DIMACS Workshop on Mathematical Methods for High Performance Data Mining Applications

April 27 - 28, 1998
Convocation Room, E-Quad, Olden Street, Princeton University

Helene E. Kulsrud, CCR-P/Institute for Defense Analyses, laney@ccr-p.ida.org
Robert Grossman, University of Illinois at Chicago and Magnify, Inc., grossman@uic.edu
Presented under the auspices of the DIMACS Special Year on Massive Data Sets

Data Mining is the automatic discovery of patterns, associations, changes, and anomalies in data sets. Scaling data mining to massive data sets requires new algorithms and combining techniques from high performance computing and data management with techniques from statistics and machine learning. This workshop will emphasize techniques to scale and parallelize algorithms. Of particular interest is the application of methods to specific types of problems and the areas of success and failure. Some techniques of interest are: tree-based statistical methods, graphical models, linear algebra, neural nets, combinatorial methods, meta learning, model selection and model averaging, and applications of data mining to information retrieval.

Call for Participation:
Presentations are solicited on the various mathematical methods and the applications to which they have been applied. There will be a small amount of support available for participants. If you would like to give a talk, please send an abstract to laney@ccr-p.ida.org.

On line registration is encouraged or send email including your name, institutional affiliation, email address, and the dates you plan to attend to Ms. Sandy Barbu at barbu@cs.princeton.edu

Next: Call for Participation
Workshop Index
DIMACS Homepage
Contacting the Center
Document last modified on April 2, 1998.