Semantic framework for natural language programming

• US 20050273335A1
• Filed: 09/15/2004
• Published: 12/08/2005
• Est. Priority Date: 04/23/2004
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Olsen, Mari J.B., Parkinson, David J., Cipollone, Domenic J., Calcagno, Michael V., Chang, Su Chin, Shahani, Ravi C.
• Granted Patent: US 8,201,139 B2
• Time in Patent Office: Days
• US Class Current: 704/257
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 8/20 Software design
• CPC Class Codes: G06F 8/315 Object-oriented languages
• CPC Class Codes: G10L 15/18 using natural language modelling
• CPC Class Codes: G10L 15/1815 Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning
• CPC Class Codes: Y10S 707/99944 Object-oriented database structure

Abstract

A framework for generating a semantic interpretation of natural language input includes an interpreter, a first set of types, and a second set of types. The interpreter is adapted to mediate between a client application and one or more analysis engines to produce interpretations of the natural language input that are valid for the client application. The first set of types is adapted to define interactions between the interpreter and the one or more analysis engines. The second set of types is adapted to define interactions between the interpreter and the client application.

Claims

1. A framework for generating a semantic interpretation of a natural language input comprising:

an interpreter adapted to mediate between a client application and one or more analysis engines to produce interpretations of the natural language input that are valid for the client application;

a first set of types adapted to define interactions between the interpreter and the one or more analysis engines; and

a second set of types adapted to define interactions between the interpreter and the client application.

2. The framework of claim 1 further comprising:

declarative schema describing a semantic model of natural language features of the client application;

wherein the interpreter is initialized with the declarative schema.

3. The framework of claim 2 wherein the semantic model comprises:

a description of relationships among a plurality of semantic modeling types derived from semantic types defined in a linguistic object model.

4. The framework of the claim 1 further comprising:

a linguistic object model comprising a set of modeling types adapted to model semantic elements of a natural language, the set of modeling types being independent of any particular natural language.

5. The framework of claim 4 wherein the modeling types comprise:

procedural rules adapted to place context-based constraints on validity on an instance of one or more of the plurality of semantic modeling types when called by the interpreter.

6. The framework of claim 5 wherein the context-based constraint causes the instance to be discarded if invalid.

7. The framework of claim 4 wherein the modeling types further comprise:

restriction clauses adapted to describe a relationship between types of the second set of types;

wherein the second set of types comprises semantic types some of which are created by a developer.

8. The framework of claim 6 wherein the restriction clauses are adapted to allow for procedural rules to determine acceptance of the relationship at runtime.

9. The framework of claim 1 further comprising:

a programming language comprising a set of programming constructs adapted to target the interpreter using types corresponding to the second set of types.

10. A framework for instantiating an interpretation of a natural language input to natural language features of a client application, the framework comprising:

an interpreter initialized with declarative schema associated with the client application, the interpreter adapted to receive a natural language input and to communicate with one or more natural language analysis engines to produce interpretations of the natural language input, the interpreter adapted to instantiate one or more of the interpretations in the client application.

11. The framework of claim 10 further comprising:

a set of types defining interactions between the client application, the interpreter and the one or more natural language analysis engines.

12. The framework of claim 10 wherein the instantiated one of the interpretations corresponds to at least one of the natural language features in the client application.

13. The framework of claim 10 wherein the declarative schema describes a semantic model of natural language features of the client application.

14. The framework of claim 13 wherein the semantic model comprises:

a plurality of semantic modeling types at least some of which are derived from semantic types defined in a linguistic object model; and

a description of relationships among the plurality of semantic modeling types.

15. The framework of claim 10 further comprising:

a linguistic object model comprising a set of modeling types adapted to model semantic elements of a natural language, the set of modeling types being independent of any particular natural language.

16. The framework of claim 15 wherein the modeling types comprise:

procedural rules adapted to place context-based constraints on the instantiation of the plurality of semantic modeling types when called by the interpreter.

17. The framework of claim 10 further comprising:

a programming language comprising a set of programming constructs adapted to target the interpreter using types corresponding to the set of types.

18. A natural language processing framework comprising:

an interpreter adapted to mediate between a client application and one or more analysis engines to produce interpretations of a natural language input, the interpreter adapted to instantiate interpretations that correspond to natural language features of the client application; and

a set of types adapted to define interactions between the interpreter and the one or more analysis engines.

19. The natural language processing framework of claim 18 wherein the interpretations comprise:

linguistic objects representing elements of the natural language input that are mapped to semantic types by the one or more analysis engines.

20. The natural language processing framework of claim 18 wherein the interpreter is initialized with declarative schema of the client application.

21. The natural language processing framework of claim 20 wherein the semantic model comprises:

a description of relationships among a plurality of semantic modeling types derived from semantic types defined in a linguistic object model.

22. The natural language processing framework of claim 18 further comprising:

a linguistic object model comprising a set of modeling types adapted to model semantic elements of a natural language, the set of modeling types being independent of any particular natural language.

23. The natural language processing framework of claim 18 wherein the interpreter and the set of types are stored on a computer readable medium.