Adaptive task framework

• US 20070106496A1
• Filed: 11/09/2005
• Published: 05/10/2007
• Est. Priority Date: 11/09/2005
• Status: Abandoned Application

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Katariya, Sanjeev, Ramsey, William, Barklund, Jonas
• Application Number:
  US11/270,407
• Publication Number:
  US 20070106496A1
• Time in Patent Office: Days
• US Class Current: 704/9
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates
• CPC Class Codes: G06F 40/295 Named entity recognition
• CPC Class Codes: G10L 15/18 using natural language modelling

Abstract

The subject disclosure pertains to systems and methods for performing natural language processing in which natural language input is mapped to a task. The system includes a task interface for defining a task, the associated data and the manner in which the task data is interpreted. Furthermore, the system provides a framework that manages the tasks to facilitate natural language processing. The task interface and framework can be used to provide natural language processing capabilities to third party applications. Additionally, the task framework can learn or be trained based upon feedback received from the third party applications.

Claims

1. A natural language processing framework, comprising:

a task component that defines one or more tasks;

a task retrieval component to process the tasks;

a slot-filling component to analyze data associated with the task; and

at least one application to execute the task.

2. The framework of claim 1, further comprising an interface component for interacting with a natural language processor.

3. The framework of claim 2, further comprising a component to process at least one query from an application.

4. The framework of claim 2, further comprising a logging component to enable adaptive changes within the natural language processor.

5. The framework of claim 4, further comprising a feedback component that is monitored by the logging component to determine the adaptive changes.

6. The framework of claim 5, further comprising at least one learning component that is trained from the feedback component.

7. The framework of claim 1, the task retrieval component employs a query to select one or more tasks from a collection of tasks.

8. The framework of claim 7, the task retrieval component automatically determines a task to be retrieved based upon keywords in the query.

9. The framework of claim 7, further comprising a component to index tasks based at least in part upon the keywords or other metadata.

10. The framework of claim 7, further comprising a component to pass user context information for automated selection of a desired task.

11. The framework of claim 1, the slot-filling component provides a matching of a list of tokens from a natural language input or query with one or more task parameters.

12. The framework of claim 11, the slot-filling component generates one or more possible mappings of tokens to one or more slots of a task.

13. The framework of claim 12, the slot-filling component is trained from feedback data.

14. The framework of claim 13, the slot-filling component generates a score or rank for a possible mapping of tokens to one or more task slots.

15. The framework of claim 14, further comprising an annotation component that includes one or more annotations that mark or indicate the significance of other tokens.

16. The framework of claim 15, the slot-filling component produces a list of up to a maximum number of requested semantic solutions, where a semantic solution is a representation of a mapping of tokens to slots that is employed by applications.

17. The framework of claim 1, further comprising a computer readable medium having computer readable instructions stored thereon for executing the task component, the task retrieval component, or the slot-filling component.

18. A natural language processing method, comprising:

defining one or more tasks for a natural language application;

automatically filling the tasks with data relevant to the application; and

automatically mapping the tasks to one token or query from the natural language application.

19. The method of claim 18, further comprising logging user feedback associated with the task.

20. A natural language processing system, comprising:

means for processing one or more tasks for a natural language application;

means for filling the tasks with one or more parameters of an application;

means for mapping the tasks to the application; and

means for interfacing to the task or the application.