• S. Parthasarathy 

  • Ohio State University
    ( )
  • H. Kargupta 

  • University of Maryland (Baltimore County Campus)  
  • V. Kumar 

  • University of Minnesota 
  • D. Skillicorn

  • Queens University, Canada 
  • M.J. Zaki

  • Rensselaer Polytechnic Institute


  • G. Agrawal , Ohio State University  
  • E. Bertino, DSI, University of Milan, Italy 
  • D. Cheung, University of Hong Kong, Hong Kong 
  • A. Choudhary, Northwestern University, USA 
  • J. Gehrke, Cornell University, USA 
  • R. Grossman, University of Illinois-Chicago, USA  
  • Ron Musick, iKuni Inc., USA  
  • G. Ostrouchov, Oak Ridge National Labs, USA  
  • S. Orlando, University of Venice, Italy  
  • B. H. Park , Oak Ridge National Labs, USA  
  • Y. Pan, Georgia State University, USA  
  • R. Rastogi , Lucent Technologies, USA  
  • P. Scheuermann, Northwestern University  
  • K. Sivakumar, Washington State University, USA  
  • D. Talia, DEIS, University of Calabria, Italy  
  • G. Williams, CSIRO, Aust. Nat. Univ., Australia
  • Additional Reviewers:

  • B. Rutt, Ohio State University
  • A. Ghoting, Ohio State University
  • HPDM: High Performance, Pervasive, and Data Stream Mining

    6th International Workshop on High Performance Data Mining:
    Pervasive and Data Stream Mining (HPDM:PDS'03)
    May 2003
    Program Details

    in conjunction with

    Third International SIAM Conference on Data Mining

    Workshop History: This is the 6th workshop on this theme held annually. The first four held in conjunction with IPDPS were held at Orlando ( HPDM'98), San Juan ( HPDM'99), Cancun (HPDM'00). and San Francisco (PDDM01). Last years workshop was held along-side the SIAM conference with a special focus on mobile and location-aware data mining issues (HPDM:RLM). This years workshop while continuing its focus on traditional areas of high performance data mining, also embraces the new areas of pervasive and datastream mining and management.

    This workshop will focus on the emerging field of high performance, pervasive, and data stream mining. Over the years the definition of high performance computing has taken on various forms as a function of the types of technical and creative uses and the underlying semantics of the applications driving them. Traditional definitions often refer to the problem of using high end parallel computers to meet the need of scientific applications. However, high performance computing can also include the need for fast sequential algorithms that target memory and I/O performance. The last decade has seen the growth and importance of grid computing where resources and data are physically distributed. This has led to the development of high performance distributed algorithms over the computational grid.

    More recently we have seen the advent of mobile and distributed computing environments and sensor networks that offer a completely new set of data mining applications. Accessing and analyzing data in a ubiquitous environment offer many challenges. They include distributed resource aware algorithms, limited access to possibly privacy sensitive data, novel challenges in human-computer interaction, and algorithms for handling data streams. Topics of interest include but are not limited to:

    • Resource and location-aware mining algorithms.
    • Data mining in mobile environments.
    • Grid-based data mining and management.
    • Theoretical foundations of resource-aware data mining.
    • Systems support for resource and location aware data mining.
    • Data stream mining and management.
    • Efficient, scalable, disk-based, parallel and distributed algorithms for large-scale data mining and pre-procesing and post-processing tasks.
    • Parallel or distributed techniques for incremental, exploratory and interactive mining.
    • Frameworks for KDD systems, and parallel or distributed mining.
    • Applications of parallel and distributed data mining (PDDM) in business, science, engineering, medicine, and other disciplines.
    • Theoretical foundation of PDDM.

    The organizers plan to have one invited talk, preferably by an exponent in the areas of special interest (mobile data mining, mining over the information grid, or mining over streaming datasets). The organizers plan to have one panel on a topic to be determined. Panelists will include some of the top people working in the area of high performance data mining. In addition it is anticipated that there will be two contributed sessions and one invited session on various topics of interest to the workshop.

      Important Dates:
    • Paper Submissions due (deadline extended):
      February 10 2003 by 8AM EST.
    • Notification to authors:
      March 10 2003.
    • Final papers due:
      March 30 2003.

    Submission Information: Authors are instructed to submit articles that meet the following criteria:

    • No more than 20 double spaced pages excluding bibliography.
    • At least 11pt. font.
    • One title page (not included in the above 20) containing:
      • the title
      • names and affiliations of all authors
      • contact author information
      • abstract of no more than 100 words
    • PS or PDF format only.
    Papers that do not meet the above criteria may not be reviewed. You can submit by emailing the PS or PDF file to . Use the subject header "HPDM:PDS03:submission" when you email the submission. Hardcopy submissions may be sent to:
      Dr. Srinivasan Parthasarathy
      DL 395, 2015 Neil Ave.,
      Ohio State University,
      Columbus Ohio-43210
    Please be aware that hardcopy submissions must arrive at the above address by the due date in order to be considered by the program committee.

    Additional Information:

    • Following the tradition in previous workshops in addition to contributed papers there will be one or two invited talks and/or a panel discusssion relevant to the theme of the workshop.
    • Program committee members will be instructed to consider work-in-progress papers with novel ideas and such papers are welcomed by the co-chairs.

    Maintained by: Srinivasan Parthasarathy <>