Fall Seminar Series - December 11, 2003
University of Minnesota
School of Statistics
College of Liberal Arts
Sex, Lies, and Video Disks: Statistical Lessons of a Computer
Scientist
Joseph A.
Konstan
Department of Computer Science and Engineering
University of Minnesota
Thursday,
December 11, 2003
4:00 PM,
115
Ford Hall
Minneapolis,
East Bank Campus
Social
at 3:30 PM, 300
Ford Hall
Abstract
This talk covers two very different Internet-based research projects.
In the first part, I discuss a 1000-subject Internet sex survey of Latino
gay men to draw lessons about the design, deployment, and reliability of
such surveys. Of particular note is the surprising finding that more than
10% of our responses were duplicate responses submitted to receive multiple
payments -- indeed, over 5% of our responses came from a single individual!
The second part of the talk covers collaborative filtering -- a set of statistical
algorithms for producing personal recommendations out of a sparse matrix
of opinions. I will briefly present the MovieLens system -- a research system
we've been using for over six years -- and discuss recent results and future
research directions.
Joseph A. Konstan is Associate Professor of Computer Science and Engineering
at the University of Minnesota. His research addresses a variety of human-computer
interaction issues related to filtering, comprehending, organizing, and automating
large and complex data sets. He is probably best known for his work in collaborative
filtering (the GroupLens project).
Dr. Konstan received his Ph.D. from the University of California, Berkeley
in 1993. He is an ACM Distinguished Lecturer and IEEE Distinguished Visitor.
He currently serves as Chair of ACM SIGCHI -- the 5000-member Special Interest
Group on Computer-Human Interaction.