Crowdsourcing responses in a query processing system

• US 10,503,739 B2
• Filed: 04/20/2017
• Issued: 12/10/2019
• Est. Priority Date: 04/20/2017
• Status: Active Grant

• Current Assignee: Breville USA Incorporated (Breville Group Ltd.)
• Original Assignee: Breville USA Incorporated (Breville Group Ltd.)
• Inventors: Natkin, Michael Joshua, Young, Christopher Charles
• Primary Examiner(s): Stevens, Robert
• Application Number:
  US15/493,032
• Publication Number:
  US 20180307687A1
• Time in Patent Office: 964 Days
• Field of Search: 707728
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34

Abstract

Embodiments are disclosed for a query processing system configured to receive user submitted questions and provide answers to the user submitted questions. In an example embodiment a system receives a user query from a user. The system then determines if a response is available that matches an intent of the user query above a threshold confidence level. The systems then prompts the user for permission to enable access to the user query to other users. If permission is granted, the system then enables access to the user query to the user users. Responses to the user query received from one or more of the other users are then output to the user, for example, via a computing device or an appliance. Further, in some embodiments, feedback received from the user regarding output responses can be used to train machine learning models associated with the system.

Claims

1. A system comprising:

a processing unit; and

a memory unit communicatively coupled to the processing unit, the memory unit including instructions, which when executed by the processing unit, cause the system to;

receive a first user query from a first user;

determine if a response is available that matches an intent of the first user query above a threshold level of confidence;

enable access to the first user query to a plurality of other users if an automated response is not available that matches the intent of the first user query above the threshold level of confidence;

receive one or more user responses to the first user query from one or more of the plurality of other users;

output, to the first user, the one or more user responses to the first user query;

receive feedback from the first user regarding the output of the one or more user responses, wherein the feedback includes an indication that a particular response, from the one or more user responses, sufficiently answers the first user query;

store data corresponding to the particular response in a data repository

wherein storing the data corresponding to the particular response in the data repository includes applying a tag to the data corresponding to the particular response, the tag indicative that the particular response is supplied by another user;

receive a second user query from a second user, the second user query similar to the first user query;

determine, using a machine learning model, that the particular response to the first user query matches an intent of the second user query above the threshold level of confidence; and

output the particular response to the second user, the output including the tag indicative that the particular response is supplied by another user.

2. The system of claim 1, wherein receiving the first user query from the first user comprises:

receiving a message from the first user; and

interpreting, using natural language processing, the received message as the first user query.

3. The system of claim 2, wherein the message received from the first user includes a text message and/or an audio message.

4. The system of claim 1, wherein enabling access to the first user query to the plurality of other users includes:

prompting the first user for permission to enable access to the first user query to other users; and

receiving a confirmation indicative of permission to enable access to the first user query to other users.

5. The system of claim 1, wherein the memory unit includes further instructions, which when executed by the processing unit, cause the system to further:

train the machine learning model based on the feedback received from the first user.

6. The system of claim 1, wherein storing the data corresponding to the particular response in the data repository includes:

associate the data corresponding to the particular response with data corresponding to the first user query and/or data corresponding to one or more other user queries that are similar to the first user query.

7. The system of claim 1, wherein enabling access to the first user query includes posting the first user query using any of:

a chatbot;

an online forum;

or

a social media platform.

8. The system of claim 1, wherein the first user query relates to any of:

food;

food preparation;

or

operation of a cooking appliance.

9. The system of claim 1, wherein the message is received from the first user via any of:

a client computing device;

or

a cooking appliance.

10. The system of claim 1, wherein the one or more responses are output to the first user via any of:

a client computing device;

or

a cooking appliance.

11. A method comprising:

receiving a first user query from a first user;

determining if a response is available that matches an intent of the first user query above a threshold level of confidence;

prompting the first user for permission to enable access to the first user query to a plurality of other users if a response is not available that matches the intent of the first user query above the threshold level of confidence;

enabling access to the first user query to the plurality of other users in response to receiving a confirmation of permission from the first user;

receiving one or more user responses to the first user query from one or more of the plurality of other users;

outputting, to the first user, the one or more user responses to the first user query;

receiving feedback from the first user regarding the output of the one or more user responses;

determining, based on the received feedback from the first user, that a particular user response of the output one or more user responses answers the first user query;

receiving a second user query from a second user, the second user query similar to the first user query;

determining that the particular user response to the first user query matches an intent of the second user query above the threshold level of confidence; and

outputting the particular user response to the second user.

12. The method of claim 11, wherein receiving the first user query from the first user includes:

receiving a message from the first user; and

interpreting, using natural language processing, the received message as the first user query.

13. The method of claim 11, further comprising:

training a machine learning model based on the feedback received from the first user.

14. The method of claim 11, further comprising:

storing data corresponding to user query and one or more of the output user responses in a repository based on the received feedback from the first user.

15. The method of claim 11, wherein the first user query relates to any of:

food;

food preparation;

or

operation of a cooking appliance.

16. A food preparation guidance system comprising:

a network interface through which to communicate, via a computer network, with a client computing device and a cooking appliance, the food preparation guidance system operable to receive inputs from the client computing device and control a cooking process by the cooking appliance based on the inputs from the user computing device;

a processing unit communicatively coupled to the network interface; and

a memory unit communicatively coupled to the processing unit, the memory unit including instructions, which when executed by the processing unit, cause the food preparation guidance system to;

receive, via the network interface, a message from a user;

interpret, using natural language processing, the received message as a user query corresponding to food preparation;

determine if a response is available that matches an intent of the user query above a threshold level of confidence;

prompt the user to enable access to the user query to a plurality of other users if a response is not available that matches the intent of the first user query above the threshold level of confidence;

enable access to the user query to the plurality of other users in response to receiving a confirmation of permission from the user;

receive one or more user responses to the user query from one or more of the plurality of other users;

output, to the first user, the received one or more user responses to the user query; and

receive feedback, from the user, regarding the output one or more user responses, the feedback including an indication that a particular user response, of the received one or more user responses, satisfactorily answers the user query corresponding to food preparation.

17. The food preparation guidance system of claim 16, wherein the memory unit includes further instructions, which when executed by the processing unit, cause the food preparation guidance system to:

train a machine learning model based on the feedback received from the user.

18. The food preparation guidance system of claim 16, wherein enabling access to the user query includes posting the user query using any of:

a chatbot;

an online forum;

or

a social media platform.

19. The food preparation guidance system of claim 16, wherein the message received from the user includes a text message and/or an audio message.

20. The food preparation guidance system of claim 16, wherein the memory unit includes further instructions, which when executed by the processing unit, cause the food preparation guidance system to:

output, to the user, a particular available response to the user query before prompting the user to enable access to the user query to the plurality of other users, if the particular available response matches the intent of the user query above the threshold level of confidence.

21. The food preparation guidance system of claim 16, wherein the message is received via any of:

the client computing device;

or

the cooking appliance.

22. The food preparation guidance system of claim 16, wherein the one or more received user responses are output to the user via any of:

the client computing device;

or

the cooking appliance.

23. The food preparation guidance system of claim 16, wherein the user query relates to any of:

food;

food preparation;

or

operation of the cooking appliance.