The Internet Ordered For You
Personalization of user experiences is a key differentiator of Yahoo!'s content strategy. How does one personalize the browsing experience of millions of users world-wide? How are user's interests inferred from their browsing habits? How does a personalization model adapt to changes in user interests? At Yahoo!, we take a machine learned approach to model users’ interests and surface relevant content. In this talk, we will provide an overview of the different components that went into building a web-scale personalization system and the machine learning challenges therein. We shall draw parallels between a personalization system and other web-mining areas, specifically content enrichment, information retrieval and recommendation systems. At the end of this presentation, we hope to motivate those in audience with an interest in machine learning to think about how he/she would go about solving the problem of personalization at scale.
Suju Rajan is a Senior Manager with the Personalization and Content Science team at Yahoo! Labs. At Yahoo!, she works on personalizing user experience, measured by the relevance and timeliness of content surfaced to the users. Her research interests are in content enrichment, user modeling and information retrieval. Dr. Rajan received her PhD from the University of Texas at Austin, focusing on semi-supervised and active learning based classification for dynamic environments.