131
reads

Po Yao Niu

2014-01-17
15:00:00 - 15:30:00

308 , Mathematics Research Center Building (ori. New Math. Bldg.)

In everyday life, we rely on recommendations from other people either by word of mouth, recommendation letters, movies and book reviews in print. In this talk, we will emphasize on web applications including PChome, Youtube (Google) or large movie ratings dataset. The goal of recommender system found on webs is to match items from an inventory for each user visit in some context to optimize long-term business objectives. It often involves learning on constructing user profiles (feature construction) and finding right metric to optimize. The challenge is lacking enough data to learn all we want to learn and as quickly as we would like to learn. In this presentation, we will address basic concepts and methods. If time is allowed, we also present illustration with a few case studies.