Java implementation of Matrix factorization model for recommendation
Matrix Factorization uses latent model for recommendation. Basically, it is the process of completing the empty entries in the matrix, where the matrix is usualy very sparse. Each enty in the matrix represents the interaction between user and item pair, usually denotes the rating given by user.
The code implements the matrix factorization techniques proposed in the paper: http://www2.research.att.com/~volinsky/papers/ieeecomputer.pdf