions sent to authors; papers accepted
with no more than minor revisions accepted to
the special issue.

Dec 27, 2003 Final versions of accepted papers should be received
in the format specified for full submissions, using
Kluwer style guidelines.

An online version of this call for papers can be found at:

http://www.hpl.hp.com/personal/Nina_Mishra/MLJ-clustering.html