PRIVACY-AWARE PERSONALIZED CONTENT FOR THE SMART HOME

• US 20160260135A1
• Filed: 03/04/2015
• Published: 09/08/2016
• Est. Priority Date: 03/04/2015
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Zomet, Asaf, Urbach, Shlomo Reuben
• Granted Patent: US 10,453,098 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06Q 30/0269 based on user profile or attribute
• CPC Class Codes: H04L 12/2812 describing content present in a home automation network, e.g. audio video content retrieval from the Internet G06F16/95
• CPC Class Codes: H04L 12/2829 involving user profiles according to which the execution of a home appliance functionality is automatically triggered
• CPC Class Codes: H04L 12/2834 Switching of information between an external network and a home network access arrangements H04L12/2856
• CPC Class Codes: H04L 67/12 specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34

Abstract

In one embodiment, a computing system may operate within a local area network. The computing system may include a network interface configured to receive a set of content items from a content server located remotely from the computing system and the local area network, a storage element for storing the set of content items, and a processor. The processor may be configured to determine first data relating to people, objects, or some combination thereof, select at least one content item from the set of content items based at least in part on the first data relating to people, objects, or some combination thereof without communicating the first data to the content server or any other computing device outside of the local area network, and communicate the selected at least one content item to a user of the computing system.

Claims

1. A computing system operable within a local area network, the computing system comprising:

a network interface configured to receive a set of content items from a content server located remotely from the computing system and the local area network;

a storage element for storing the set of content items; and

a processor configured to;

determine first data relating to people, objects, or some combination thereof;

select at least one content item from the set of content items based at least in part on the first data relating to people, objects, or some combination thereof without communicating the first data to the content server or any other computing device outside of the local area network; and

communicate the selected at least one content item to a user of the computing system.

2. The computing system of claim 1, wherein the network interface is configured to receive the first data sent from a network interface of a second computing system, and the second computing system comprises one or more sensors configured to obtain the first data.

3. The computing system of claim 2, wherein the network interfaces of the computing system and the second computing system are configured to communicate the first data within the local area network and the local area network is inaccessible to the content server.

4. The computing system of claim 1, comprising a sensor configured to sense the first data, wherein the sensor comprises a camera, a microphone, a temperature sensor, a haptic sensor, a global positioning system (GPS), an occupancy sensor, or some combination thereof, and the first data comprises person identification, person age, person gender, audio data, visual data, object identification, object class, haptic data, temperature, textual data, location, radio frequency identification, or some combination thereof.

5. The computing system of claim 1, wherein the set of content items is received by the network interface from the content server sending the set of content items, the computing system retrieving the set of content items from the content server and indexing the set of content items in the storage element, or some combination thereof.

6. The computing system of claim 1, wherein the set of content items received from the content server comprises advertisements for products, services, or some combination thereof, television show listings, movie show times, shopping recommendations, software application alerts, search result rankings, or some combination thereof.

7. The computing system of claim 1, wherein selecting at least one content item from the set of content items comprises generating a score for each content item of the set of content items using a classifier of the computing system, the classifier trains one or more models with at least the first data, initial scores assigned by the content server to the set of content items, and labels based on user conversion, wherein the labels represent one of two groupings related to showing or not showing the set of content items.

8. The computing system of claim 7, comprising only selecting the at least one content item when at least one content item receives a score generated by the classifier that is higher than a threshold amount and, when more than one content item receives a score higher than the threshold amount, the processor selects the content item with the highest score.

9. The computing system of claim 7, wherein the one or more models are included in a support vector machine used by the classifier that recognizes patterns of a user based on the user'"'"'s interaction with certain content items that is indicative of user conversion and the computing system tracks the user'"'"'s interaction with the presented at least one content item to update the one or more models.

10. The computing system of claim 7, wherein the classifier trains the one or more models with parameters comprising locale, demographics, search history, or some combination thereof, and the classifier is trained for all smart-device environments, for a specific smart-device environment, or for each device configuration.

