Granted Patent B1

US 11,481,559 · App. 16/716,335 · Granted Oct 25, 2022

Inventors: Azita Asefi (Vacaville, CA); Suhas D. Sankolli (Charlotte, NC); Senthil K. Subramaniam (San Ramon, CA); Alexander Davakis (Ormond Beach, FL); Rupali S. Shah (San Jose, CA); Alan W. Hecht (San Francisco, CA)

Assignee: Wells Fargo Bank, N.A.

G06F40/30G06F9/543G06F40/247G06F40/295

Abstract

Claims (77)

1. A system for delivering an interactive natural language chatbot session, the system comprising:

at least one processor; and

memory including instructions that, when executed by the at least one processor, cause the at least one processor to perform operations to:

obtain a first dialogue configuration ruleset for a first intent for a first entity;

receive first user input including a first text string from a chatbot session;

identify the first intent from the chatbot session through evaluation of the first text string by a natural language processing engine and the first dialogue configuration ruleset;

generate a dialogue stack for the chatbot session that includes a first state comprising the first intent and a corresponding first progress indicator based on the first dialogue configuration ruleset;

receive second user input that includes a second text string from the chatbot session;

identify a second intent from the chatbot session through evaluation of the second text string by the natural language processing engine and a second dialogue configuration ruleset;

determine that the second intent is different than the first intent;

update the first progress indicator in the dialogue stack to indicate that processing of the first intent has been paused;

generate a second state comprising the second intent and a corresponding second progress indicator based on the second dialogue configuration ruleset in the dialogue stack;

display, in the chatbot session, a second response to the second user input based on the second dialogue configuration ruleset;

update the second progress indicator based on a workflow for the second intent included in the second dialogue configuration ruleset;

send a prompt to continue with the first intent;

determine that there are no intervening intents awaiting completion based on progress indicators within the dialogue stack;

receive an indication to continue with the first intent;

update the first progress indicator to indicate that the first intent is ready based on determining that there are no intervening intents awaiting completion and receiving the indication to continue with the first intent; and

update the first state and the second state based on the second response.

2. The system of claim 1 , the memory further comprising instructions that cause the at least one processor to perform operations to:

present a configuration console to a dialogue designer, wherein the configuration console includes an entity creation user interface, an intent creation user interface, and an instruction user interface;

receive a set of user inputs from the configuration console; and

generate the first dialogue configuration ruleset using the received set of user inputs.

3. The system of claim 2 , wherein the configuration console includes a design-time application programmer interface (API) configuration user interface element and a connection between the natural language processing engine and the first dialogue configuration ruleset is established using the API configuration user interface element.

4. The system of claim 1 , wherein the natural language processing engine is a third-party natural language processor.

5. The system of claim 1 , wherein the first state indicates a processing state of the first intent.

6. The system of claim 1 , wherein the first dialogue configuration ruleset includes at least one of a set of synonyms for the entity, a set of utterances for the intent, and an association between the entity and the intent.

7. At least one non-transitory machine-readable medium for instructions for delivering an interactive natural language chatbot session that, when executed by at least one processor, cause the at least one processor to perform operations to:

obtain a first dialogue configuration ruleset for a first intent for a first entity;

receive first user input including a first text string from a chatbot session;

identify the first intent from the chatbot session through evaluation of the first text string by a natural language processing engine and the first dialogue configuration ruleset;

generate a dialogue stack for the chatbot session that includes a first state comprising the first intent and a corresponding first progress indicator based on the first dialogue configuration ruleset;

receive second user input that includes a second text string from the chatbot session;

identify a second intent from the chatbot session through evaluation of the second text string by the natural language processing engine and a second dialogue configuration ruleset;

determine that the second intent is different than the first intent;

update the first progress indicator in the dialogue stack to indicate that processing of the first intent has been paused;

generate a second state comprising the second intent and a corresponding second progress indicator based on the second dialogue configuration ruleset in the dialogue stack;

display, in the chatbot session, a second response to the second user input based on the second dialogue configuration ruleset;

update the second progress indicator based on a workflow for the second intent included in the second dialogue configuration ruleset;

send a prompt to continue with the first intent;

determine that there are no intervening intents awaiting completion based on progress indicators within the dialogue stack,

receive an indication to continue with the first intent;

update the first progress indicator to indicate that the first intent is ready based on determining that there are no intervening intents awaiting completion and receiving the indication to continue with the first intent; and

update the first state and the second state based on the second response.

