公开(公告)号：US10904599B2

公开(公告)日：2021-01-26

申请号：US15994776

申请日：2018-05-31

Applicant: Adobe Inc.

Inventor： Somdeb Sarkhel ,  Viswanathan Swaminathan ,  Shuo Yang ,  Saayan Mitra ,  Lakshmi Shivalingaiah ,  Jason Boyer ,  Dwight Rodgers

IPC: H04N21/45 ,  H04N21/25 ,  H04N21/466 ,  H04N21/258 ,  G06K9/62 ,  G06N20/00

Abstract: This disclosure relates to methods, non-transitory computer readable media, and systems that determine multiple personas corresponding to a user account for digital content and train a persona classifier to predict a given persona (from among the multiple personas) for content requests associated with the user account. By using the persona classifier, the disclosed methods, non-transitory computer readable media, and systems accurately detect a given persona for a content request upon initiation of the request. Based on determining the given persona, in some implementations, the methods, non-transitory computer readable media, and systems generate a digital-content recommendation for presentation on a client device associated with the user account.

公开(公告)号：US11914665B2

公开(公告)日：2024-02-27

申请号：US17675290

申请日：2022-02-18

Applicant: Adobe Inc.

Inventor： Matvey Kapilevich ,  Margarita R. Savova ,  Anup Bandigadi Rao ,  Tung Thanh Mai ,  Lakshmi Shivalingaiah ,  Liron Goren Snai ,  Charles Menguy ,  Vijeth Lomada ,  Moumita Sinha ,  Harleen Sahni

IPC: G06F16/248 ,  G06F16/9538 ,  G06F16/28 ,  G06F16/901 ,  G06N20/00

CPC classification number: G06F16/9538 ,  G06F16/248 ,  G06F16/283 ,  G06F16/9024 ,  G06N20/00

Abstract: Multi-modal machine-learning model training techniques for search are described that overcome conventional challenges and inefficiencies to support real time output, which is not possible in conventional training techniques. In one example, a search system is configured to support multi-modal machine-learning model training. This includes use of a preview mode and an expanded mode. In the preview mode, a preview segment is generated as part of real time training of a machine learning model. In the expanded mode, the preview segment is persisted as an expanded segment that is used to train and utilize an expanded machine-learning model as part of search.

公开(公告)号：US10601953B2

公开(公告)日：2020-03-24

申请号：US15273580

申请日：2016-09-22

Applicant: Adobe Inc.

Inventor： Payal Bajaj ,  Sumit Shekhar ,  Lakshmi Shivalingaiah ,  George Horia Galatanu

IPC: H04L29/08

Abstract: Various embodiments disambiguate users who share media content accounts to provide persona-based experience individualization. Personas correspond to commonly observed channel watching patterns among media content customers. Decomposition of the media content account into personas is achieved by analyzing many accounts, e.g., millions of accounts, on media content platforms. By analyzing accounts, a recommendation system can individualize the channel watching experience in media content accounts.

公开(公告)号：US20230267158A1

公开(公告)日：2023-08-24

申请号：US17675290

申请日：2022-02-18

Applicant: Adobe Inc.

Inventor： Matvey Kapilevich ,  Margarita R. Savova ,  Anup Bandigadi Rao ,  Tung Thanh Mai ,  Lakshmi Shivalingaiah ,  Liron Goren Snai ,  Charles Menguy ,  Vijeth Lomada ,  Moumita Sinha ,  Harleen Sahni

IPC: G06F16/9538 ,  G06F16/901 ,  G06F16/28

CPC classification number: G06F16/9538 ,  G06F16/9024 ,  G06F16/283 ,  G06N20/00

Abstract: Multi-modal machine-learning model training techniques for search are described that overcome conventional challenges and inefficiencies to support real time output, which is not possible in conventional training techniques. In one example, a search system is configured to support multi-modal machine-learning model training. This includes use of a preview mode and an expanded mode. In the preview mode, a preview segment is generated as part of real time training of a machine learning model. In the expanded mode, the preview segment is persisted as an expanded segment that is used to train and utilize an expanded machine-learning model as part of search.

公开(公告)号：US20210056458A1

公开(公告)日：2021-02-25

申请号：US16545224

申请日：2019-08-20

Applicant: Adobe Inc.

Inventor： Margarita Savova ,  Matvey Kapilevich ,  Lakshmi Shivalingaiah ,  Anup Rao ,  Alexandru Ionut Hodorogea ,  Harleen Singh Sahni

IPC: G06N20/00 ,  G06N5/02

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods for intelligently predicting a persona class of a client device and/or target user utilizing an overlap-agnostic machine learning model and distributing persona-based digital content to the client device. In particular, in one or more embodiments, the persona classification system can learn overlap-agnostic machine learning model parameters to apply to user traits in real-time or in offline batches. For example, the persona classification system can train and utilize an overlap-agnostic machine learning model that includes an overlap-agnostic embedding model, a trained user-embedding generation model, and a trained persona prediction model. By applying the learned overlap-agnostic machine learning model parameters to the target user traits, the persona classification system can predict a persona class for sending digital content based on the predicted persona class.