Personalization at Yahoo: From Mobile Data to User Preferences

Apr 23

Thursday, April 23, 2015

9:30 am - 10:30 am
Gross Hall 330


Andrew Gelfand, Yahoo

Abstract: Personalization is both a ubiquitous and unappreciated area of research. Everything at Yahoo, from the set of news stories displayed to you on to the order in which blogs are presented to you on tumblr, is tailored to meet individual user preferences. Yet most users are unaware that any such tailoring has even occurred. In this talk, I¿ll introduce the concept of personalization and describe the four main components of any personalization task. I¿ll then discuss some recent work on building profiles of user preferences from rich, mobile data collected by Yahoo Aviate, an intelligent launcher for Android devices. Bio: Andrew Gelfand is a Scientist in the Personalization, Search and Knowledge Group at Yahoo Labs. Andrew's research interests are in the area of machine learning and statistics, with a focus on inference, learning and decision making in graphical models. He received a Ph.D. in Computer Science from the University of California, Irvine, a masters degree in Operations Research from George Washington University and a B.S. from Tufts University. Prior to joining Yahoo, Andrew was a visiting scholar at Los Alamos National Lab and he also worked at Decisive Analytics, an applied research firm based in Washington, DC.