Friday, April 8 | 11:00 AM - 12:00 PM
Washington University in St. Louis
Matrix factorization (MF) is exactly what it sounds like: the factorization of a matrix into a product of matrices. In Data Mining, implementations of MF like principal component analysis (PCA) have long been used for dimension reduction and exploratory analysis. More recently, MF has also become popular for training recommender systems, such as those used to produce movie suggestions to Netflix subscribers. In this talk, I'll first discuss the applications and differences of the PCA, ICA, and NMF methods. I'll then use the example of Netflix recommendations to introduce the challenge of applying MF to datasets with high sparsity, and how MF can be used to identify good recommendations by filling in missing values in the sparse data matrix.
If you are a member of the WashU community, login with your WUSTL Key to interact with events, personalize your calendar, and get recommendations.Login with WUSTL Key
If you are not a member of the WashU community, please login via one of the options below to interact with our calendar.