Providing a task description name space map for the information worker

• US 8,700,385 B2
• Filed: 04/04/2008
• Issued: 04/15/2014
• Est. Priority Date: 04/04/2008
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Dumais, Susan T., Gates, William H. III, Shoroff, Srikanth, Ehrenberg, Michael, Harris, Jensen M., Wolf, Richard J., Goodman, Joshua T., Megiddo, Eran
• Primary Examiner(s): SPOONER, LAMONT M
• Application Number:
  US12/098,189
• Publication Number:
  US 20090254336A1
• Time in Patent Office: 2,202 Days
• Field of Search: 704/1, 704/9, 704/10, 704/257, 704/270.1, 707706-708, 715/811, 717/108, 717/112
• US Class Current: 704/9
• CPC Class Codes: G06F 9/451 Execution arrangements for user interfaces

Abstract

Providing for generation of a task oriented data structure that can correlate natural language descriptions of computer related tasks to application level commands and functions is described herein. By way of example, a system can include an activity translation component that can receive a natural language description of an application level task. Furthermore, the system can include a language modeling component that can generate the data structure based on an association between the description of the task and at least one application level command utilized in executing the computer related task. Once generated, the data structure can be utilized to automate computer related tasks by input of a human centric description of those tasks. According to further embodiments, machine learning can be employed to train classifiers and heuristic models to optimize task/description relationships and/or tailor such relationships to the needs of particular users.

Claims

1. A system that provides a task oriented data structure to correlate natural language task descriptions to at least one of a plurality of application level commands available for inclusion in the task oriented data structure, comprising:

an activity tracking component that monitors performance of a first application level task and registers at least one first application level command utilized in accomplishing the first application level task, the at least one first application level command utilized in accomplishing the first application level task being one of the plurality of application level commands;

an activity translation component that obtains a first natural language task description for the first application level task;

a language modeling component that generates a first task oriented data structure based on the first natural language task description and the at least one first application level command utilized in accomplishing the first application level task, the first task oriented data structure facilitates execution of the at least one first application level command such that the first task oriented data structure specifies the at least one first application level command to be executed and causes the execution, at least in part, of the at least one first application level command; and

a namespace execution component that receives a second natural language task description for a second application level task and causes the execution of the second application level task based at least in part on the second natural language task description and the first task oriented data structure, the first natural language task description and the second natural language task description differing at least in part such that the at least one first application level command utilized in accomplishing the first application level task differs at least in part from at least one second application level command utilized in accomplishing the second application level task.

2. The system of claim 1, the first natural language task description includes a description of the first application level task, instructions pertaining to accomplishing the first application level task, or user profile information qualifying or contextualizing the first application level task, or a combination thereof.

3. The system of claim 1, comprising a template compilation component that generates a namespace template for the first application level task, the template incorporates a plurality of natural language descriptions of the first application level task, and associates the descriptions with a plurality of application level commands utilized to accomplish the first application level task.

4. The system of claim 3, further comprising at least one of:

a task extrapolation component that tracks an application utilized to accomplish the first application level task and identifies an additional application, command, or function, or a combination thereof, pertinent to accomplishing the first application level task or a related task and includes the additional application, command or function into the namespace template;

or

a namespace extrapolation component that references a second data structure and incorporates commands or task descriptions from the second data structure into the namespace template.

5. The system of claim 3, the association between the first natural language task description and the at least one first application level command is based at least in part on an association incorporated within the namespace template.

6. The system of claim 1, the activity translation component further comprises at least one of:

a query component that submits a request for a first natural language command description related to utilization of the first application level command, why the first application level command is being performed, what portion of the first application level task the first application level command is related to, when the first application level command or the first application level task is interrupted, when completion of an ancillary task is utilized to accomplish the first application level task, when the first application level task is initiated, or when the first application level task is complete, or a combination thereof;

or

an input component that receives a response to the request, the response forms at least a portion of the first natural language task description.

7. The system of claim 1, comprising a user profile component that stores information related to preferences for searching data, extrapolating the task to related tasks, borrowing commands or tasks from a namespace template, or deconstructing problem solving techniques related to a task, or a combination thereof.

8. The system of claim 1, comprising a machine learning component that utilizes heuristics or trained classifiers to assist the activity translation component and language map component in associating descriptions and commands and generating the first task oriented data structure.

