公开(公告)号：US20150088953A1

公开(公告)日：2015-03-26

申请号：US14493308

申请日：2014-09-22

Applicant: Infosys Limited

Inventor： Hari Manassery Koduvely ,  Sarbendu Guha ,  Arun Yadav ,  David C. Gladbin ,  Naveen Chandra Tewari ,  Utkarsh Gupta

IPC: G06F17/16 ,  G06F17/18

CPC classification number: G06F17/16 ,  G06K9/6249

Abstract: The present invention provides a method and system for distributed probabilistic matrix factorization. In accordance with a disclosed embodiment, the method may include partitioning a sparse matrix into a first set of blocks on a distributed computer cluster, whereby a dimension of each block is MB rows and NB columns. Further, the method shall include initializing a plurality of matrices including first mean matrix Ū, a first variance matrix Ũ, a first prior variance matrix ŨP, a second mean matrix V, a second variance matrix {tilde over (V)}, and a second prior variance matrix {tilde over (V)}P, by a set of values from a probability distribution function. The plurality of matrices can be partitioned into a set of blocks on the distributed computer cluster, whereby each block can be of a shorter dimension K, and the plurality of matrices can be updated iteratively until a cost function of the sparse matrix converges.

Abstract translation: 本发明提供了一种分布式概率矩阵分解的方法和系统。 根据所公开的实施例，该方法可以包括将稀疏矩阵划分为分布式计算机集群上的第一组块，由此每个块的维度是MB行和NB列。 此外，该方法应包括初始化多个矩阵，包括第一均方矩阵Ū，第一方差矩阵Ũ，第一先验方差矩阵ŨP，第二平均矩阵V，第二方差矩阵（V） 通过来自概率分布函数的一组值的第二先验方差矩阵{tilde over（V）} P。 多个矩阵可以被划分成分布式计算机集群上的一组块，由此每个块可以具有更短的维度K，并且可以迭代地更新多个矩阵，直到稀疏矩阵的成本函数收敛。