Personalized Prediction and Recommender Systems
Xiaotong Shen, School of Statistics, University of Minnesota
Personalized prediction predicts a user's preference for a large number of items through user-specific as well as content-specific information, based on a very small amount of observed preference scores. The problem of this kind involves unknown parameters of high-dimensionality in the presence of a very high percentage of missing observations. In this situation, the predictive accuracy depends on how to pool the information from similar users and items. Two major approaches are collaborative filtering and content-based filtering. Whereas the former utilizes the information on users that think alike for a specific item, the latter acts on characteristics of the items that a user prefers, on which two kinds of recommender systems Grooveshark and Pandora are built. In this talk, I will present our recent developments in regularized latent-factor modeling and scalable computation based on a ``divide-and-conquer'' strategy. Special attention will be devoted to the impacts of missing patterns and user-specific social networks on personalized prediction.
Note: Seminars are free and open to the public. Reception to follow.