公开(公告)号：US20150206196A1

公开(公告)日：2015-07-23

申请号：US14160510

申请日：2014-01-21

Applicant: Facebook, Inc.

Inventor： Chinmay Deepak Karande ,  Mark Rabkin ,  Max Christian Eulenstein

IPC: G06Q30/02

CPC classification number: G06Q30/0275

Abstract: A social networking system modifies a bid amount associated with advertisements in an advertising campaign based on a target average price paid associated with the advertising campaign. A bid amount is determined from the target average price paid and associated with advertisements from the advertising campaigns in various advertisement selection processes. From advertisement selection processes in which an advertisement from the advertising campaign was selected, an average amount charged to the advertiser is determined. Based on the target average price paid and the average amount charged to the advertiser, the bid amount is modified. In subsequent advertisement selection processes, the modified bid amount is associated with advertisements from the advertising campaign.

Abstract translation: 基于与广告活动相关联的目标平均价格，社交网络系统修改与广告活动中的广告相关联的出价金额。 投标金额是根据各种广告选择过程中广告活动中的广告投放的目标平均价格和广告相关来确定的。 从选择广告活动的广告的广告选择处理中，确定向广告商收取的平均金额。 根据目标平均价格和广告客户收取的平均金额，出价金额将被修改。 在随后的广告选择过程中，修改的出价金额与广告活动的广告相关联。

公开(公告)号：US20170161276A1

公开(公告)日：2017-06-08

申请号：US14960172

申请日：2015-12-04

Applicant: Facebook, Inc.

Inventor： Max Christian Eulenstein ,  Lauren Elizabeth Scissors ,  Alexander Peysakhovich ,  Lars Seren Backstrom ,  Lu Wang

IPC: G06F17/30

CPC classification number: G06F17/30867 ,  G06F17/3053 ,  G06Q50/01

Abstract: A social networking system builds a quality controlled and desired population-representative pool of human raters to provide ratings on content items to improve a feed ranking model used for providing its users with more relevant content. The system identifies a pool of candidate human raters for providing ratings on a feed of content items. For each candidate human rater of the pool of candidate human raters, the system presents a feed of content items based on a feed ranking model, obtains ratings on the feed of content items, and determines a score representing the consistency of the obtained ratings, the representativeness of the pool of human raters, or the relevance of the content provided by the ranking model. The system uses the computed scores to modify the ranking model used to present content to its users for improving the relevance of the presented content.

公开(公告)号：US10740690B2

公开(公告)日：2020-08-11

申请号：US15469399

申请日：2017-03-24

Applicant: Facebook, Inc.

Inventor： Jeffrey William Pasternack ,  David Vickrey ,  Justin MacLean Coughlin ,  Prasoon Mishra ,  Austen Norment McDonald ,  Max Christian Eulenstein ,  Jianfu Chen ,  Kritarth Anand ,  Polina Kuznetsova

IPC: G06N20/00 ,  G06F16/35 ,  G06N5/04 ,  G06N20/20

Abstract: An online system predicts topics for content items. The online system provides one or more topic labels for a user to apply concurrently while a user is composing a post, in response to requests periodically received from the user's device. A request includes information such as content composed by the user and contextual information. The online system employs machine learning techniques to analyze content composed by a user and contextual information thereby to predict topic labels. Different machine learning models for classifying individual topic labels, identifying relevant topic labels, and/or detecting changes in existing topic predictions are developed. Some machine learning models predict topics for full content and some predict topics for partial content. The online system trains the machine learning models to ensure accurate topic predictions are provided timely. The online system employs various machine learning model training methods such as active training and gradient training.

公开(公告)号：US20170161667A1

公开(公告)日：2017-06-08

申请号：US14960179

申请日：2015-12-04

Applicant: Facebook, Inc.

Inventor： Max Christian Eulenstein ,  Lauren Elizabeth Scissors ,  Alexander Peysakhovich ,  Lars Seren Backstrom ,  Lu Wang

IPC: G06Q10/06 ,  G06Q50/00 ,  G06F17/30

CPC classification number: G06Q10/06398 ,  G06Q50/01

Abstract: A social networking system builds a quality controlled and desired population-representative pool of human raters to provide ratings on content items to improve a feed ranking model used for providing its users with more relevant content. The system identifies a pool of candidate human raters for providing ratings on a feed of content items. For each candidate human rater of the pool of candidate human raters, the system presents a feed of content items based on a feed ranking model, obtains ratings on the feed of content items, and determines a score representing the consistency of the obtained ratings, the representativeness of the pool of human raters, or the relevance of the content provided by the ranking model. The system uses the computed scores to modify the ranking model used to present content to its users for improving the relevance of the presented content.

公开(公告)号：US10540627B2

公开(公告)日：2020-01-21

申请号：US14960179

申请日：2015-12-04

Applicant: Facebook, Inc.

