Method and Apparatus for Dynamic Modification of Command Weights in a Natural Language Understanding System

• US 20070225969A1
• Filed: 06/11/2007
• Published: 09/27/2007
• Est. Priority Date: 09/03/2003
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Inventors: Coffman, Daniel, Ramaswamy, Ganesh, Kleindienst, Jan
• Granted Patent: US 7,533,025 B2
• Time in Patent Office: Days
• US Class Current: 704/9
• CPC Class Codes: G06F 40/44 Statistical methods, e.g. probability models
• CPC Class Codes: G06F 40/58 Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation
• CPC Class Codes: G10L 2015/228 of application context

Abstract

A method and system for dynamically assigning weights to the subset of commands in a natural language dialog system based on prior context of the user'"'"'s interaction with the system. The search space of the translation process may be reduced when some context information is available. A user presents input to the natural language understanding system. The system translates the user input into a formal command and calculates a weight value for a next set of formal commands based on the formal command. The command weights may then be dynamically boosted for the next set of formal commands before executing the formal command. The exemplary aspects of the present invention reduce the time needed to complete a task since the search space of the translation process may be reduced if context information is available and improve the accuracy of the process by using knowledge that users regularly use repeating patterns for repeating tasks.

Claims

1. A method for context-based dynamic assignment of weights for formal commands in a natural language understanding system, comprising:

receiving a user input;

translating the user input into a formal command;

determining a weight value for a next set of formal commands based on the formal command;

dynamically boosting the command weights for the next set of formal commands; and

executing the formal command.

2. The method of claim 1, further comprising:

receiving a second user input; and

translating the second user input into a formal command based on the weight value.

3. The method of claim 1, wherein calculating a weight value for the next set of formal commands includes one of a given command history, access method information, and application context.

4. The method of claim 1, wherein dynamically boosting the command weights for the set of formal commands includes ranking the formal commands based on their corresponding conditional probability.

5. The method of claim 1, wherein translating the user input into a formal command includes converting the user input into text and converting the text into a formal command.

6. The method of claim 1, further comprising:

resetting the boosted command weights to a default value.

7. The method of claim 6, wherein the step of resetting the boosted command weights includes automatically resetting the command weights to the default value after the next set of formal commands are executed.

8. The method of claim 6, wherein the step of resetting the boosted command weights includes issuing a reset command by the dialog manager.

11. The method of claim 1, wherein translating the user input into a formal command includes creating a subset of formal commands to narrow a search space for determining the formal command corresponding to the user input.

9. (canceled)

10. (canceled)

12. A data processing system for context-based dynamic assignment of weights for formal commands in a natural language understanding system, comprising:

means for receiving a user input;

means for translating the user input into a formal command;

means for determining a weight value for a next set of formal commands based on the formal command;

means for dynamically boosting the command weights for the next set of formal commands; and

means for executing the formal command.

13. The data processing system of claim 12, further comprising:

means for receiving a second user input; and

means for translating the second user input into a formal command based on the weight value.

14. The data processing system of claim 12, wherein the means for calculating a weight value for the next set of formal commands includes one of a given command history, access method information, and application context.

15. The data processing system of claim 12, wherein the means for dynamically boosting the command weights for the set of formal commands includes ranking the formal commands based on their corresponding conditional probability.

16. The data processing system of claim 12, wherein the means for translating the user input into a formal command includes converting the user input into text and converting the text into a formal command.

17. The data processing system of claim 12, further comprising:

means for resetting the boosted command weights to a default value.

18. The data processing system of claim 17, wherein the means for resetting the boosted command weights includes automatically resetting the command weights to the default value after the next set of formal commands are executed.

19. The data processing system of claim 17, wherein the means for resetting the boosted command weights includes issuing a reset command by the dialog manager.

22. The data processing system of claim 12, wherein the means for translating the user input into a formal command includes creating a subset of formal commands to narrow a search space for determining the formal command corresponding to the user input.

20. (canceled)

21. (canceled)

23. A system for context-based dynamic assignment of weights for formal commands in a natural language understanding system, the system comprising:

user interface for generating a user input;

a natural language understanding system, wherein the user input is translated into a formal command;

a command booster for calculating a weight value for a next set of formal commands based on the formal command and dynamically boosting the command weights for the next set of formal commands; and

a command executor for executing the formal command.

24. A computer program product comprising a computer readable medium having encoded thereon computer usable program code for context-based dynamic assignment of weights for formal commands in a natural language understanding system, the computer program product comprising:

computer usable program code for receiving a user input;

computer usable program code for translating the user input into a formal command;

computer usable program code for determining a weight value for a next set of formal commands based on the formal command;

computer usable program code for dynamically boosting the command weights for the next set of formal commands; and

computer usable program code for executing the formal command.

25. The computer program product of claim 24, further comprising:

instructions for receiving a second user input; and

instructions for translating the second user input into a formal command based on the weight value.

26. The computer program product of claim 24, wherein the instructions for calculating a weight value for the next set of formal commands includes one of a given command history, access method information, and application context.

27. The computer program product of claim 24, wherein the instructions for dynamically boosting the command weights for the set of formal commands includes ranking the formal commands based on their corresponding conditional probability.

28. The computer program product of claim 24, wherein the instructions for translating the user input into a formal command includes converting the user input into text and converting the text into a formal command.

29. The computer program product of claim 24, further comprising:

instructions for resetting the boosted command weights to a default value.

30. The computer program product of claim 29, wherein the instructions for resetting the boosted command weights includes automatically resetting the command weights to the default value after the next set of formal commands are executed.

31. The computer program product of claim 29, wherein the instructions for resetting the boosted command weights includes issuing a reset command by the dialog manager.

34. The computer program product of claim 24, wherein the instructions for translating the user input into a formal command includes creating a subset of formal commands to narrow a search space for determining the formal command corresponding to the user input.

32. (canceled)

33. (canceled)