公开(公告)号：US20190073592A1

公开(公告)日：2019-03-07

申请号：US15694321

申请日：2017-09-01

Applicant: Facebook, Inc.

Inventor： Enming Luo ,  Yang Mu ,  Emanuel Alexandre Strauss ,  Taiyuan Zhang ,  Daniel Olmedilla de la Calle

IPC: G06N3/08 ,  G06F21/62

Abstract: A content review system for an online system automatically determines if received content items to be displayed to users violate any policies of the online system. The content review system generates a semantic vector representing the semantic features of a content item, for example, using a neural network. By comparing the semantic vector for the content item with semantic vectors of content items previously determined to violate one or more policies, the content review system determines whether the content item also violates one or more policies. The content review system may also maintain templates corresponding to portions of semantic vectors shared by multiple content items. An analysis of historical content items that conform to the template is performed to determine a probability that received content items that conform to the template violate a policy.

公开(公告)号：US11195099B2

公开(公告)日：2021-12-07

申请号：US15694321

申请日：2017-09-01

Applicant: Facebook, Inc.

Inventor： Enming Luo ,  Yang Mu ,  Emanuel Alexandre Strauss ,  Taiyuan Zhang ,  Daniel Olmedilla de la Calle

IPC: G06N3/08 ,  G06N3/04 ,  G06F21/62 ,  G06Q10/06 ,  G06Q50/00 ,  G06Q30/00

Abstract: A content review system for an online system automatically determines if received content items to be displayed to users violate any policies of the online system. The content review system generates a semantic vector representing the semantic features of a content item, for example, using a neural network. By comparing the semantic vector for the content item with semantic vectors of content items previously determined to violate one or more policies, the content review system determines whether the content item also violates one or more policies. The content review system may also maintain templates corresponding to portions of semantic vectors shared by multiple content items. An analysis of historical content items that conform to the template is performed to determine a probability that received content items that conform to the template violate a policy.

公开(公告)号：US10936952B2

公开(公告)日：2021-03-02

申请号：US15694339

申请日：2017-09-01

Applicant: Facebook, Inc.

Inventor： Enming Luo ,  Yang Mu ,  Emanuel Alexandre Strauss ,  Taiyuan Zhang ,  Daniel Olmedilla de la Calle

IPC: G06N3/08 ,  G06F21/62 ,  G06N3/04 ,  G06Q50/00 ,  G06Q10/10

Abstract: A content review system for an online system automatically determines if received content items to be displayed to users violate any policies of the online system. The content review system generates a semantic vector representing the semantic features of a content item, for example, using a neural network. By comparing the semantic vector for the content item with semantic vectors of content items previously determined to violate one or more policies, the content review system determines whether the content item also violates one or more policies. The content review system may also maintain templates corresponding to portions of semantic vectors shared by multiple content items. An analysis of historical content items that conform to the template is performed to determine a probability that received content items that conform to the template violate a policy.

公开(公告)号：US20190073593A1

公开(公告)日：2019-03-07

申请号：US15694339

申请日：2017-09-01

Applicant: Facebook, Inc.

Inventor： Enming Luo ,  Yang Mu ,  Emanuel Alexandre Strauss ,  Taiyuan Zhang ,  Daniel Olmedilla de la Calle

IPC: G06N3/08 ,  G06F21/62

Abstract: A content review system for an online system automatically determines if received content items to be displayed to users violate any policies of the online system. The content review system generates a semantic vector representing the semantic features of a content item, for example, using a neural network. By comparing the semantic vector for the content item with semantic vectors of content items previously determined to violate one or more policies, the content review system determines whether the content item also violates one or more policies. The content review system may also maintain templates corresponding to portions of semantic vectors shared by multiple content items. An analysis of historical content items that conform to the template is performed to determine a probability that received content items that conform to the template violate a policy.