Real-time guidance for content collection

• US 10,248,716 B2
• Filed: 02/19/2016
• Issued: 04/02/2019
• Est. Priority Date: 02/19/2016
• Status: Active Grant

• Current Assignee: Accenture Global Solutions Limited (Accenture PLC)
• Original Assignee: Accenture Global Solutions Limited (Accenture PLC)
• Inventors: Neland, Joshua
• Primary Examiner(s): Vo, Truong V
• Application Number:
  US15/048,664
• Publication Number:
  US 20170242920A1
• Time in Patent Office: 1,138 Days
• CPC Class Codes: G06F 16/3322 using system suggestions G06F16/3325 takes precedence
• CPC Class Codes: G06F 16/3346 using probabilistic model
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 40/08 Insurance

Abstract

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for providing real-time guidance for content collection. One of the methods includes receiving user input from a user through a user interface presentation, determining, from the received user input using a first model, one or more provided data elements occurring in the user input, determining, from the one or more provided data elements occurring in the user input using a second model, one or more intended tasks, determining, for each intended task of the one or more intended tasks using a third model, one or more suggested data elements, ranking the one or more suggested data elements, and updating the user interface presentation with a user interface element suggesting that the user provide the one or more needed data elements.

Claims

1. A computer-implemented method comprising:

receiving user input from a user through a user interface, the user input comprising a plurality of terms;

determining, by providing the user input as input to a first model, one or more data elements in the user input and one or more data element labels that each describe a respective data element, wherein each data element includes one or more terms of the plurality of terms;

determining, from among a plurality of tasks and by providing the one or more data elements and the one or more data element labels that each describe a respective data element as input to a second model, one or more predicted tasks for the provided user input, wherein each task of the plurality of tasks is associated with a respective plurality of data elements and data element labels of each data element for completing the task;

determining, by providing each task of the one or more tasks as input to a third model, one or more additional data elements for one or more suggested data element labels that are not included in the one or more data elements in the user input, wherein each additional data element for each suggested data element label includes additional information that is for completing the task and that has not yet entered been entered;

and

updating the user interface with a user interface element suggesting that the user provide the one or more additional data elements.

2. The method of claim 1, wherein updating the user interface presentation comprises updating the user interface presentation without receiving a user submission action.

3. The method of claim 1, wherein receiving user input from the user through a user interface comprises receiving unstructured text through a text box of the user interface.

4. The method of claim 1, wherein receiving user input from the user through a user interface comprises receiving voice input from the user.

5. The method of claim 1, further comprising:

ranking the one or more suggested data elements

wherein the user interface element includes the ranking of each suggested data element, and

wherein ranking the one or more suggested data elements comprises;

determining, for each suggested data element, a likelihood that the suggested data element will be omitted by the user for the one or more predicted tasks; and

ranking the one or more suggested data elements according to the likelihood that the suggested data element will be omitted by the user for the one or more predicted tasks.

6. The method of claim 5, wherein ranking the one or more suggested data elements comprises:

determining, by providing each suggested data element as input to a fourth model, a score that represents a measure of accuracy or quality of predicted task if the suggested data element is added; and

ranking the one or more suggested data elements according to the measure of accuracy or quality of the predicted task.

7. The method of claim 6, further comprising:

obtaining training examples that indicate, for a particular data element and a particular task, a change in a measure of accuracy or quality of the predicted task if the particular data element is used to perform the predicted task; and

training the fourth model using the obtained training examples.

8. The method of claim 6, wherein the score represents a likelihood of a predicted task having a particular outcome.

9. A system comprising:

one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising;

receiving user input from a user through a user interface, the user input comprising a plurality of terms;

determining, by providing the user input as input to a first model, one or more data elements in the user input and one or more data element labels that each describe a respective data element, wherein each data element includes one or more terms of the plurality of terms;

determining, from among a plurality of tasks and by providing the one or more data elements and the one or more data element labels that each describe a respective data element as input to a second model, one or more predicted tasks for the provided user input, wherein each task of the plurality of tasks is associated with a respective plurality of data elements and data element labels of each data element for completing the task;

determining, by providing each task of the one or more tasks as input to a third model, one or more additional data elements for one or more suggested data element labels that are not included in the one or more data elements in the user input, wherein each additional data element for each suggested data element label includes additional information that is for completing the task and that has not yet entered been entered;

and

updating the user interface with a user interface element suggesting that the user provide the one or more additional data elements.

