NATURAL LANGUAGE INTERFACE FOR DRIVING ADAPTIVE SCENARIOS

• US 20070106497A1
• Filed: 03/28/2006
• Published: 05/10/2007
• Est. Priority Date: 11/09/2005
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Katariya, Sanjeev, Ramsey, William
• Granted Patent: US 7,627,466 B2
• Time in Patent Office: Days
• US Class Current: 704/9
• CPC Class Codes: G06F 40/295 Named entity recognition
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 40/35 Discourse or dialogue representation
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command

Abstract

A “Natural Language Script Interface” (NLSI), provides an interface and query system for automatically interpreting natural language inputs to select, execute, and/or otherwise present one or more scripts or other code to the user for further user interaction. In other words, the NLSI manages a pool of scripts or code, available from one or more local and/or remote sources, as a function of the user'"'"'s natural language inputs. The NLSI operates either as a standalone application, or as a component integrated into existing applications. Natural language inputs may be received from a variety of sources, and include, for example, computer-based text or voice input, handwriting or text recognition, or any other human or machine-readable input from one or more local or remote sources. In various embodiments, machine learning techniques are used to improve script selection and processing as a function of observed user interaction with selected scripts.

Claims

1. A method for identifying actions by semantic analysis of a natural language input, comprising:

receiving a natural language input from a user;

performing a semantic analysis of the natural language input to identify probable matches to one or more scripts as a function of an initial semantic mapping associated with each script;

probabilistically scoring and sorting the identified scripts to construct a rank ordered list of probable scripts; and

presenting the rank ordered list of probable scripts to the user via a user interface.

2. The method of claim 1 wherein performing the semantic analysis of the natural language input comprises:

deconstructing the natural language input to create a set of one or more tokens from the natural language input; and

probabilistically evaluating the tokens with respect to the initial semantic mapping associated with each script.

3. The method of claim 1 further comprising:

receiving context information associated with the natural language input; and

wherein the semantic analysis is performed with respect to the received context information.

4. The method of claim 1 wherein the scripts within the rank ordered list of probable scripts are user selectable via the user interface

5. The method of claim 4 wherein user selection of any of the scripts initiates execution of the user selected scripts.

6. The method of claim 1 wherein one or more of the rank ordered list of probable scripts are automatically executed.

7. The method of claim 4 wherein the user interface provides means for allowing user entry of additional data into Slots within user selected scripts for providing the scripts with information required for successful execution of the selected script.

8. The method of claim 1 further comprising extracting information from the natural language input as a part of the semantic analysis for populating applicable Slots within one or more of the scripts for providing those scripts with information required for successful execution of those scripts.

9. The method of claim 1 further comprising machine learning means for learning updated semantic analysis models for performing improved semantic analysis of future natural language inputs as a function of user interaction with the rank ordered list of probable scripts via the user interface.

10. A system for providing a list of probable tasks to a user in response to a user input, comprising using one or more computing devices to perform steps for:

constructing a set of initial natural language mapping criteria from a set of available tasks;

receiving a natural language input from a user;

performing a semantic analysis of the natural language input relative to the initial natural language mapping criteria to identify a set of one or more highest probability tasks;

extracting one or more values from the natural language input for use in populating one or more Slots of one or more of the set of highest probability tasks;

scoring the tasks comprising the set of highest probability tasks as a function of the semantic analysis and the population of the one or more Slots of those tasks;

sorting the scored tasks in order of highest to lowest scores to create a rank ordered list of probable tasks; and

presenting the rank ordered list of probable tasks to the user.

11. The system of claim 10 further comprising steps for receiving context information associated with the natural language input and including an evaluation of the context information in the semantic analysis of the natural language input.

12. The system of claim 10 wherein the scripts within the rank ordered list of probable tasks are user selectable via a user interface.

13. The system of claim 12 wherein user selection of any of the tasks initiates execution of the user selected tasks.

14. The system of claim 10 wherein one or more of the rank ordered list of probable tasks are automatically executed.

15. The system of claim 12 further comprising steps for user entry of additional data via the user interface into one or more Slots within one or more user selected tasks.

16. The system of claim 12 further comprising steps for:

observing user interactions with the rank ordered list of probable tasks via the user interface; and

using the observed user interactions for learning an updated semantic analysis model for use in performing the semantic analysis with respect to further natural language inputs.

17. A computer-readable medium having computer-executable instructions for presenting user selectable tasks to a user via a user interface, said computer executable instructions comprising means for:

receiving a natural language input from a user;

analyzing the natural language input relative to an initial mapping criteria extracted from one or more sets of available tasks to identify a set of one or more tasks having a highest probability of match to the natural language input;

populating one or more variables of one or more of the highest probability tasks with one or more data elements extracted from the natural language input;

computing a score for each of the highest probability tasks as a function of a probability of match and the populated variables associated with each of those tasks;

sorting the highest probability tasks as a function of the score associated with each of those tasks; and

presenting the sorted tasks to the user via a graphical user interface.

18. The computer-readable medium of claim 17 further comprising computer-executable instructions for:

observing user interactions with the sorted tasks via the user interface; and

using the observed user interactions for learning an updated semantic analysis model for use in analyzing further natural language inputs.

19. The computer-readable medium of claim 17 further comprising computer-executable instructions for initiating execution of the tasks selected via the user interface.

20. The computer-readable medium of claim 17 further comprising computer-executable instructions for facilitating user entry of data elements via the user interface for populating one or more variables of one or more of the sorted tasks.