9. The system of claim 1, comprising a task association component that infers an ancillary task and recommends initiation of the ancillary task in conjunction with execution of the at least one first application level command.

10. The system of claim 1, the first application level task includes, spreadsheet management, data entry, document creation and formatting, content authoring, database management, computer-aided shopping, online research, securing travel reservations, identifying entertainment opportunities, identifying sales or business opportunities, researching sales information, balancing a budget, locating a person or group of people, locating a physical or digital resource, or a combination thereof.

11. The system of claim 9, the ancillary task includes locating product reviews, obtaining product price information, deconstructing problem solving techniques related to accomplishing a task, identifying problem solving techniques of colleagues, fellow employees, or experts, or identifying a namespace template associated with a task, or a combination thereof.

12. The system of claim 9, the language modeling component provides a task-focused context for the first natural language task description based on the association with the at least one first application level command.

13. A method for monitoring an information worker for facilitating a natural language task-oriented data structure, comprising:

monitoring a device-related activity connected with execution of an application level task and registering at least one application level command utilized in the monitored execution of the application level task, the at least one application level command utilized in the monitored execution of the application level task being one of a plurality of application level commands available for inclusion in the task oriented data structure;

obtaining a natural language description of the activity to assist in defining a relationship between the application level task and the activity; and

analyzing at least one of the device-related activity, the obtained natural language description, or the application level task for constructing a natural language task oriented data structure that associates the natural language description with the activity or with the task to facilitate execution of the at least one application level command, wherein the task oriented data structure specifies the at least one application level command to be executed and facilitates execution of the at least one application level command by causing the execution, at least in part, of the at least one application level command; and

receiving a different natural language task description for a different application level task and causing the execution of the different application level task based at least in part on the different natural language task description and the task oriented data structure, the natural language task description and the different natural language task description differing at least in part such that the at least one application level command utilized in accomplishing the application level task differs at least in part from at least one application level command utilized in accomplishing the different application level task.

14. The method of claim 13, comprising at least one of the following:

providing a template for exemplifying a feature of the task;

implementing the feature based on a second natural language description related to the obtained natural language description;

or

developing a high-level task description language based at least in part on the data structure.

15. The method of claim 13, comprising at least one of:

identifying a sub-task related to execution of the application level task;

associating the sub-task with a second device-related activity that is pertinent to completion of the sub-task;

obtaining a second natural language description pertaining to the sub-task or the device-related activity, or both;

incorporating the sub-task, second device-related activity, or second natural language description into a namespace template associated with the task;

or

recommending initiation of the sub-task in conjunction with implementing the application level task.

16. A system that generates a high-level task description language for accomplishing device implemented tasks by way of natural language commands, comprising:

one or more processors;

one or more computer readable storage media storing instructions that when executed by the one or more processors perform a method comprising;

tracking an application level activity related to execution of an application level task and registering at least one first application level command utilized in the tracked execution of the application level task, the at least one first application level command utilized in the tracked execution of the application level task being one of a plurality of application level commands available for inclusion in the task oriented data structure;

obtaining a natural language description of the application level activity or the application level task;

generating a data structure that creates a relationship between the task and the natural language description to facilitate execution of the at least one first application level command, wherein the data structure specifies the at least one first application level command to be executed and facilitates execution of the at least one application level command by causing the execution, at least in part, of the at least one first application level command; and

receiving a different natural language description for a different application level task and causing the execution of the different application level task based at least in part on the different natural language description and the task oriented data structure, the natural language task description and the different natural language task description differing at least in part such that the at least one first application level command utilized in accomplishing the application level task differs at least in part from at least one second application level command utilized in accomplishing the different application level task.

17. The system of claim 16, the method further comprising compiling a task description language that aggregates the data structure with a second data structure, the second data structure provides a relationship between a second task and a second natural language description.

18. The system of claim 16, the method further comprising at least one of:

identifying a sub-task related to execution of the task;

associating the sub-task with an activity that is pertinent to completion of the sub-task;

receiving a natural language description of the sub-task or the activity, or both;

incorporating the sub-task, activity, and description into a namespace template associated with the task;

or

recommending initiation of the sub-task in conjunction with implementing the task.