US 11,755,924 B2
Machine learning for home understanding and notification
Donald Madden, Columbia, MD (US); and Allison Beach, Leesburg, VA (US)
Assigned to ObjectVideo Labs, LLC, Tysons, VA (US)
Filed by ObjectVideo Labs, LLC, Tysons, VA (US)
Filed on May 20, 2019, as Appl. No. 16/416,851.
Claims priority of provisional application 62/673,523, filed on May 18, 2018.
Prior Publication US 2019/0354875 A1, Nov. 21, 2019
Int. Cl. G06N 5/02 (2023.01); H04L 12/28 (2006.01); G06N 20/00 (2019.01)
CPC G06N 5/02 (2013.01) [G06N 20/00 (2019.01); H04L 12/2803 (2013.01)] 23 Claims
OG exemplary drawing
 
1. A home monitoring system comprising:
one or more processors; and
at least one computer-readable storage medium couple to the one or more processors having stored thereon instructions which, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
receiving, from one or more devices in a home network that are each physically at a property, activity data that represents activity of one or more objects or one or more people at the property;
generating, using the activity data and a machine learning process, a knowledge graph including two or more nodes, wherein each node represents a corresponding object, person, or routine and wherein each link connecting two nodes in the two or more nodes represents a relationship between the two nodes;
generating, using the knowledge graph that includes the two or more nodes each of which represents the corresponding object, person, or routine and each link connecting two nodes represents the relationship between the two nodes, one or more rules;
determining, using the one or more rules and second activity data, an occurrence of a particular rule of the one or more rules;
in response to determining the occurrence of the particular rule of the one or more rules, causing presentation, by a user device, of a notification responsive to the particular rule;
receiving, from the user device, user feedback responsive to the notification, wherein the user feedback includes a natural language label for the particular rule, and the natural language label for the particular rule is a phrase entered by a user in an application interface of the user device; and
in response to receiving the user feedback, updating a node in the knowledge graph from which the particular rule was generated using the user feedback that includes the natural language label for the particular rule.