公开(公告)号：US10592513B1

公开(公告)日：2020-03-17

申请号：US14599407

申请日：2015-01-16

Applicant: GOOGLE LLC

Inventor： Lucian Florin Cionca ,  Junbin Teng ,  Andre Rohe ,  Harish Chandran ,  Yumio Saneyoshi

IPC: G06F16/20 ,  G06F16/2457

Abstract: In one aspect, a method includes identifying a first user viewing a first set of posts at a social networking service, the first set of posts including one or more posts, determining that the level of engagement of the first user at the social networking service is below a predetermined level, generating a second set of posts in response to determining that the level of engagement of the first user at the social networking service is below a predetermined level, the second set of posts including one or more posts not previously seen by the user and providing the second set of posts for display to the user. Other aspects can be embodied in corresponding systems and apparatus, including computer program products.

公开(公告)号：US10055463B1

公开(公告)日：2018-08-21

申请号：US14946233

申请日：2015-11-19

Applicant: Google LLC

Inventor： Hariharan Chandrasekaran ,  Madhavi Yenugula ,  Harish Chandran

IPC: G06F17/30

CPC classification number: G06F16/24578 ,  G06F16/951

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for feature-based ranking adjustment. In one aspect, a method includes finalizing rankings of resources based on detected features, and for each resource for which a ranking is not finalized, finalizing the respective resources or demoting the resources based on the detection of features common to the resources with the finalized rankings and the resources with the unfinalized rankings.

公开(公告)号：US20230162091A1

公开(公告)日：2023-05-25

申请号：US18070195

申请日：2022-11-28

Applicant: GOOGLE LLC

Inventor： Archit Gupta ,  Hariharan Chandrasekaran ,  Harish Chandran

IPC: G06N20/00 ,  G06F16/783 ,  G06F16/33 ,  G06F16/9536 ,  G06F16/583 ,  G06N7/01 ,  G06N5/02

CPC classification number: G06N20/00 ,  G06F16/783 ,  G06F16/334 ,  G06F16/9536 ,  G06F16/583 ,  G06N7/01 ,  G06N5/02

Abstract: Training and/or utilizing a machine learning model to generate request agnostic predicted interaction scores for electronic communications, and to utilization of request agnostic predicted interaction scores in determining whether, and/or how, to provide corresponding electronic communications to a client device in response to a request. A request agnostic predicted interaction score for an electronic communication provides an indication of quality of the communication, and is generated independent of corresponding request(s) for which it is utilized. In many implementations, a request agnostic predicted interaction score for an electronic communication is generated “offline” relative to corresponding request(s) for which it is utilized, and is pre-indexed with (or otherwise assigned to) the electronic communication. This enables fast and efficient retrieval, and utilization, of the request agnostic interaction score by computing device(s), when the electronic communication is responsive to a request.

公开(公告)号：US10409818B1

公开(公告)日：2019-09-10

申请号：US15228756

申请日：2016-08-04

Applicant: Google LLC

Inventor： Matthew Hayes ,  Hariharan Chandrasekaran ,  Harish Chandran

IPC: G06F16/00 ,  G06F16/2455 ,  G06F16/248 ,  G06F16/901 ,  G06F16/2457

Abstract: Methods, systems, apparatus, including computer programs encoded on computer storage medium, for a bottom-up approach for generating high-quality content streams. In one aspect, the method includes actions of obtaining data identifying a plurality of content items, generating a plurality of queries for the particular topic, and for each query of the plurality of queries: obtaining a set of search results for the query that identify content items identified in the obtained data, and determining, from the search results for the query, a respective quality score for each of one or more quality characteristics. The method may also include actions such as identifying one or more first high-quality queries from the plurality of queries based on the respective quality scores for the one or more quality characteristics, and populating a stream of content for display on the user device using search results for the one or more first high-quality queries.

公开(公告)号：US11514353B2

公开(公告)日：2022-11-29

申请号：US15795204

申请日：2017-10-26

Applicant: Google LLC

Inventor： Archit Gupta ,  Hariharan Chandrasekaran ,  Harish Chandran

IPC: G06N5/02 ,  G06N7/00 ,  G06N20/00 ,  G06F16/783 ,  G06F16/33 ,  G06F16/9536 ,  G06F16/583

Abstract: Training and/or utilizing a machine learning model to generate request agnostic predicted interaction scores for electronic communications, and to utilization of request agnostic predicted interaction scores in determining whether, and/or how, to provide corresponding electronic communications to a client device in response to a request. A request agnostic predicted interaction score for an electronic communication provides an indication of quality of the communication, and is generated independent of corresponding request(s) for which it is utilized. In many implementations, a request agnostic predicted interaction score for an electronic communication is generated “offline” relative to corresponding request(s) for which it is utilized, and is pre-indexed with (or otherwise assigned to) the electronic communication. This enables fast and efficient retrieval, and utilization, of the request agnostic interaction score by computing device(s), when the electronic communication is responsive to a request.

公开(公告)号：US10296512B1

公开(公告)日：2019-05-21

申请号：US15275261

申请日：2016-09-23

Applicant: Google LLC

Inventor： Harish Chandran ,  Ka Hung Hui

IPC: G06F16/2457 ,  G06F16/248 ,  G06F16/438 ,  G06F16/28 ,  G06F16/48

Abstract: Aspects of the subject technology relate to systems and methods for action-based content scoring. Scores associated with a content item are determined. Each of the scores is generated by a different predictive model and associated with a respective user interaction type. A composite score for the content item is determined based on at least one of the scores. The content item is provided for display in a content stream associated with a user based on the composite score.

公开(公告)号：US20190130304A1

公开(公告)日：2019-05-02

申请号：US15795204

申请日：2017-10-26

Applicant: Google LLC

Inventor： Archit Gupta ,  Hariharan Chandrasekaran ,  Harish Chandran

IPC: G06N99/00 ,  G06N5/02 ,  G06N7/00

Abstract: Training and/or utilizing a machine learning model to generate request agnostic predicted interaction scores for electronic communications, and to utilization of request agnostic predicted interaction scores in determining whether, and/or how, to provide corresponding electronic communications to a client device in response to a request. A request agnostic predicted interaction score for an electronic communication provides an indication of quality of the communication, and is generated independent of corresponding request(s) for which it is utilized. In many implementations, a request agnostic predicted interaction score for an electronic communication is generated “offline” relative to corresponding request(s) for which it is utilized, and is pre-indexed with (or otherwise assigned to) the electronic communication. This enables fast and efficient retrieval, and utilization, of the request agnostic interaction score by computing device(s), when the electronic communication is responsive to a request.