Method and apparatus for federated understanding

• US 20030233223A1
• Filed: 06/14/2002
• Published: 12/18/2003
• Est. Priority Date: 06/14/2002
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Wang, Kuansan
• Granted Patent: US 7,502,730 B2
• Time in Patent Office: Days
• US Class Current: 704/4
• CPC Class Codes: G06F 16/986 Document structures and storage, e.g. HTML extensions
• CPC Class Codes: G06F 40/295 Named entity recognition
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99936 Pattern matching access
• CPC Class Codes: Y10S 707/99944 Object-oriented database structure
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing

Abstract

A method and apparatus are provided for resolving user input that crosses different application domains to identify a discourse item such as a command or entity. To resolve the input, a discourse item definition is selected based on the user input. The discourse item definition is associated with a first application and comprises at least one defining entity that is to be used to identify a discourse item. An entity definition associated with a second application is selected based on the user input, wherein the entity defined by the definition can fulfill the role of the defining entity in the discourse item definition. An entity is then identified using the entity definition and the user input. The discourse item is identified based on the identified entity, the discourse item definition and the user input.

Claims

1. A method of resolving user input that crosses different application domains to identify a discourse item, the method comprising:

selecting a discourse item definition associated with a first application based on user input, the discourse item definition comprising at least one defining entity that is to be used to identify a discourse item;

selecting an entity definition associated with a second application based on the user input, the entity definition defining an entity that can fulfill the role of the defining entity in the discourse item definition;

identifying an entity based on the entity definition and at least a portion of the user input; and

identifying a discourse item based on the discourse item definition, the identified entity and the user input.

2. The method of claim 1 wherein selecting a discourse item definition comprises matching the user input to a grammar rule associated with the discourse item definition and using the grammar rule to select the discourse item definition.

3. The method of claim 2 wherein selecting an entity definition comprises matching a portion of the user input to a grammar rule associated with the entity definition and using the grammar rule to select the entity definition.

4. The method of claim 1 wherein identifying an entity comprises associating a portion of the user input with a portion of the entity definition.

5. The method of claim 4 wherein identifying an entity further comprises passing the portion of the user input associated with the entity definition to a domain expert that is capable of using the portion of the user input to identify an entity.

6. The method of claim 5 wherein the domain expert is identified in the entity definition.

7. The method of claim 6 wherein the domain expert is associated with the second application but not the first application.

8. The method of claim 5 wherein selecting a discourse item definition comprises using the discourse item definition to construct an outer discourse semantic structure.

9. The method of claim 8 wherein selecting an entity definition comprises using the entity definition to construct an entity portion of a discourse semantic structure and placing the entity portion of the discourse semantic structure in the outer discourse semantic structure.

10. The method of claim 9 wherein the domain expert identifies the entity and places the identified entity in the outer discourse semantic structure in place of the entity portion of the discourse semantic structure.

11. The method of claim 5 wherein identifying a discourse item further comprises passing the identified entity to a second domain expert that is capable of using the identified entity to identify the discourse item.

12. A computer-readable medium providing computer-executable instructions for performing steps to provide and use extensible discourse semantic definitions to resolve a user input into a discourse item, the steps comprising:

providing a first discourse semantic definition comprising a slot having an entity of a type;

allowing more than one discourse semantic definition to be provided for entities that are the same type as the slot;

constructing an initial discourse semantic structure based on the first discourse semantic definition and user input;

inserting into the initial discourse semantic structure a discourse semantic structure that is constructed based on at least one of the discourse semantic definitions for an entity of the same type as the slot and user input to produce an expanded discourse semantic structure; and

using the expanded discourse semantic structure to resolve the user input.

13. The computer-readable medium of claim 12 wherein inserting a discourse semantic structure comprises matching the user input to a grammar rule, utilizing the grammar rule to identify which of the discourse semantic definitions should be used to construct the discourse semantic structure, and constructing the discourse semantic structure from the identified discourse semantic definition.

14. The computer-readable medium of claim 12 wherein allowing more than one discourse semantic definition to be provided for entities of the same type as the slot comprises allowing discourse semantic definitions to be added that have the same entity type as the type of the slot.

15. The computer-readable medium of claim 14 wherein adding a discourse semantic definition does not change the first discourse semantic definition.

16. The computer-readable medium of claim 12 wherein allowing more than one discourse semantic definition comprises allowing a first discourse semantic definition to specify a first domain expert and a second discourse semantic definition to specify a second domain expert different from the first domain expert.

17. The computer-readable medium of claim 12 wherein allowing more than one discourse semantic definition comprises allowing a first discourse semantic definition to have a first number of slots and a second discourse semantic definition to have a second number of slots different from the first number.

18. The computer-readable medium of claim 12 wherein allowing more than one discourse semantic definition comprises allowing a first discourse semantic definition that is associated with a first application and a second discourse semantic definition that is associated with a second application.

19. The computer-readable medium of claim 12 wherein using the discourse semantic structure to resolve the user input comprises sending a portion of the user input to a domain expert identified in one of the discourse semantic definitions provided for entities of the same type as the slot.

20. The computer-readable medium of claim 19 wherein using the discourse semantic structure to resolve the user input further comprises:

receiving a change to the discourse semantic structure from the domain expert; and

using the change to the discourse semantic structure and a behavior template to generate an output to the user.

21. The computer-readable medium of claim 20 wherein the behavior template is associated with a same application that is associated with the domain expert.

22. The computer-readable medium of claim 21 wherein the behavior template associated with one application is different from the behavior template associated with another application even though both behavior templates are used to resolve an entity of the same type.

23. A method of resolving user input to identify a discourse item, the method comprising:

associating a first part of the user input with a first application;

associating a second part of the user input with a second application that is independent of the first application;

utilizing application-specific programs associated with the first application to generate user interfaces to identify a first entity from the user input;

utilizing application-specific programs associated with the second application to generate user interfaces to identify a second entity from the user input; and

using the first entity and the second entity to identify the discourse item.

24. The method of claim 23 wherein the application-specific programs comprise domain experts.

25. The method of claim 24 wherein the application-specific programs further comprise behavior templates that define user interfaces to be provided to the user.

26. The method of claim 23 wherein using the first entity and second entity to identify the semantic discourse item comprises identifying a command to be executed by a third application.