Method and system for conveying an example in a natural language understanding application

• US 10,192,543 B2
• Filed: 05/10/2016
• Issued: 01/29/2019
• Est. Priority Date: 12/15/2005
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Inventors: Balchandran, Rajesh, Boyer, Linda M., Lewis, James R., Metz, Brent D.
• Primary Examiner(s): Ortiz-Sanchez, Michael
• Application Number:
  US15/151,277
• Publication Number:
  US 20160253991A1
• Time in Patent Office: 994 Days
• CPC Class Codes: G06F 40/40 Processing or translation of natural language natural language analysis G06F40/20; semantic analysis G06F40/30
• CPC Class Codes: G10L 15/063 Training
• CPC Class Codes: G10L 15/18 using natural language modelling
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models
• CPC Class Codes: G10L 2015/0638 Interactive procedures
• CPC Class Codes: H04M 2203/2061 Language aspects
• CPC Class Codes: H04M 3/4936 Speech interaction details speech recognition per se G10L15/00

Abstract

A method (300) and system (100) is provided to add the creation of examples at a developer level in the generation of Natural Language Understanding (NLU) models, tying the examples into a NLU sentence database (130), automatically validating (310) a correct outcome of using the examples, and automatically resolving (316) problems the user has using the examples. The method (300) can convey examples of what a caller can say to a Natural Language Understanding (NLU) application. The method includes entering at least one example associated with an existing routing destination, and ensuring an NLU model correctly interprets the example unambiguously for correctly routing a call to the routing destination. The method can include presenting the example sentence in a help message (126) within an NLU dialogue as an example of what a caller can say for connecting the caller to a desired routing destination. The method can also include presented a failure dialogue for displaying at least one example that failed to be properly interpreted to ensure that ambiguous or incorrect examples are not presented in a help message.

Claims

1. A system for facilitating development, in a natural language understanding (NLU) application development environment, of an NLU model associated with an NLU application, the system comprising:

at least one processor;

a database storing information used for training one or more NLU models;

at least one non-transitory computer-readable storage medium encoded with instructions that, when executed by the at least one processor, cause the at least one processor to perform;

obtaining at least one expected user entry and a corresponding desired routing destination;

applying the NLU model to the at least one expected user entry to determine whether the NLU model associates the at least one expected user entry with the desired routing destination;

when it is determined that the NLU model associates the at least one expected user entry with the desired routing destination, selecting the at least one expected user entry for presentation to a user in a help message of the NLU application as an example of input the user could provide to be routed to the desired routing destination; and

when it is determined that the NLU model does not associate the at least one expected user entry with the desired routing destination;

training the NLU model using training data accessed in the database to associate the at least one expected user entry with the desired routing destination; and

validating that the trained NLU model associates the at least one expected user entry with the desired routing destination.

2. The system of claim 1, wherein the applying comprises:

interpreting, via the NLU model, the at least one expected user entry to determine an actual routing destination for the at least one expected user entry; and

determining whether the NLU model associates the at least one expected user entry with the desired routing destination by comparing the actual routing destination to the desired routing destination.

3. The system of claim 2, wherein, when it is determined that the NLU model does not associate the at least one expected user entry with the desired routing destination, the instructions further cause the at least one processor to perform:

presenting a failure dialog to a developer of the NLU application indicating that the actual routing destination does not match the desired routing destination.

4. The system of claim 1, wherein training the NLU model to associate the at least one expected user entry with the desired routing destination comprises:

training the NLU model in accordance with a specified training frequency and/or a statistical weighting for the at least one expected user entry.

5. The system of claim 1, wherein when it is determined that the NLU model does not associate the at least one expected user entry with the desired routing destination, the instructions further cause the at least one processor to perform:

adding the at least one expected user entry to an NLU entry data set associated with the NLU model; and

training the NLU model using the data in the NLU entry data set.

6. The system of claim 1, wherein the instructions further cause the at least one processor to perform:

presenting the at least one expected user entry in the help message of the NLU application.

7. The system of claim 6, wherein the help message comprises a preamble followed by the at least one expected user entry.

