Granted Patent B2

US 11,170,161 · App. 15/977,641 · Granted Nov 9, 2021

Abstract

Claims (63)

1. A method comprising:

receiving, by a computing device, a first document image of a paper-based form document;

generating, based on the first document image, a chatbot configured to engage in a conversation with a first user via a chat interface;

receiving, by the chatbot and via the chat interface, a message from a first user device associated with the first user;

based on a determination that the message corresponds to the first document image, engaging, by the chatbot and based on a first conversation model associated with the paper-based form document, in a conversation with the first user;

determining, based on the conversation, that a response message associated with another conversation model should be sent to the first conversation model;

updating, based on a second message associated with the another conversation model, the first conversation model; and

sending, to an application associated with the paper-based form document and based on one or more responses received from the first user during the conversation, information indicated by the paper-based form document.

2. The method of claim 1 , wherein the second message is received from a second user device associated with a second user to supplement the chatbot in the conversation.

3. The method of claim 1 , wherein the chatbot is associated with the first conversation model, the method further comprising:

identifying the another conversation model corresponding to a second document image, wherein a comparison between the first document image and the second document image satisfies a similarity threshold; and

adding content from the another conversation model to the first conversation model.

4. The method of claim 3 , further comprising replacing a first content in the first conversation model with a second content from the another conversation model.

5. The method of claim 4 , wherein the replacing the first content in the first conversation model with the second content from the another conversation model comprises:

determining, based on an aggregation of user responses to the first content, a rating of the first content; and

replacing the first content with the second content based on a comparison between the rating of the first content and a rating of the second content.

6. The method of claim 1 , further comprising:

identifying text corresponding to an input location of the first document image based on a detection of a line extending from an area adjacent to the input location to an area adjacent to the text.

7. The method of claim 1 , further comprising:

generating, based on text corresponding to a location in the first document image that requires input from a second user, a question; and

initiating a dialogue with the second user, using the question.

8. The method of claim 1 , further comprising:

identifying explanatory text within the first document image; and

adding to the first conversation model associated with the chatbot, dialogue corresponding to the explanatory text.

9. The method of claim 8 , further comprising:

receiving, during the conversation, a question from the first user; and

answering, based on the dialogue corresponding to the explanatory text, the question from the first user.

10. The method of claim 8 , further comprising:

determining, by the computing device and based on a first question, a data validation model; and

after determining, based on the data validation model, that input from the first user is not valid, providing output to the first user based on the explanatory text.

11. A method comprising:

receiving, by a computing device, a first document;

generating first document text by performing optical character recognition on the first document;

identifying, by the computing device and using natural language processing, explanatory text and interrogatory text within the first document text;

generating a first conversation model based on the explanatory text and the interrogatory text; and

engaging, by the computing device and based on the first conversation model, in a conversation with a first user, wherein the conversation comprises:

sending, to a user device associated with the first user, a first question;

determining, based on the conversation, that a response message associated with another conversation model should be sent to the first conversation model; and

updating, based on the response message, the first conversation model.

12. The method of claim 11 , wherein the response message is received from a second user device associated with a second user to supplement the first conversation model.

13. The method of claim 12 , further comprising:

based on the receiving the response message, removing a question, corresponding to the response message, from the first conversation model.

14. The method of claim 12 , further comprising:

receiving a response from the first user, wherein the response is unrelated to the first conversation model;

identifying the another conversation model related to the response, wherein the another conversation model is based on a second document; and

continuing, based on the another conversation model, the conversation with the first user.

15. The method of claim 11 , further comprising:

determining a second document that is related to the first document; and

initiating a dialogue with the first user based on the another conversation model that is generated from the second document.

16. A system comprising:

a document engine that receives a computerized image of a first document, and generates, based on the first document, a first question and an instruction corresponding to the first question;

a conversation model generator that creates a first conversation model comprising the first question, the instruction, and an expected answer format for a response to the first question;

a plurality of chatbot generators, each configured to automatically create a chatbot for a different one of a plurality of different communication interfaces, wherein the chatbot is configured to engage in a conversation with a first user according to the first conversation model, wherein the conversation comprises:

sending, to a user device associated with the first user, the first question;

determining, based on the conversation, that a response message associated with another conversation model should be sent to the first conversation model; and

updating, based on the response message, the first conversation model; and

a backend handler that triggers a workflow in response to completing the conversation with the first user, wherein the workflow comprises a response of the first user to the first question.

17. The system of claim 16 wherein the plurality of different communication interfaces comprise an instant messenger and a voice-activated personal assistant.

18. The system of claim 16 , wherein the conversation model generator is configured to:

identify the another conversation model corresponding to a second document based on a comparison between the first document and the second document satisfying a similarity threshold; and

add content from the another conversation model to the first conversation model.

19. The system of claim 16 , wherein the response message is received from a second user device associated with a second user to supplement the chatbot in the conversation.

20. The system of claim 16 , wherein the document engine is configured to identify a second document that is related to the first document, and wherein the chatbot is configured to initiate a dialogue with a second user based on the another conversation model that is generated from the second document.

Assignments (11)