11. The computing system of claim 1, wherein the processor is configured to use the first data to determine which user is occupying a smart-device environment and selects at least one content item that is relevant to properties of the user based on the first data, the properties comprising age, gender, fashion-taste, mood, preferred activities, medical condition, or some combination thereof.

12. A method of operating a computing system, comprising:

obtaining, via a network interface located within a local area network, a set of content items from a content server located remotely from the computing system and the local area network;

storing, in a storage element located within the local area network, the set of content items;

determining, via a processor located within the local area network, first data relating to people, objects, or some combination thereof;

selecting, via the processor and within the local area network, at least one content item from the set of content items based at least in part on the first data relating to people, objects, or some combination thereof without communicating the first data to the content server or any other computing device outside of the local area network; and

communicating the selected at least one content item to a user of the computing system.

13. The method of claim 12, comprising generating final scores for each content item of the set of content items by:

training a binary classifier using machine learning with the first data, initial scores provided by the content server for each content item of the set of content items, or both, wherein the binary classifier groups each content item of the set of content items into one of two groups including a first group of content items to display and a second group of content items to not display based on user conversion of the communicated at least one content item, and generating the final scores for each content item of the set of content items in the groups.

14. The method of claim 12, wherein the first data is received at the network interface of the computing system from a second network interface of a second computing system, wherein the second computing system collects the first data via one or more sensors and sends the first data via the second network interface to the network interface using the local area network, and the local area network is inaccessible to the content server.

15. The method of claim 12, comprising:

recognizing a person based on the first data;

selecting at least one content item from the set of content items tailored for the person based on one or more properties of the person determined using the first data, wherein the one or more properties comprise a person identification, gender, an age, a fashion-taste, a style, a mood, a preferred activity, a medical condition, a product brand preference, or some combination thereof.

16. The method of claim 12, comprising:

sensing raw first data using one or more sensors that is processed to determine the first data, wherein;

the raw first data is processed using facial recognition, voice recognition, object recognition, pattern recognition, optical character recognition, or some combination thereof; and

the raw first data is processed by the processor, by a second processor of a second computing system that senses the raw data but does not communicate the selected at least one content item to the user of the computing system, or some combination thereof.

17. One or more tangible, non-transitory machine-readable media of a computing system comprising instructions to cause a processor to:

obtain, via a network interface located within a local area network, a set of content items from a content server located remotely from the computing system and the local area network;

store, in a storage element located within the local area network, the set of content items;

determine, via a processor located within the local area network, first data relating to people, objects, or some combination thereof;

select, via the processor and within the local area network, at least one content item from the set of content items based at least in part on the first data relating to people, objects, or some combination thereof without communicating the first data to the content server or any other computing device outside of the local area network; and

communicating the selected at least one content item to a user of the computing system.

18. The media of claim 17, wherein the set of content items is selected by the content server to include diverse content comprising:

advertisements for a product, service, or both, television shows schedule, movie show time recommendations, shopping recommendations, software application alerts, search result ranking changes, or some combination thereof; and

the processor generates final scores for each content item of the set of content items based at least on the first data, initial scores associated with each content item of the set of content items, or both, wherein the final scores cause the set of content items to be rearranged so that the selected at least one content item is personalized based on a user that is determined to be occupying a home according to the first data or is expected to be occupying the home according to an occupancy schedule.

19. The media of claim 17, wherein the instructions cause the processor to:

generate final scores for each content item of the set of content items comprising;

training a binary classifier using machine learning with the first data, initial scores provided by the content server for each content item of the set of content items, or both, wherein the binary classifier groups each content item of the set of content items into one of two groups including a first group of content items to communicate to the user and a second group of content items to not communicate to the user based on user conversion of the presented at least one content item, and generating the final scores for each content item of the set of content items in the groups.

20. The media of claim 17, wherein determining the first data comprises:

receiving the first data via one or more sensors communicatively coupled to the processor;

or

receiving the first data from the network interface, wherein the network interface receives the first data from a second computing system that detects the first data via one or more sensors and sends the first data over the local area network that is inaccessible to the content server;

or

some combination thereof.