Inventor： Max Christian Eulenstein ,  Lauren Elizabeth Scissors ,  Alexander Peysakhovich ,  Lars Seren Backstrom ,  Lu Wang

IPC: G06Q10/06 ,  G06Q50/00

Abstract: A social networking system builds a quality controlled and desired population-representative pool of human raters to provide ratings on content items to improve a feed ranking model used for providing its users with more relevant content. The system identifies a pool of candidate human raters for providing ratings on a feed of content items. For each candidate human rater of the pool of candidate human raters, the system presents a feed of content items based on a feed ranking model, obtains ratings on the feed of content items, and determines a score representing the consistency of the obtained ratings, the representativeness of the pool of human raters, or the relevance of the content provided by the ranking model. The system uses the computed scores to modify the ranking model used to present content to its users for improving the relevance of the presented content.

公开(公告)号：US10120945B2

公开(公告)日：2018-11-06

申请号：US14960172

申请日：2015-12-04

Applicant: Facebook, Inc.

Inventor： Max Christian Eulenstein ,  Lauren Elizabeth Scissors ,  Alexander Peysakhovich ,  Lars Seren Backstrom ,  Lu Wang

IPC: G06F17/30 ,  G06Q50/00

Abstract: A social networking system builds a quality controlled and desired population-representative pool of human raters to provide ratings on content items to improve a feed ranking model used for providing its users with more relevant content. The system identifies a pool of candidate human raters for providing ratings on a feed of content items. For each candidate human rater of the pool of candidate human raters, the system presents a feed of content items based on a feed ranking model, obtains ratings on the feed of content items, and determines a score representing the consistency of the obtained ratings, the representativeness of the pool of human raters, or the relevance of the content provided by the ranking model. The system uses the computed scores to modify the ranking model used to present content to its users for improving the relevance of the presented content.

公开(公告)号：US20180276561A1

公开(公告)日：2018-09-27

申请号：US15469399

申请日：2017-03-24

Applicant: Facebook, Inc.

Inventor： Jeffrey William Pasternack ,  David Vickrey ,  Justin MacLean Coughlin ,  Prasoon Mishra ,  Austen Norment McDonald ,  Max Christian Eulenstein ,  Jianfu Chen ,  Kritarth Anand ,  Polina Kuznetsova

IPC: G06N99/00 ,  G06N7/00 ,  G06F17/30

CPC classification number: G06N20/00 ,  G06F16/353 ,  G06N5/041 ,  G06N20/20

Abstract: An online system predicts topics for content items. The online system provides one or more topic labels for a user to apply concurrently while a user is composing a post, in response to requests periodically received from the user's device. A request includes information such as content composed by the user and contextual information. The online system employs machine learning techniques to analyze content composed by a user and contextual information thereby to predict topic labels. Different machine learning models for classifying individual topic labels, identifying relevant topic labels, and/or detecting changes in existing topic predictions are developed. Some machine learning models predict topics for full content and some predict topics for partial content. The online system trains the machine learning models to ensure accurate topic predictions are provided timely. The online system employs various machine learning model training methods such as active training and gradient training.

公开(公告)号：US20170186029A1

公开(公告)日：2017-06-29

申请号：US14983449

申请日：2015-12-29

Applicant: Facebook, Inc.

Inventor： Cassidy Jake Morris ,  Max Christian Eulenstein ,  Tanmoy Chakraborty ,  Joshua Elliot Geller ,  Nikola Mihajlovic

IPC: G06Q30/02 ,  G06Q50/00

CPC classification number: G06Q30/0243 ,  G06Q30/0275 ,  G06Q50/01

Abstract: An online system, such as a social networking system, displays a plurality of advertisements to users. The system selects an ad to display to a user based on a bidding system. The system receives feedback and user engagement data for an ad to compare the ad to other ads that are targeted to a similar group of users, to generate a relevance score. The relevance score can be provided to an advertiser as a way to quantify the effectiveness of the ad, and it reflects user engagement with the advertisement. In some embodiments, a projected relevance score can be calculated for a prospective advertisement by analyzing the content of the prospective ad prior to receiving user engagement data by comparing the prospective advertisement's content to other ads for which user engagement data does exist.

公开(公告)号：US20170161277A1

公开(公告)日：2017-06-08

申请号：US14960181

申请日：2015-12-04

Applicant: Facebook, Inc.

Inventor： Max Christian Eulenstein ,  Lauren Elizabeth Scissors ,  Alexander Peysakhovich ,  Lars Seren Backstrom ,  Lu Wang ,  Virot Chiraphadhanakul

IPC: G06F17/30

CPC classification number: G06F16/9535 ,  G06Q30/0282 ,  G06Q50/01

Abstract: A social networking system builds a quality controlled and desired population-representative pool of human raters to provide ratings on content items to improve a feed ranking model used for providing its users with more relevant content. The system identifies a pool of candidate human raters for providing ratings on a feed of content items. For each candidate human rater of the pool of candidate human raters, the system presents a feed of content items based on a feed ranking model, obtains ratings on the feed of content items, and determines a score representing the consistency of the obtained ratings, the representativeness of the pool of human raters, or the relevance of the content provided by the ranking model. The system uses the computed scores to modify the ranking model used to present content to its users for improving the relevance of the presented content.