home · mobile · calendar · colloquia · 2000-2001 · 

Colloquium - Bay

Detecting Differences Between Groups
Stephen Bay
University of California, Irvine

I will discuss the problem of detecting and describing the differences between several contrasting groups from observational data. These groups can be different classes of objects, such as male or female students, or the same group over time, e.g., freshman students in 1993 through 1998. My goal is to find contrast sets: conjunctions of attributes and values that differ meaningfully in their probability distribution across the groups. The main research issues are dealing with the enormous search space, controlling the search error to limit false discoveries, and summarizing the results for an end user. Throughout the talk, I will present examples based on an analysis of student record data from UC Irvine. I will also discuss some applications for a tool that finds contrast sets in characterizing the errors a classifier makes, multivariate discretization, and spatial clustering.

Hosted by Michael Mozer.
Refreshments will be served immediately following the talk in ECOT 831.

Department of Computer Science
University of Colorado Boulder
Boulder, CO 80309-0430 USA
May 5, 2012 (14:13)