Aspect pre-selection using machine learning

Patent metadata

Field	Value
Publication number	US-12314830-B2
Application number	US-202318523674-A
Country	US
Kind code	B2
Filing date	Nov 29, 2023
Priority date	Nov 20, 2017
Publication date	May 27, 2025
Grant date	May 27, 2025

Abstract

Aspect pre-selection techniques using machine learning are described. In one example, an artificial assistant system is configured to implement a chat bot. A user then engages in a first natural-language conversation. As part of this first natural-language conversation, a communication is generated by the chat bot to prompt the user to specify an aspect of a category that is a subject of a first natural-language conversation and user data is received in response. Data that describes this first natural-language conversation is used to train a model using machine learning. Data is then received by the chat bot as part of a second natural-language conversation. This data, from the second natural-language conversation, is processed using the model as part of machine learning to generate the second search query to include the aspect of the category automatically and without user intervention.

First claim

What is claimed is:

1. A method comprising:
receiving an input selecting an object in a digital image of a plurality of digital images;
identifying the object in the digital image by processing the digital image with an object recognition module;
extracting a category of the object based on the identifying;
selecting, based on the category, a model from a plurality of models, wherein respective models of the plurality of models are associated with different categories, and wherein the respective models are trained using machine learning and previous data corresponding to the different categories;
generating data describing at least one aspect associated with the category of the object using the model;
performing a search for digital content that pertains to the object based on the object and the data describing the at least one aspect; and
outputting, in real time, the digital content by superimposing the digital content onto the plurality of digital images.

2. The method as described in claim 1 , wherein the input selecting the object in the digital image is received from a user, and wherein the model is specific for the user.

3. The method as described in claim 2 , wherein the previous data corresponding to the different categories is received from the user.

4. The method as described in claim 1 , wherein the plurality of digital images are representative of a camera feed, and wherein the digital image is received from the camera feed.

5. The method as described in claim 4 , wherein outputting the digital content comprises rendering the digital content relative to a view of the object in the camera feed.

6. The method as described in claim 4 , wherein outputting the digital content comprises displaying the digital content as a part of the camera feed.

7. The method as described in claim 1 , wherein the digital content comprises at least one product available for purchase.

8. The method as described in claim 1 , wherein the previous data is received in response to a communication generated as a part of a natural-language conversation.

9. A computing device comprising:
a processing system; and
a computer-readable storage medium storing instructions that, responsive to execution by the processing system, causes the processing system to perform operations comprising:
displaying a digital image of a plurality of digital images;
receiving an input selecting an object in the digital image;
identifying the object in the digital image by processing the digital image with an object recognition module;
extracting a category of the object based on the identifying;
selecting, based on the category, a model from a plurality of models, wherein respective models of the plurality of models are associated with different categories, and wherein the respective models are trained using machine learning and previous data corresponding to the different categories;
generating data describing at least one aspect associated with the category of the object using the model;
performing a search for digital content that pertains to the object based on the object and the data describing the at least one aspect; and
outputting the digital content by superimposing the digital content onto the plurality of digital images.

10. The computing device as described in claim 9 , wherein the input selecting the object in the digital image is received from a user, wherein the model is specific for the user, and wherein the previous data corresponding to the different categories comprises user data received from the user during a previous interaction with the computing device.

11. The computing device as described in claim 9 , wherein performing the search for the digital content that pertains to the object based on the object and the data describing the at least one aspect comprises:
generating text based on the object identified in the digital image;
identifying additional text based on the data describing the at least one aspect associated with the category of the object;
generating a search query based on the text and the additional text; and
initiating the search using the search query.

12. The computing device as described in claim 9 , wherein the operations further comprise displaying an indication that the data describing the at least one aspect is included in the search.

13. The computing device as described in claim 12 , wherein the indication is user selectable to initiate a subsequent search that is not based on the data describing the at least one aspect.

14. A computer-readable storage medium storing executable instructions that, responsive to execution by a processing system, causes the processing system to perform operations comprising:
receiving, via a user interface, an input selecting an object in a digital image of a plurality of digital images;
identifying the object in the digital image by processing the digital image with an object recognition module;
extracting a category of the object based on the identifying;
selecting, based on the category, a model from a plurality of models, wherein respective models of the plurality of models are associated with different categories, and wherein the respective models are trained using machine learning and previous data corresponding to the different categories;
generating data describing at least one aspect associated with the category of the object using the model;
performing a search for digital content that pertains to the object based on the object and the data describing the at least one aspect; and
outputting, in real time via the user interface, the digital content by superimposing the digital content onto the plurality of digital images.

15. The computer-readable storage medium as described in claim 14 , wherein the input selecting the object in the digital image is received from a user, wherein the model is specific for the user, and wherein the previous data corresponding to the different categories comprises user data received from the user during a previous interaction with the user interface.

16. The computer-readable storage medium as described in claim 14 , wherein the plurality of digital images are representative of a camera feed, wherein the digital image is received from the camera feed, and wherein outputting the digital content comprises rendering the digital content in the user interface relative to a view of the object in the camera feed.

17. The computer-readable storage medium as described in claim 14 , wherein the object recognition module is trained using training digital images that are tagged with corresponding identifications by users of a commerce service provider system.

18. The computer-readable storage medium as described in claim 14 , wherein the digital content comprises at least one product available for purchase.

19. The computer-readable storage medium as described in claim 14 , wherein the previous data is received in response to a communication generated as a part of a natural-language conversation.

20. The computing device as described in claim 9 , wherein the previous data is received in response to a communication generated as a part of a natural-language conversation.

Assignees

• Ebay Inc

Inventors

• Abdallah Farah
• Enyedi Robert
• Srivastava Amit
• Lee Elaine
• Gaskill Braddock Craig
• Lancewicki Tomer
• Zhang Xinyu
• Vasudevan Jayanth
• Bouchon Dominique Jean

Classifications

• G06F16/248

  Presentation of query results · CPC title

• G06N20/00

  Machine learning · CPC title

• G06F40/30

  Semantic analysis · CPC title

• G06Q10/10

  Office automation; Time management · CPC title

• G06F16/90332

  Natural language query formulation or dialogue systems · CPC title