Friday
Organizers:
James Abello, DIMACS, [email protected]
Graham Cormode, Bell Laboratories, [email protected]
Presented under the auspices of the Special Focus on Computational and Mathematical Epidemiology.
*********************************************************************
Data Mining is now a staple part of Computer Science, and has been applied in a wide variety of different areas. It covers a diverse set of topics from algorithms, statistics and discrete mathematics, with the general goal of identifying patterns in data in order to draw inferences and make predictions. This tutorial brings together experts from Data Mining to introduce the key ideas and techniques from:
* Probability, Decision Trees and Bayesian Statistics
http://www.cs.cmu.edu/~awm/
* Machine Learning, Classifiers and Boosting
http://www.cs.princeton.edu/~schapire/
* Data Stream Analysis and Clustering
http://dimacs.rutgers.edu/~graham/pubs/epidcluster.pdf
* Graph Mining http://www.mgvis.com
* Applications to Biology and Epidemiology
The goal is to allow people with little or no knowledge of data mining to understand the basic techniques, and get a flavor of the general methodology and style of results. This tutorial is aimed to be of interest to researchers wishing to work in data mining, and also to researchers from outside computer science who wish to understand these methods in order to apply them. The tutorial includes short talks on applications to problems in epidemiology and biology in order to put the general techniques described into perspective.
****************************************************************
Workshop Program:
Thursday, March 23, 2006
Probability and Machine Learning Tutorials
8:15 - 9:00 Breakfast and Registration
9:00 - 9:15 Introductory Remarks
Fred Roberts, DIMACS Director
9:15 - 10:15 Selected Problems in Epidemiology
Nina Feffermann, DIMACS and Tufts University
The Mathematical Formulation of the Foot-and-mouth Disease Epidemic
Component of the Decision Support System Developed at LLNL
Tanya Kostova, LLNL
10:15 - 10:30 Break
10:30 - 1:00 Probability for Data Miners Tutorial,
Brigham Anderson, Carnegie Mellon University
1:00 - 2:00 Lunch
2:00 - 4:30 Machine Learning Tutorial
Rob Schapire, Princeton University
4:30 - 4:45 Break
4:45 - 5:45 Contributed Talks
5:45 - 7:45 Dinner and Reception
Friday, March 24, 2006
Data Streaming and Graph Mining Tutorials
8:15 - 9:00 Breakfast and Registration
9:00 - 9:15 Introductory Remarks
9:15 - 11:45 Data Streaming and Clustering Tutorial
Graham Cormode, Bell Labs
11:45 - 1:00 Lunch
1:00 - 3:15 Graph Mining Tutorial
James Abello, DIMACS and Ask-Research
3:15 - 3:30 Break
3:30 - 5:00 Closing Section - To what extent are the presented data mining
techniques useful for epidemiological research?
Moderators: James Abello, Graham Cormode, Ni