SMART MESSAGING FOR COMPUTER-IMPLEMENTED DEVICES

• US 20170339076A1
• Filed: 08/09/2017
• Published: 11/23/2017
• Est. Priority Date: 02/14/2012
• Status: Active Grant

• Current Assignee: Salesforce.com, Inc.
• Original Assignee: Salesforce.com, Inc.
• Inventors: PATIL, DIPAK
• Granted Patent: US 10,205,684 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 40/274 Converting codes to words; Guess-ahead of partial word inputs
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 10/107 Computer-aided management of electronic mailing [e-mailing]
• CPC Class Codes: H04L 51/02 using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages

Abstract

Various computer-implemented systems and methods are provided here for purposes of smart messaging. A machine learning application can obtain message context data from a user device, and process the message context data to determine a predicted conversation pattern of the conversation that is taking place between the particular users. The message context data is indicative of context of a conversation taking place between particular users during a messaging session. Based on the predicted conversation pattern, the machine learning application can then determine options for suggested responses that are predicted to compete at least part of a response message as part of the conversation. Each of the options for suggested responses corresponds to the message context data indicative of the context of the conversation. In response to selection of one of the suggested responses, that suggested response can be displayed within a message field.

Claims

1. A computer-implemented method of smart messaging, the method comprising:

obtaining, at a machine learning application from a user device, message context data indicative of context of a conversation taking place between particular users during a messaging session;

processing the message context data, at the machine learning application, to determine a predicted conversation pattern of the conversation that is taking place between the particular users;

determining, at the machine learning application based on the predicted conversation pattern, options for suggested responses that are predicted to compete at least part of a response message as part of the conversation, wherein each of the options for suggested responses corresponds to the message context data indicative of the context of the conversation;

displaying one of the suggested responses within a message field in response to selection of that suggested response.

2. The method of claim 1, wherein the machine learning application is implemented at the user device.

3. The method of claim 1, wherein the machine learning application is implemented at a computing system that is communicatively coupled to the user device via a network.

4. The method of claim 1, wherein displaying comprises:

displaying the one of the suggested responses within the message field at the user device.

5. The method of claim 1, wherein the user device is a first user device, and wherein displaying comprises:

displaying the one of the suggested responses within the message field at a second user device that is part of the messaging session with the first user device.

6. The method of claim 1, wherein displaying further comprises:

automatically populating the message field with the one of the suggested responses in response to the selection input indicating selection of that suggested response; and

after automatically populating, sending the response message from the user device, the response message including the one of the suggested responses that has been automatically populated in the message field.

7. The method of claim 1, wherein the message is an email message, and wherein the message context data comprises at least one of:

text from a subject line of a prior email message of the conversation and message content of the prior email message of the conversation.

8. The method of claim 1, wherein the message is a text message, and wherein the message context data comprises:

message content of a prior text message of the conversation.

9. The method of claim 1, wherein processing comprises:

processing the message context data, at the machine learning application, to determine, based on prior conversation patterns between those particular users, a predicted messaging scenario of the conversation between the particular users, wherein the predicted messaging scenario includes;

the predicted conversation pattern of the conversation that is taking place between the particular users who are participants in the conversation.

10. The method of claim 1, wherein the machine learning application is configured to learn certain characteristics of conversations comprising:

conversation patterns, commonly used phrases, and whether a conversation uses formal or informal language, and wherein the method further comprises;

updating, after processing the message context data, the machine learning application such that a predictive nature of the machine learning application improves suggested responses over time.

11. The method of claim 1, wherein each of the options for suggested responses comprise:

different recommended text that corresponds to the message context data and indicates suggested language to complete at least part of the response message; and

wherein the method further comprises;

providing the options for suggested responses for presentation at the user device;

selecting one of the options for suggested responses, resulting in a selected option; and

wherein automatically populating the message field comprises;

automatically populating the message field with the selected option.

