Thursday, September 21

MS1
Mining Scientific Data

10:30 AM-12:30 PM
Potomac

Rapid advances in high performance computing and the availability of massive storage systems to manage and assimilate data have given a significant impetus to data mining in the scientific sector. Data mining is now recognized as a key computational technology, supporting traditional analysis, visualization, and design tasks. The applications driving these developments also influence the basis, assumptions and methodological issues underlying the effective application of data mining. For example, diverse applications in mineral prospecting, computer aided design, bio-informatics and computational steering are being viewed in the data mining framework. This has led to a very effective cross-fertilization of computational techniques from both continuous and discrete perspectives. While a large number of conferences and workshops have been devoted to data mining and knowledge discovery, the problem of mining scientific data poses important research issues that will be of interest and relevance to a CS&E audience. The speakers in this minisymposium will provide an overview and discuss various aspects of mining scientific data.

10:30:-10:55 Mining Scientific Data: An Overview

Ananth Y. Grama, Organizer; and Naren Ramakrishnan, Virginia Polytechnic Institute and State University

11:00-11:25 Mining Scientific Simulation Datasets

Eui-Hong (Sam) Han, George Karypis, and Vipin Kumar, Army HPC Research Center and University of Minnesota, Minneapolis, USA; and Chandrika Kamath, Lawrence Livermore National Laboratory, USA

11:30-11:55 Extracting Comparative Strengths of Alternative Designs from Benchmarking Data

Raúl E. Valdés-Pérez, Carnegie Mellon University, USA

12:00-12:25 Spatial Aggregation in Scientific Data Mining

Christopher Bailey-Kellogg and Bruce Randall Donald, Dartmouth College, USA; and Feng Zhao, Xerox Palo Alto Research Center, USA