10. The system of claim 9, wherein updating the user interface presentation comprises updating the user interface presentation without receiving a user submission action.

11. The system of claim 9, wherein receiving user input from the user through a user interface comprises receiving unstructured text through a text box of the user interface.

12. The system of claim 9, wherein receiving user input from the user through a user interface comprises receiving voice input from the user.

13. The system of claim 9, further comprising:

ranking the one or more suggested data elements

wherein the user interface element includes the ranking of each suggested data element, and

wherein ranking the one or more suggested data elements comprises;

determining, for each suggested data element, a likelihood that the suggested data element will be omitted by the user for the one or more predicted tasks; and

ranking the one or more suggested data elements according to the likelihood that the suggested data element will be omitted by the user for the one or more predicted tasks.

14. The system of claim 13, wherein ranking the one or more suggested data elements comprises:

determining, by providing each suggested data element as input to a fourth model, a score that represents a measure of accuracy or quality of predicted task if the suggested data element is added; and

ranking the one or more suggested data elements according to the measure of accuracy or quality of the predicted task.

15. The system of claim 14, wherein the operations further comprise:

obtaining training examples that indicate, for a particular data element and a particular task, a change in a measure of accuracy or quality of the predicted task if the particular data element is used to perform the predicted task; and

training the fourth model using the obtained training examples.

16. The system of claim 14, wherein the score represents a likelihood of a predicted task having a particular outcome.

17. A computer program product, encoded on one or more non-transitory computer storage media, comprising instructions that when executed by one or more computers cause the one or more computers to perform operations comprising:

receiving user input from a user through a user interface, the user input comprising a plurality of terms;

determining, by providing the user input as input to a first model, one or more data elements in the user input and one or more data element labels that each describe a respective data element, wherein each data element includes one or more terms of the plurality of terms;

determining, from among a plurality of tasks and by providing the one or more data elements and the one or more data element labels that each describe a respective data element as input to a second model, one or more predicted tasks for the provided user input, wherein each task of the plurality of tasks is associated with a respective plurality of data elements and data element labels of each data element for completing the task;

determining, by providing each task of the one or more tasks as input to a third model, one or more additional data elements for one or more suggested data element labels that are not included in the one or more data elements in the user input, wherein each additional data element for each suggested data element label includes additional information that is for completing the task and that has not yet entered been entered;

and

updating the user interface with a user interface element suggesting that the user provide the one or more additional data elements.

18. The computer program product of claim 17, wherein updating the user interface presentation comprises updating the user interface presentation without receiving a user submission action.

19. The computer program product of claim 17, wherein receiving user input from the user through a user interface comprises receiving unstructured text through a text box of the user interface.

20. The computer program product of claim 17, wherein receiving user input from the user through a user interface comprises receiving voice input from the user.

21. The computer program product of claim 17, further comprising:

ranking the one or more suggested data elements

wherein the user interface element includes the ranking of each suggested data element, and

wherein ranking the one or more suggested data elements comprises;

determining, for each suggested data element, a likelihood that the suggested data element will be omitted by the user for the one or more predicted tasks; and

ranking the one or more suggested data elements according to the likelihood that the suggested data element will be omitted by the user for the one or more predicted tasks.

22. The computer program product of claim 21, wherein ranking the one or more suggested data elements comprises:

determining, by providing each suggested data element as input to a fourth model, a score that represents a measure of accuracy or quality of predicted task if the suggested data element is added; and

ranking the one or more suggested data elements according to the measure of accuracy or quality of the predicted task.

23. The computer program product of claim 22, wherein the operations further comprise:

obtaining training examples that indicate, for a particular data element and a particular task, a change in a measure of accuracy or quality of the predicted task if the particular data element is used to perform the predicted task; and

training the fourth model using the obtained training examples.

24. The computer program product of claim 22, wherein the score represents a likelihood of a predicted task having a particular outcome.