CONALD, June 11-13 Conference on Automated Learning and Discovery
General Information Submission Instructions Registration Workshops Travel and Accommodation Committees
Plenary Speakers

Tom Dietterich

Stuart Geman

David Heckerman

Michael Jordan

Daryl Pregibon

Herb Simon

Robert Tibshirani


Daryl Pregibon
Realtime Learning and Discovery in Large Scale Networks


In many financial and communications industries, data from networks are becoming increasingly important. Two aspects of network data that are particularly challenging concern the size of the data stream and it's timeliness. Indeed we have been concerned with industrial strength applications that consist of hundreds of millions of transactions (i.e., packets, conversations) that need to be analyzed, summarized and acted upon, within 24 hours. Furthermore some applications (e.g., fraud and network intrusion detection) require real-time analysis and intervention.

We will share our experiences in these applications, stressing statistical and computational issues. Several examples of web-based delivery systems will be presented that convey the scope and scale of what is possible today --- and what might be coming tomorrow. [Joint work with RA Becker, C Cortes, and AR Wilks]


Daryl Pregibon is Head of Statistics Research at AT&T Labs. His group is focused on providing the computational and theoretical foundation for the application of statistics to very large data sets. He is Past-Chair of the premier datamining conference (KDD) and remains the Chair of the Committee on Applied and Theoretical Statistics at the National Academy of Science.
More Information

Contact conald@cs.cmu.edu for more information

The conference is sponsored by CMU's newly created Center for Automated Learning and Discovery.