Date: Wednesday, December 30, 1998 9:51 AM
From: Hannu Toivonen htt@rni.helsinki.fi
Subject: Workshop: Euro-Par'99, Toulouse, France, Aug 31 - Sep 3, 1999
Web: http://www.enseeiht.fr/europar99/programme/ws22.html
Call for papers
Euro-Par'99
Toulouse, France
August 31 - September 3, 1999
Topic 22: High-Performance Data Mining and Knowledge Discovery
Topic Committee :
Global chair : David Skillicorn (Queen's University, Canada)
Vice-chair : Vipin Kumar (University of Minnesota, USA)
Vice-chair : Hannu Toivonen (University of Helsinki, Finland)
Local chair : Domenico Talia (ISI-CNR, Rende, Italy)
Description :
-------------
Data mining extracts conceptual information from raw data, often originally
collected for some other purpose. Both the size of the data available
and the complexity of the algorithms required for concept extraction
suggest that high-performance computing has an important role to play.
Indeed, high-performance and parallel versions of standard data
mining techniques are beginning to mature; and novel approaches that
presuppose parallelism are also being investigated.
High-performance data mining work has evolved from two directions: users
whose performance needs have outgrown uniprocessor platforms, and
parallel computing researchers who perceive an application domain
with long-term potential. This is a good moment to explore
cross-fertilisation between these two communities.
Topics of interest include:
---------------------------
Parallel algorithms for data mining
Scalability of existing algorithms
Principled approaches to sampling
Communication requirements for parallel data mining
High-performance presentation of results (winnowing, visualization)
Data representation and storage for fast acces
Data mining in text (e.g. parallel latent semantic indexing)
High performance data warehousing and OLAP
For more details see
http://www.enseeiht.fr/europar99/
or e-mail to:
europar99@enseeiht.fr.
- January 31st 1999 : Final Date for Submissions
Copyright © 1999 KDnuggets