DIMACS Center, CoRE Building, Rutgers University

**Organizers:****James Abello**, DIMACS, abello@dimacs.rutgers.edu**Graham Cormode**, DIMACS, graham@dimacs.rutgers.edu**Kenton Morgan**, University of Liverpool, k.l.morgan@liverpool.ac.uk**David Ozonoff**, Boston University, dozonoff@bu.edu

Presented under the auspices of the Special Focus on Computational and Mathematical Epidemiology.

Epidemiology is an observational science that concerns itself with finding and explaining patterns of health and disease in populations, usually of humans, but also populations of animals, insects and plants. Data mining is an active area of research interested in finding algorithms for describing latent patterns in often very large data sets. This Working Group has the objective of fostering collaboration between these two disciplines. In March of 2004 it is organizing a two-day meeting at DIMACS to bring these two groups together in a format designed to initiate such collaborations.

Meeting Format

Each day there will be a small number of presentations from epidemiologists, statisticians and computer scientists. They will give a detailed description of a specific study analyzed by methods current in today's epidemiological practice. The studies will be selected to represent a range of typical techniques such as exploratory factor analysis, multiple logistic regression, proportional hazards models, generalized additive models and other current methods. The epidemiologists will concentrate on the epidemiological aspects of the study, with some relatively brief explanations of the main ideas by the statisticians. Computer scientists or discrete mathematicians will comment on algorithmic issues and alternate approaches to the underlying questions. This will lead into a moderated audience-wide discussion of each study.

The goal is to identify some fruitful areas for applying new techniques from theoretical computer science, discrete mathematics, and statistics and to allow new collaborations among participants to be initiated. This workshop will bring together with epidemiologists experts interested in exploring data mining and algorithmic techniques potentially useful in epidemiology from a range of viewpoints. In order to facilitate a synthesis of perspectives, the workshop will be organized around half-day sessions, each with one or two talks followed by active discussion. The five primary topics of inquiry will be:

- Disease Dynamics, Epidemiological Studies and the Interpretation of Data
- Genetic, Pharmaceutical and Environmental Epidemiology
- Elements of Statistical Learning and Models
- Regression Methods, Support Vector Machines and Cluster Analysis
- Association Rules and Lattice/Graph Theoretical Methods

The workshop will also include a problem section and a report on the first Epidemiological DIMACS Challenge that is beginning on September 2003. Through short presentations and discussions that include all participants, the workshop aims to foster research cooperation among Epidemiologists, Computer Scientists and Mathematicians. The goal is to identify fundamental epidemiological problems that can benefit from efficient computational, statistical and mathematical models that can aid in the processing and understanding of combined epidemiological, and immunological data.

**This working group is by invitation only. **

Next: List of Participants

Workshop Index

DIMACS Homepage

Contacting the Center

Document last modified on June 18, 2003.