公开(公告)号：US11821742B2

公开(公告)日：2023-11-21

申请号：US16584444

申请日：2019-09-26

Applicant: Snap Inc.

Inventor： Thomas Kay Brown

IPC: G01C21/34 ,  G06N20/00 ,  G06Q30/0251 ,  H04W4/12 ,  H04W4/23 ,  H04W4/42 ,  H04W4/021

CPC classification number: G01C21/3476 ,  G01C21/3423 ,  G01C21/3484 ,  G06N20/00 ,  G06Q30/0261 ,  H04W4/12 ,  H04W4/021 ,  H04W4/23 ,  H04W4/42

Abstract: Systems and methods are provided for performing operations comprising: establishing a plurality of geofences each associated with a respective point of interest; determining a current location of a user and a current time; applying a trained machine learning model to the current location and the current time to predict a given geofence of the plurality of geofences that will be traversed by the user at a future time, the machine learning model being trained to establish a relationship between travel times and paths of the user and a set of geofences of the plurality of geofences that is traversed by the paths during the travel times; retrieving the point of interest associated with the predicted given geofence; and automatically generating a notification relating to the retrieved point of interest for presentation to the user before the user traverses the given geofence associated with the retrieved point of interest.

公开(公告)号：US20210095986A1

公开(公告)日：2021-04-01

申请号：US16584444

申请日：2019-09-26

Applicant: Snap Inc.

Inventor： Thomas Kay Brown

IPC: G01C21/34 ,  H04W4/12 ,  G06N20/00 ,  G06Q30/02

Abstract: Systems and methods are provided for performing operations comprising: establishing a plurality of geofences each associated with a respective point of interest; determining a current location of a user and a current time; applying a trained machine learning model to the current location and the current time to predict a given geofence of the plurality of geofences that will be traversed by the user at a future time, the machine learning model being trained to establish a relationship between travel times and paths of the user and a set of geofences of the plurality of geofences that is traversed by the paths during the travel times; retrieving the point of interest associated with the predicted given geofence; and automatically generating a notification relating to the retrieved point of interest for presentation to the user before the user traverses the given geofence associated with the retrieved point of interest.