公开(公告)号：US20180012236A1

公开(公告)日：2018-01-11

申请号：US15203719

申请日：2016-07-06

Applicant: Facebook, Inc.

Inventor： Yue Zhuo ,  Meihong Wang ,  Mahmud Sami Tas ,  Jie Xu ,  Kristin S. Hendrix ,  Vibhi Kant

IPC: G06Q30/02 ,  G06N99/00

CPC classification number: G06Q30/0201 ,  G06N20/00

Abstract: Systems, methods, and non-transitory computer-readable media can select one or more content items that are associated with one or more interactions that each at least meet a specified interaction metric threshold. Data associated with the one or more content items can be acquired. A classifier can be developed based on the data associated with the one or more content items. At least some of the one or more content items can be identified, based on the classifier, as having at least a threshold confidence score of being interaction-bait content.

公开(公告)号：US20170177578A1

公开(公告)日：2017-06-22

申请号：US14975433

申请日：2015-12-18

Applicant: Facebook, Inc.

Inventor： Mahmud Sami Tas ,  Vibhi Kant ,  Gregory Matthew Marra ,  Dean Griffin Eckles ,  Yaroslav O. Dudin

IPC: G06F17/30 ,  H04L29/08 ,  G06N7/00

Abstract: Systems, methods, and non-transitory computer-readable media can obtain at least one content item to be scored with respect to a user, wherein the score is used to determine whether the content item is included in a content feed of the user. A value for a first event is determined based at least in part on a likelihood of the user performing an action with respect to the content item, wherein the likelihood of the user performing the action is greater than an average user of the social networking system. The score for the content item is determined based at least in part on the value for the first event, wherein the first value is dampened to reduce its contribution to the score.

公开(公告)号：US10789258B2

公开(公告)日：2020-09-29

申请号：US14975433

申请日：2015-12-18

Applicant: Facebook, Inc.

Inventor： Mahmud Sami Tas ,  Vibhi Kant ,  Gregory Matthew Marra ,  Dean Griffin Eckles ,  Yaroslav O. Dudin

IPC: G06F16/00 ,  G06F16/2457 ,  H04W12/08 ,  H04W4/21 ,  G06N20/00

Abstract: Systems, methods, and non-transitory computer-readable media can obtain at least one content item to be scored with respect to a user, wherein the score is used to determine whether the content item is included in a content feed of the user. A value for a first event is determined based at least in part on a likelihood of the user performing an action with respect to the content item, wherein the likelihood of the user performing the action is greater than an average user of the social networking system. The score for the content item is determined based at least in part on the value for the first event, wherein the first value is dampened to reduce its contribution to the score.