Invention Grant
US07870083B2 Systems and methods for generating predictive matrix-variate T models
有权
用于生成预测矩阵变量T模型的系统和方法
- Patent Title: Systems and methods for generating predictive matrix-variate T models
- Patent Title (中): 用于生成预测矩阵变量T模型的系统和方法
-
Application No.: US11869886Application Date: 2007-10-10
-
Publication No.: US07870083B2Publication Date: 2011-01-11
- Inventor: Shenghuo Zhu , Kai Yu , Yihong Gong
- Applicant: Shenghuo Zhu , Kai Yu , Yihong Gong
- Applicant Address: US NJ Princeton
- Assignee: NEC Laboratories America, Inc.
- Current Assignee: NEC Laboratories America, Inc.
- Current Assignee Address: US NJ Princeton
- Agent Bao Tran; Joseph Kolodka
- Main IPC: G06F15/18
- IPC: G06F15/18 ; G06F15/00
Abstract:
Systems and methods are disclosed to predict one or more missing elements from a partially-observed matrix by receiving one or more user item ratings; generating a model parameterized by matrices U, S, V; applying the model to display an item based on one or more predicted missing elements; and applying the model at run-time and determining UiTSVj.
Public/Granted literature
- US20090099984A1 SYSTEMS AND METHODS FOR GENERATING PREDICTIVE MATRIX-VARIATE T MODELS Public/Granted day:2009-04-16
Information query
