DIMACS Workshop on Mathematical Methods for High Performance Data Mining Applications

Tentative Program of Workshop:

Monday, April 27, 1998

8:30 Registration/Coffee

Session I
Chair: Helene Kulsrud, CCR-P

9:00 Welcome
     Julia Abrahams, Dimacs

9:10 The Year of Massive Data Sets
     Robert Sedgewick, Princeton University

9:20 Overview
     Helene Kulsrud, CCR-P

9:30  Finding and Exploiting Relationships in Large Text Collections
      Michael Littman, Duke

10:30 Coffee Break 

11:00 Clustering Speech Data by Language Using SVD
      Randall P. Dahlberg, DOD
   
12:00 Lunch

Session II
Chair: William Harrod, SIG/CRAY

1:00 Scaling Tree Based Classifiers to Massive Data Sets
     Robert Grossman, U of Illinois and Magnify

2:00 MDL Designs of Classification Trees
     Jorma Rissanen, IBM

3:00  Tea Break

3:30  Discovery of Patterns in Very Large Dimension Data Sets 
      using Hypergraph Models
      Vipin Kumar, University of Minnesota

4:30  Parallel Data Mining for Rules in Very Large Databases
      Mohammad Zaki, University of Rochester

5:30  Reception - Prospect House

Tuesday, April 28, 1998

Session III
Chair: Frank Brickle, CCR-P

9:00  Low-rank Approximations for Information Retrieval
      Samtosh Vempala, MIT 

10:00 Coffee Break

10:30 n-Gram Term Weighting: A Comparative Analysis
      Claudia Pearce, DOD

11:45 Combining Multiple Classification Trees
      William Shannon, Washington University

12:45 Lunch

Session IV
Chair: Robert Grossman, U. of Illinois and Magnify, Inc.

2:00  Bayesian Model Averaging:  An Introduction

3:00  Graphical Models and Variational Approximation
      Michael Jordan, MIT 

4:00  Tea Break

4:30  Wrap Up Discussion

Previous: Participation

Next: Registration

Workshop Index

DIMACS Homepage