The Third Annual
Journal of Information Technology & Politics Conference
May 16 & 17, 2011 – University of Washington - Seattle, WA

JITP 2011 Keynote Speaker: Jaime Teevan, Microsoft Research

Title: “Using Large Scale Log Analysis to Understand Human Behavior”

Abstract: In recent years, the rise of cloud-based services has made it possible to capture human interaction with computing systems on a scale previously unimaginable. These large-scale logs represent traces of human behavior, seen through the lenses of the tools people use. Unlike other methods for observing behavior, logs provide a complete picture of what people actually do for real task in real-world situations, including behaviors people might normally hide. Additionally, the scale at which log data can be collected gives a good picture of subtle behavior changes or unusual activities that could not possibly be observed by looking at only hundreds (or even thousands) of people. However, there are challenges to understanding log data. Logs show us what people do, but not why they do what they do. Little is known about the people who create the log data; not their age, not their gender, nor even whether events observed at different times or on different machines are made by the same person. Even less is known about the user’s motivations or success. Often apparently identical log data can be created with very different motivations; there are can even be adversarial motives for people to create traces in the log data that are not real. This talk will provide a practical overview of log analysis, with examples from what we have learned from real world Web search and browsing logs.

Join Now!
 

[Conference Home Page] [Conference Speakers & Authors] [Registered Participants]
[JITP.net] [Join the JITP Reviewer Database] [Browse the JITP archives]