Elisa Redmiles
DATE AND TIME CHANGE
Monday, Apr 9, 2018 at 12:00 PM in 380 Soda Hall
Title: User Demand for Spam and Dancing Pigs: Measuring Security Behavior
Abstract: Accurately modeling security behavior decision-making is critical to thinking about when, why, and how to improve end-user security. In this talk, I will present the results of two studies, which use different forms of measurement to model security-behavior choices: suggesting when and with which users it is most important, or most rational, to intervene. In the first study, we used behavioral economics experiments to define a mathematic model for end-user security decision-making including the rationality of that decision-making. Our results suggest a move away from "one-size-fits-all" emphasis on security toward cost-sensitive and context-aware personalized security recommendations. In the second study, we conducted a large-scale analysis of de-identified, aggregated user-spam interactions on Facebook (n=600,000) that allow us to predict user risk of falling for spam (e.g., propensity to click). Our results suggest that resources may be best utilized improving the security behavior (avoidance of spam clicking) for a subset of high-risk users, and suggest a number of avenues that may be effective.
Bio: Elissa Redmiles is a Ph.D. student at the University of Maryland in Computer Science. Her research focuses on using computational, economic, and social science methodologies to understand and mitigate digital threats to users' well-being. She focuses particularly on security and algorithmic threats and her research on these topics has appeared in Scientific American, Business Insider, and other popular press publications. Elissa frequently collaborates with the Max Planck Institute for Software Systems, the University of Zurich, and Facebook. She is the recipient of a NSF Graduate Research Fellowship, a National Science Defense and Engineering Graduate Fellowship, and a Facebook Fellowship. Prior to pursuing her Ph.D., Elissa held Marketing Management and Software Engineering roles at IBM and was a Data Science for Social Good Fellow at the University of Chicago.