Parallel Data Mining Workshop 2011

• 13:30-14:30 Invited talk by Pr. Gagan Agrawal
• 14:30-15:00 Aerial Root Classifers for Predicting Missing Values in Data Stream Decision Tree Classifcation, Yang Hang, Simon Fong, Wei Chen
• 15:00-15:30 Coffee break
• 15:30-16:00 Monitoring Multiple Streams with Dynamic Time Warping using Graphic Processors, Jason Chang, Mi-Yen Yeh
• 16:00-16:30 Supporting Dynamic Load Balancing in a Parallel Data Mining Middleware, Tekin Bicer, Gagan Agrawal

The workshop notes are available in PDF

Today is an exciting time to be a data mining researcher. There is a huge quantity of available data, and a high demand for knowledge extraction from this data that only data mining can provide. Moreover, with the democratization of parallel computing solutions, be it multicore processors, GPUs, clusters or grids, there has never been more raw computing power available.

There is a catch however: parallel computing power does not come for free. Designing efficient parallel algorithms requires new skills, many of which are yet to discover. Most current programming language does not make parallel programming easy, leading to new research on different ways of programming. Then performance analysis, which is a crucial point for data mining algorithms, get much more complex, as the scaling with the number of computing elements must also be studied. Bad performance can be caused by an inefficient usage of the underlying parallel system architecture. These problems quickly arise as data mining algorithms often need to perform complex computations on huge volumes of data, with irregular and unpredictable computing loads, and generating large quantities of intermediate data and/or results. This pushes the parallel architectures to their limits, forcing data mining researchers to get a precise understanding of how these architectures work and to rethink existing algorithms.

The PDM workshop provides a venue for all data mining researchers working on parallel data mining, whatever the parallel support used. Despite differences in support, the main problems of parallelism remain the same, discussions between researchers working on different parallel supports or different domains of data mining will allow cross-fertilization and help everyone to progress.

Submissions of early research and innovative ideas are highly encouraged.

Researchers applying parallel data mining to real-world data and applications are also most welcome.

• New parallel data mining algorithms, especially in pattern mining, classification, clustering
• Multicore CPU data mining
• GPU data mining
• Cluster-based or grid-based distributed data mining
• Cloud data mining
• Peer-to-peer data mining
• Performance analysis of parallel data mining algorithms
• Theoretical foundations of parallel data mining
• Cache-aware / Cache-oblivious data mining algorithms
• Parallel data-mining over streams
• Parallel implementations of data mining algorithms with programming languages well adapted to parallelism, for example functional languages
• Applications using parallel data mining in either an industrial or a scientific context

Please upload your paper in PDF format at: SUBMIT HERE

The paper should have at most 10 pages and use the SIAM style available at: http://www.siam.org/tex/books/soda2e.all.

More details on SIAM paper formatting can be found here. The paper must clearly present author information including name, affiliation and email.

Each paper will receive at least two reviews to determine acceptance.

Attendees are required to register for SDM 2010 so that no separate registration is needed for this workshop. A one-day registration for the conference is available. Workshop attendees do not have to register at the complete conference rate. Click here for more details.

Chair: Alexandre Termier (Grenoble University, France)

• Srinivasan Parthasarathy (Ohio State University, USA)
• Georges Karypis (University of Minnesota, USA)
• Shirish Tatikonda (IBM Almaden, USA)
• Geoffrey C. Fox (Indiana University, USA)
• Anne Laurent (University of Montpellier 2, France)
• Jean-Francois Méhaut (Grenoble University, France)
• Benjamin Négrevergne (Grenoble University, France)
• Claudio Lucchese (ISTI-CNR Pisa, Italy)
• Nicolas Hanusse (Bordeaux University, France)
• Sadok Ben Yahia (Tunis-El Manar University, Tunisia)
• Mario Rosario Guarracino (ICAR-CNR, Italy)
• Domenico Talia (University of Calabria, Italy)
• Maurice Tchuente (University of Yaoundé 1, Cameroon)

Alexandre Termier
first name [dot] last name [at] imag.fr
LIG (Laboratoire d'Informatique de Grenoble)
Université Joseph Fourier
681 rue de la Passerelle
B.P. 72, 38402 Saint Martin d'Hères
FRANCE
Phone: +33 4 76 82 72 07
Fax: +33 4 76 82 72 87
http://membres-liglab.imag.fr/termier/