Balancing Discovery & Continuation in Recommendations
When a member logs on to Netflix, she may be in one or a combination of different watching modes: discovering a new content, continuing to watch a partially-watched movie or a TV show, playing one of the contents she had put in her play list during an earlier session, etc. If, for example, we can reasonably predict when a member is more likely to be in the continuation mode, and which videos she is more likely to resume, it makes sense to place those videos in more prominent places of the home page. In this talk we focus on understanding the discovery vs. continuation behavior and explain how we have used machine learning to improve the member experience by learning a personalized balance between those two modes.
Hossein Taghavi received the B.Sc. degree from Sharif University of Technology, Tehran, Iran, in 2003, and the Ph.D. degree from the University of California at San Diego, La Jolla, in 2008, in electrical engineering. From 2008 to 2011, he was with Qualcomm, Inc. in San Diego, CA, conducting research and development on algorithms and systems for wireless networks. From 2011 to 2013, he was with Opera Solutions, LLC., in San Diego, CA, applying machine learning and data analytics to a variety of business applications. Since 2013, he has been with Netflix, Inc., contributing to algorithms and software driving the personalization and recommendation systems for Netflix members. His current interests include machine learning, recommendation systems, and distributed computing.