公开(公告)号：US10657574B2

公开(公告)日：2020-05-19

申请号：US15264068

申请日：2016-09-13

Applicant: ADOBE INC.

Inventor： Hung Bui ,  Branislav Kveton ,  Suvash Sedhain ,  Nikolaos Vlassis ,  Jaya Kawale

IPC: G06Q30/06 ,  G06F7/08

Abstract: Techniques disclosed herein provide more efficient and more relevant item recommendations to users in large-scale environments in which only positive interest information is known. The techniques use a rank-constrained formulation that generalizes relationships based on known user interests in items and/or use a randomized singular value decomposition (SVD) approximation technique to solve the formulation to identify items of interest to users in an efficiently, scalable manner.

公开(公告)号：US11354720B2

公开(公告)日：2022-06-07

申请号：US16847156

申请日：2020-04-13

Applicant: Adobe Inc.

Inventor： Hung Bui ,  Branislav Kveton ,  Suvash Sedhain ,  Nikolaos Vlassis ,  Jaya Kawale

IPC: G06Q30/06 ,  G06F7/08

Abstract: Techniques disclosed herein provide more efficient and more relevant item recommendations to users in large-scale environments in which only positive interest information is known. The techniques use a rank-constrained formulation that generalizes relationships based on known user interests in items and/or use a randomized singular value decomposition (SVD) approximation technique to solve the formulation to identify items of interest to users in an efficiently, scalable manner.

公开(公告)号：US20200242678A1

公开(公告)日：2020-07-30

申请号：US16847156

申请日：2020-04-13

Applicant: Adobe Inc.

Inventor： Hung Bui ,  Branislav Kveton ,  Suvash Sedhain ,  Nikolaos Vlassis ,  Jaya Kawale

IPC: G06Q30/06 ,  G06F7/08

Abstract: Techniques disclosed herein provide more efficient and more relevant item recommendations to users in large-scale environments in which only positive interest information is known. The techniques use a rank-constrained formulation that generalizes relationships based on known user interests in items and/or use a randomized singular value decomposition (SVD) approximation technique to solve the formulation to identify items of interest to users in an efficiently, scalable manner.