公开(公告)号：US11604968B2

公开(公告)日：2023-03-14

申请号：US15838287

申请日：2017-12-11

申请人： Meta Platforms, Inc.

发明人： Myle Arif Ott ,  Aaron Bryan Adcock ,  Yaniv Shmueli ,  Peng-Jen Chen ,  Wenbo Yuan ,  Junfei Wang

IPC分类号： G06N3/08 ,  G06N5/02 ,  G06N3/04 ,  G06F7/08 ,  H04L67/306 ,  H04L67/52 ,  H04L67/10 ,  G01S19/48

摘要： In one embodiment, a method includes receiving, from a client system associated with a user of an online social network, data indicating that the user is located at a first geographic location at a first time; accessing a first embedding representing a first place-entity corresponding to the first geographic location; accessing multiple second embeddings representing multiple respective second place-entities each corresponding to a second geographic location; calculating, a similarity metric between the embedding representing the first place-entity and each of the embeddings representing the second place-entities; ranking each of the second place-entities based on their calculated similarity metrics; and sending, to the client system, information associated with one or more second geographic locations corresponding to one or more second place-entities having a ranking greater than a threshold ranking.

公开(公告)号：US20230206034A1

公开(公告)日：2023-06-29

申请号：US18182052

申请日：2023-03-10

申请人： Meta Platforms, Inc.

发明人： Myle Arif Ott ,  Aaron Bryan Adcock ,  Yaniv Shmueli ,  Peng-Jen Chen ,  Wenbo Yuan ,  Junfei Wang

IPC分类号： G06N3/047 ,  G06N3/08 ,  G06F7/08 ,  H04L67/306 ,  H04L67/52 ,  G06N5/02

CPC分类号： G06N3/047 ,  G06N3/08 ,  G06F7/08 ,  H04L67/306 ,  H04L67/52 ,  G06N5/02 ,  H04L67/10

摘要： In one embodiment, a method includes accessing a place-entities graph comprising a plurality of place-entity nodes, in which each place-entity node representing a place-entity corresponding to a particular geographic location, and identifying a place-entity cluster within the place-entities graph. The place-entity cluster comprises a plurality of place-entity nodes corresponding to a plurality of place-entities corresponding to the same geographic location. The method includes accessing embeddings representing the plurality of place-entities corresponding to the place-entity cluster. Each embedding is a point in a d-dimensional embedding space. The method includes calculating, using a machine-learning model, a cluster-quality score of the place-entity cluster based on the embeddings. The cluster-quality score represents a probability that the place-entities correspond to a valid geographic location. The method further includes identifying the place-entities as corresponding to an invalid geographic location based on a determining that the cluster-quality score is less than a threshold cluster-quality score.

公开(公告)号：US20230186035A1

公开(公告)日：2023-06-15

申请号：US17889116

申请日：2022-08-16

申请人： Meta Platforms, Inc.

发明人： Ann Lee ,  Peng-Jen Chen ,  Holger Schwenk ,  Jiatao Gu ,  Wei-Ning Hsu

IPC分类号： G06F40/42 ,  G10L15/06 ,  G10L21/0208

CPC分类号： G06F40/42 ,  G10L15/063 ,  G10L21/0208 ,  G10L2015/0631

摘要： In one embodiment, a method includes accessing a first utterance of a content by a first speaker, generating first discrete speech units from the first utterance based on a speech-learning model, wherein each of the first discrete speech units is associated with a speech cluster, accessing second utterances of the content by second speakers different from the first speaker, and training a speech normalizer by processing each of the second utterances using the speech normalizer to generate second discrete speech units and updating the speech normalizer by using the first discrete speech units as an optimization target for the second discrete speech units associated with each of the second utterances.