8. The at least one non-transitory machine-readable medium of claim 7 , further comprising instructions that cause the at least one processor to perform operations to:

present a configuration console to a dialogue designer, wherein the configuration console includes an entity creation user interface, an intent creation user interface, and an instruction user interface;

receive a set of user inputs from the configuration console; and

generate the first dialogue configuration ruleset using the received set of user inputs.

9. The at least one non-transitory machine-readable medium of claim 8 , wherein the configuration console includes a design-time application programmer interface (API) configuration user interface element and a connection between the natural language processing engine and the first dialogue configuration ruleset is established using the API configuration user interface element.

10. The at least one non-transitory machine-readable medium of claim 7 , wherein the natural language processing engine is a third-party natural language processor.

11. The at least one non-transitory machine-readable medium of claim 7 , wherein the first state indicates a processing state of the first intent.

12. The at least one non-transitory machine-readable medium of claim 7 , wherein the first dialogue configuration ruleset includes at least one of a set of synonyms for the entity, a set of utterances for the intent, and an association between the entity and the intent.

13. A method for delivering an interactive natural language chatbot session, the method comprising:

obtaining a first dialogue configuration ruleset for a first intent for a first entity;

receiving first user input including a first text string from a chatbot session;

identifying the first intent from the chatbot session by evaluating the first text string using a natural language processing engine and the first dialogue configuration ruleset;

generating a dialogue stack for the chatbot session including a first state that comprises the first intent and a corresponding first progress indicator based on the first dialogue configuration ruleset;

receiving second user input that includes a second text string from the chatbot session;

identifying a second intent from the chatbot session through evaluation of the second text string by the natural language processing engine and a second dialogue configuration ruleset;

determining that the second intent is different than the first intent;

updating the first progress indicator in the dialogue stack to indicate that processing of the first intent has been paused;

generating a second state comprising the second intent and a corresponding second progress indicator based on the second dialogue configuration ruleset in the dialogue stack;

displaying, in the chatbot session, a second response to the second user input based on the second dialogue configuration ruleset;

updating the second progress indicator based on a workflow for the second intent included in the second dialogue configuration ruleset;

sending a prompt to continue with the first intent;

determining that there are no intervening intents awaiting completion based on progress indicators within the dialogue stack;

receiving an indication to continue with the first intent;

updating the first progress indicator to indicate that the first intent is ready based on determining that there are no intervening intents awaiting completion and receiving the indication to continue with the first intent; and

updating the first state and the second state based on the second response.

14. The method of claim 13 , further comprising:

presenting a configuration console to a dialogue designer, the configuration console including an entity creation user interface, an intent creation user interface, and an instruction user interface;

receiving a set of user inputs from the configuration console; and

generating the first dialogue configuration ruleset using the received set of user inputs.

15. The method of claim 14 , wherein the configuration console includes a design-time application programmer interface (API) configuration user interface element and a connection between the natural language processing engine and the first dialogue configuration ruleset is established using the API configuration user interface element.

16. The method of claim 13 , wherein the natural language processing engine is a third-party natural language processor.

17. The method of claim 13 ; wherein the first state indicates a processing state of the first intent.

18. The method of claim 13 , wherein the first dialogue configuration ruleset includes at least one of a set of synonyms for the entity, a set of utterances for the intent, and an association between the entity and the intent.

Assignments (1)

Continuity (1)

Cited By (11)

US 12,197,875 US 12,288,036 US 12,333,404 US 12,386,434 US 12,417,081 US 12,477,470 US 12,511,642 US 12,541,653 US 12,608,713 US 12,669,873 US 12,748,568