Automated device grouping

Patent metadata

Field	Value
Publication number	US-11240641-B1
Application number	US-202017096357-A
Country	US
Kind code	B1
Filing date	Nov 12, 2020
Priority date	Nov 12, 2020
Publication date	Feb 1, 2022
Grant date	Feb 1, 2022

Abstract

Techniques for automatically combining devices into a single group of devices, and splitting devices into multiple groups of devices are described. A machine learning model may process device profile data, associated with devices registered to two different users, and determine the devices should be combined into a single group of devices. Such enables a user to control each of the devices, of the two different users, but providing user inputs to a single device of the group. The machine learning model may also process device profile data, associated with devices registered to a single user, and determine the devices should be split into two or more groups of devices. Such may decrease the likelihood that a system may inadvertently control a device not intended by a user.

First claim

What is claimed is:

1. A computer-implemented method comprising:
receiving a first device identifier representing a first device connected to a first wireless router, the first device identifier being associated with a first user profile identifier;
receiving a second device identifier representing a second device connected to the first wireless router, the second device identifier being associated with a second user profile identifier;
based on determining the first device and the second device are both connected to the first wireless router, determining processing is to be performed to determine whether the first device and the second device are to be represented in a single group of devices;
after determining the processing is to be performed, determining a third device identifier associated with the first user profile identifier;
determining first device profile data associated with the first device identifier;
determining second device profile data associated with the second device identifier;
determining third device profile data associated with the third device identifier;
determining a first output data representing a first cosine similarity between the first device profile data and the second device profile data;
determining a second output data representing a second cosine similarity between the third device profile data and the second device profile data;
based on the first output data and the second output data, determining the first device identifier, the second device identifier, and the third device identifier are to be represented in a group of device identifiers associated with a physical location; and
storing first data associating the first device identifier, the second device identifier, and the third device identifier.

2. The computer-implemented method of claim 1 , further comprising:
determining a first internet protocol (IP) address associated with the first device identifier;
determining a second IP address associated with the second device identifier;
determining a number of digits that are similar between the first IP address and the second IP address; and
further based on the number of digits, determining the first device identifier, the second device identifier, and the third device identifier are to be represented in the group of device identifiers.

3. The computer-implemented method of claim 1 , further comprising:
receiving, from the first device, audio data representing a spoken natural language input;
processing the audio data to determine the spoken natural language input requests a device perform first processing, the spoken natural language input including a device name;
determining the device name is unassociated with the first user profile identifier;
determining the device name is associated with the second user profile identifier; and
based on the device name being associated with the second user profile identifier, determining the first output data.

4. The computer-implemented method of claim 1 , further comprising:
based on the first output data and the second output data, determining a distribution of cosine similarities; and
using a machine learning model, processing the distribution of cosine similarities to determine the first device identifier, the second device identifier, and the third device identifier are to be represented in the group of device identifiers.

5. A computer-implemented method comprising:
receiving a first device identifier representing a first device, the first device identifier being associated with a first user profile identifier;
determining a second device identifier representing a second device, the second device identifier being associated with a second user profile identifier;
determining first device profile data associated with the first device identifier;
determining second device profile data associated with the second device identifier;
determining a first similarity between the first device profile data and the second device profile data;
based at least in part on the first similarity, determining the first device identifier and the second device identifier are to be represented in a group of device identifiers; and
storing first data associating the first device identifier and the second device identifier.

6. The computer-implemented method of claim 5 , further comprising:
determining a third device identifier associated with the first user profile identifier;
determining third device profile data associated with the third device identifier;
determining a second similarity between the third device profile data and the second device profile data;
further based at least in part on the second similarity, determining the first device identifier and the second device identifier are to be represented in a group of device identifiers; and
storing the first data to associate the first device identifier, the second device identifier, and the third device identifier.

7. The computer-implemented method of claim 6 , further comprising:
based on the first similarity and the second similarity, determining a distribution of similarities; and
using a machine learning model, processing the distribution of similarities to determine the first device identifier, the second device identifier, and the third device identifier are to be represented in the group of device identifiers.

8. The computer-implemented method of claim 5 , further comprising:
determining a first internet protocol (IP) address associated with the first device identifier;
determining a second IP address associated with the second device identifier;
determining a second similarity between the first IP address and the second IP address; and
further based on the second similarity, determining the first device identifier and the second device identifier are to be represented in the group of device identifiers.

9. The computer-implemented method of claim 5 , further comprising:
receiving, from the first device, audio data representing a spoken natural language input;
processing the audio data to determine the spoken natural language input requests a device perform first processing, the spoken natural language input including a device name;
determining the device name is unassociated with the first user profile identifier;
determining the device name is associated with the second user profile identifier; and
based on the device name being associated with the second user profile identifier, determining the first similarity.

10. The computer-implemented method of claim 5 , further comprising:
determining the first device identifier is associated with a network device;
determining the second device identifier is associated with the network device; and
based on the first device identifier and the second device identifier both being associated with the network device, determining the first similarity.

11. The computer-implemented method of claim 5 , further comprising:
determining a third device identifier associated with the first user profile identifier;
determining a fourth device identifier associated with the first user profile identifier;
determining third device profile data associated with the third device identifier;
determining fourth device profile data associated with the fourth device identifier;
determining a second similarity between the third device profile data and the fourth device profile data;
based at least in part on the second similarity, determining the third device identifier and the fourth device identifier are to be represented in different groups of device identifiers;
storing second data associating the third device identifier and a first device group identifier; and

Assignees

• Amazon Tech Inc

Inventors

• Eberhardt Sven
• Pareek Shekhar
• Basak Aniruddha
• Brett Charles Edwin Ashton
• Salimi Amir

Classifications

• H04L67/306

  User profiles · CPC title

• H04L67/12

  specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks · CPC title

• H04W8/26

  Network addressing or numbering for mobility support · CPC title

• H04W8/186

  Processing of subscriber group data · CPC title

• H04W4/08 Primary

  User group management · CPC title