Title: The Search for Spatial and Spatio-Temporal Anomalies

Presenter: Dr. Michael D. Porter     Department of Statistics, North Carolina State University

Date and Time: Friday, January 11th, 2008 (Time TBA)

Abstract:
There is a growing need, across many disciplines, to develop methodologies for change detection in spatial and space-time processes. This talk discusses some approaches to anomaly/cluster detection (a specific type of change where the change occurs in a local region of space) in space-time point processes. The problem of detecting such changes is applicable in areas such as disease surveillance, forestry, habitat selection/animal movements, and crime and terrorism. For this problem, besides estimating the rate of change and change point, we must also estimate the anomalous spatial region. This last task is an influential, but often overlooked step in the detection process. This talk introduces some new test statistics and addresses the questions of how to search for the anomalous region, how to evaluate the detected regions, and the effect misspecification of the spatial change region has on the test performance.

Bio: Michael Porter is currently a post-doc in the Department of Statistics at North Carolina State University and SAMSI. He graduated in 2006 from Department of Systems and Information Engineering at University of Virginia. His research interests include point processes, anomaly detection, intelligent site selection, and criminology.
Seminar Location: The seminars are held at:
    Virginia Tech, Corporate Research Center
    1880 Pratt Drive, Building XV
    Seminar Room, First Floor
    Directions: Map (PDF)

Back to: NDSSL Seminar Page

Copyright Virginia Bioinformatics Institute © 2005