8. A method for facilitating, in a natural language understanding (NLU) application development environment, of an NLU model associated with an NLU application, the method comprising:

using at least one processor and a database storing information used for training one or more NLU models to perform;

obtaining at least one expected user entry and a corresponding desired routing destination;

applying the NLU model to the at least one expected user entry to determine whether the NLU model associates the at least one expected user entry with the desired routing destination;

when it is determined that the NLU model associates the at least one expected user entry with the desired routing destination, selecting the at least one expected user entry for presentation to a user during in a help message of the NLU application as an example of input the user could provide to be routed to the desired routing destination; and

when it is determined that the NLU model does not associate the at least one expected user entry with the desired routing destination;

training the NLU model using training data accessed in the database to associate the at least one expected user entry with the desired routing destination; and

validating that the trained NLU model associates the at least one expected user entry with the desired routing destination.

9. The method of claim 8, wherein the applying comprises:

interpreting, via the NLU model, the at least one expected user entry to determine an actual routing destination for the at least one expected user entry; and

determining whether the NLU model associates the at least one expected user entry with the desired routing destination by comparing the actual routing destination to the desired routing destination.

10. The method of claim 9, wherein, when it is determined that the NLU model does not associate the at least one expected user entry with the desired routing destination, further comprising:

presenting a failure dialog to a developer of the NLU application indicating that the actual routing destination does not match the desired routing destination.

11. The method of claim 8, wherein training the NLU model to associate the at least one expected user entry with the desired routing destination comprises:

training the NLU model in accordance with a specified training frequency and/or a statistical weighting for the at least one expected user entry.

12. The method of claim 8, wherein when it is determined that the NLU model does not associate the at least one expected user entry with the desired routing destination, the method further comprises:

adding the at least one expected user entry to an NLU entry data set associated with the NLU model; and

training the NLU model using the data in the NLU entry data set.

13. The method of claim 8, wherein the method further comprises:

presenting the at least one expected user entry in the help message of the NLU application.

14. The method of claim 13, wherein the help message comprises a preamble followed by the at least one expected user entry.

15. At least one non-transitory computer-readable storage medium encoded with instructions that, when executed by a computer system, cause the computer system to perform a method for facilitating, in a natural language understanding (NLU) application development environment, of an NLU model associated with an NLU application, the method comprising:

obtaining at least one expected user entry and a corresponding desired routing destination;

applying the NLU model to the at least one expected user entry to determine whether the NLU model associates the at least one expected user entry with the desired routing destination;

when it is determined that the NLU model associates the at least one expected user entry with the desired routing destination, selecting the at least one expected user entry for presentation to a user during in a help message of the NLU application as an example of input the user could provide to be routed to the desired routing destination; and

when it is determined that the NLU model does not associate the at least one expected user entry with the desired routing destination;

training the NLU model using training data accessed in an NLU database to associate the at least one expected user entry with the desired routing destination; and

validating that the trained NLU model associates the at least one expected user entry with the desired routing destination.

16. The at least one non-transitory computer-readable storage medium of claim 15, wherein the applying comprises:

interpreting, via the NLU model, the at least one expected user entry to determine an actual routing destination for the at least one expected user entry; and

determining whether the NLU model associates the at least one expected user entry with the desired routing destination by comparing the actual routing destination to the desired routing destination.

17. The at least one non-transitory computer-readable storage medium of claim 16, wherein, when it is determined that the NLU model does not associate the at least one expected user entry with the desired routing destination, the method further comprises:

presenting a failure dialog to a developer of the NLU application indicating that the actual routing destination does not match the desired routing destination.

18. The at least one non-transitory computer-readable storage medium of claim 15, wherein training the NLU model to associate the at least one expected user entry with the desired routing destination comprises:

training the NLU model in accordance with a specified training frequency and/or a statistical weighting for the at least one expected user entry.

19. The at least one non-transitory computer-readable storage medium of claim 15, wherein when it is determined that the NLU model does not associate the at least one expected user entry with the desired routing destination, the method further comprises:

adding the at least one expected user entry to an NLU entry data set associated with the NLU model; and

training the NLU model using the data in the NLU entry data set.

20. The at least one non-transitory computer-readable storage medium of claim 15, wherein the method further comprises:

presenting the at least one expected user entry in the help message of the NLU application.