12. The method of claim 11, wherein:

the options for suggested responses are provided in a user-selectable format; and

the selecting is performed in response to user selection of one of the options for suggested responses.

13. The method of claim 1, wherein the message context data comprises at least one of:

an identification of at least one participant of the messaging session;

a time associated with the messaging session;

a current date of the messaging session;

a current day of the messaging session;

location information that indicates a location of the user device;

historical message content associated with the messaging session;

historical message content of previous conversations between the particular users; and

calendar information associated with at least one participant of the messaging session.

14. A computer-implemented method of smart messaging, the method comprising:

initiating a messaging session between a first user of first user device and a second user of a second user device, wherein the first user and the second user are participants in a conversation taking place as part of the messaging session;

receiving, at a machine learning application from the first user device or the second user device, message context data associated with the messaging session that indicates a characteristic of the conversation;

determining, at the machine learning application based on the characteristic of the conversation that is indicative of context of the conversation, a predicted conversation pattern of the conversation;

determining, at the machine learning application based on the predicted conversation pattern, options for suggested responses that are predicted to compete at least part of a response message as part of the conversation to complete, wherein each of the options for suggested responses corresponds to the characteristic of the conversation indicative of the context of the conversation; and

providing, to the first user device or to the second user device, one of the suggested responses; and

displaying that suggested response within a message field at the first user device or the second user device.

15. The system of claim 14, wherein the machine learning application is implemented at the first user device.

16. The system of claim 14, wherein the machine learning application is implemented at the second user device.

17. The system of claim 14, further comprising:

a computing system that is communicatively coupled to the first user device and the second user device via a network, wherein the machine learning application is implemented at the computing system.

18. The system of claim 14, wherein determining, at a machine learning application based on a characteristic of the conversation that is indicative of context of the conversation, a predicted conversation pattern of the conversation, comprises:

determining, at the machine learning application based on the characteristic of the conversation that is indicative of context of the conversation and prior conversation patterns between those participants of the conversation, a predicted messaging scenario that includes the predicted conversation pattern of the conversation that is taking place between the first user and the second user during the messaging session, wherein the first user and the second user are users, and wherein the suggested responses are influenced by the predicted messaging scenario.

19. The system of claim 14, wherein each of the options for suggested responses comprise:

different recommended text that corresponds to the message context data and indicates suggested language to complete at least part of the response message.

20. A computer-implemented system, comprising:

a first user device; and

a system configured to execute a machine learning application, the system comprising;

a processor and a memory, wherein the memory comprises a non-transitory computer-readable-medium having computer-executable instructions for the machine learning application stored therein that, when executed by the processor, cause the processor to;

obtain, at the machine learning application from the first user device, message context data indicative of context of a conversation taking place between a first user of the first user device and a second user of a second user device;

analyze the message context data, at the machine learning application, to determine a predicted conversation pattern of the conversation that is taking place between the first user and the second user who are participating in the conversation;

generate, at the machine learning application based on the predicted conversation pattern, options for suggested responses that are predicted to compete at least part of a response message as part of the conversation, wherein each of the options for suggested responses corresponds to the message context data indicative of the context of the conversation; and

provide one of the suggested responses to either the first user device or the second user device.

21. The system of claim 20, wherein the system is part of the first user device, and wherein the machine learning application is implemented at the first user device, wherein the one of the suggested responses is presented within a message field at the first user device, and automatically populates the message field with one of the suggested responses in response to selection of that suggested response.

22. The system of claim 20, wherein the system is the second user device that is participating in the messaging session with the first user device, wherein the machine learning application is implemented at the second user device, wherein the one of the suggested responses is presented within a message field at the second user device, and automatically populates the message field with one of the suggested responses in response to selection of that suggested response.

23. The system of claim 20, wherein the system is a computing system that is communicatively coupled to the first user device via a network, wherein the machine learning application is implemented at the computing system, wherein the one of the suggested responses is presented within a message field at the first user device or the second user device, and automatically populates the message field with one of the suggested responses in response to selection of that suggested response.