"Using Context when Modeling User Behavior: Improving Fraud Detection, Neural Recommenders, and Fairness"
Alex Beutel, Google Research
Wednesday, November 15, 2017 at 2:00 P.M.
Room 368 (3rd Floor CIT)
How can we find fraudulent behavior in social networks? Can we model a user's preferences and background to predict what video or topic a user is interested in? How do we make these models work well for everyone and act fairly?
In this talk we will focus on understanding and modeling user behavior. Online, users interact not just with each other in social networks, but also with the world around them – supporting politicians, watching movies, buying clothing, searching for restaurants and finding doctors. These interactions often include insightful contextual information as attributes, such as the time of the interaction and ratings or reviews about the interaction. We demonstrate that by modeling how fraudsters work, we can more effectively detect them, catching previously undetected fraud on Facebook, Twitter, Tencent Weibo and Flipkart. Next, we will explore how neural models can be designed to make use of context for state-of-the-art recommendation, such as modeling the evolution of user preferences over time. Finally, we will explore how these models can be biased, and how we can decrease or remove these biases so that the models work well for everyone.
Alex Beutel is a Senior Research Scientist at Google Research working on neural user behavior modeling and fairness in machine learning. He received his Ph.D. in 2016 from Carnegie Mellon University's Computer Science Department, and previously received his B.S. from Duke University in computer science and physics. His Ph.D. thesis on large-scale user behavior modeling, covering recommender systems, fraud detection, and scalable machine learning, was given the SIGKDD 2017 Doctoral Dissertation Award Runner-Up. His work has appeared in KDD, WWW, ICDM, SDM, AISTATS, and TKDD; and he has given tutorials on user behavior modeling at KDD, one of the premier data mining conferences, and CCS, one of the premier security conferences. He received the Best Paper Award at KDD 2016 and ACM GIS 2010, was a finalist for best paper in KDD 2014 and ASONAM 2012, and was awarded the Facebook Fellowship in 2013 and the NSF Graduate Research Fellowship in 2011. More details can be found at http://alexbeutel.com
Host: Professor Ugur Cetintemel