ENTITY-SPECIFIC SEARCH MODEL

• US 20080005076A1
• Filed: 06/28/2006
• Published: 01/03/2008
• Est. Priority Date: 06/28/2006
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Dumais, Susan T., Naam, Ramez, Ozzie, Raymond E., Flake, Gary W., Platt, John C., Huang, Xuedong D., Goodman, Joshua T., Brill, Eric D., Meek, Christopher A., Hurst-Hiller, Oliver, Horvitz, Eric J., Gounares, Alexander G., Payne, Christopher D., Peltonen, Kyle G., Gates, William H.
• Granted Patent: US 7,822,762 B2
• Time in Patent Office: Days
• US Class Current: 707/3
• CPC Class Codes: G06F 16/903 Querying for retrieval from the web G06F16/953
• CPC Class Codes: G06F 16/9032 Query formulation

Abstract

A system that employs an explicitly and/or implicitly trained model in order to return entity-specific computer-based search results is provided. The innovation can provide for a customized search model that focuses search in connection with achieving information that is meaningful with respect to goals of an entity. The model can be used to modify a search query in accordance with a goal of the entity or to generate the search query thereby returning meaningful and/or targeted results to the user. The system can automatically gather entity-related data thereafter determining or inferring a goal as well as training the model. Moreover, the system can selectively configure (e.g., order, rank, filter) and render results to a user based upon the model.

Claims

1. A computer-implemented system that customizes computer-based search to an entity, comprising:

a model that is trained with entity-related data that corresponds to at least one of a plurality of users and a plurality of computers; and

an entity-specific search component that employs the model to modify a search query and provide entity-centric search results as a function of the trained model.

2. The system of claim 1, the entity-related data is demographic data that corresponds to at least one of the users.

3. The computer-implemented system of claim 1, further comprising a search engine that employs the search query to obtain the plurality of search results.

4. The computer-implemented system of claim 1, the model is trained with data mined from a subset of at least one of the plurality of users and the plurality of computers.

5. The computer-implemented system of claim 1, the entity-specific search component further comprises a search model generator that establishes the model based at least in part upon the entity-related data.

6. The computer-implemented system of claim 5, further comprising a goal determination component that gathers the entity-related data based at least in part upon a goal of the entity and inputs the gathered entity-related data into the search model generator.

7. The computer-implemented system of claim 6, the goal determination component comprises at least one of:

an inquiry component that queries a user to identify the goal; and

an inference component that infers the goal based at least in part upon the entity-related data.

8. The computer-implemented system of claim 7, the entity-related data includes at least one of a role of the user within the entity, an entity structure, a search location, entity product information and CRM data.

9. The computer-implemented system of claim 1, further comprising a rendering component that uses the trained model to re-rank search results prior to display to a user.

10. The computer-implemented system of claim 9, the rendering component comprises a results configuration component that organizes a subset of the search results.

11. The computer-implemented system of claim 10, the rendering component performs a utility-based analysis in connection with displaying a subset of the search results to the user, wherein the utility-based analysis factors a probability of cost of displaying a result to the user versus a probability of benefit of displaying the result to the user.

12. The computer-implemented system of claim 1, the entity-specific search component modifies the search query as a function of inferred user goal and state.

13. A computer-implemented method of performing an entity-based computer search, comprising:

receiving a search query from an entity, the entity comprises at least one of a plurality of users and a plurality of computers; and

prior to submission of the search query, modifying the search query as a function of a model that represents preferences of the entity.

14. The method of claim 13, further comprising:

establishing an entity context; and

modifying the search query based in part upon the entity context.

15. The method of claim 13, further comprising:

explicitly or implicitly training the model based upon entity feedback.

16. The method of claim 15, further comprising automatically gathering entity-related data based upon an objective of an entity.

17. The method of claim 16, further comprising automatically gathering the entity-related data from a plurality of disparate sources.

18. The method of claim 15, further comprising:

configuring a subset of the plurality of search results; and

rendering the subset of the plurality of search results.

19. A computer-implemented search system that comprises the following computer executable components:

means for determining an objective of an entity, the entity is associated with a plurality of individuals and a plurality of devices;

means for training a model based at least in part upon the objective;

means for modifying a search query based upon the model;

means for obtaining a plurality of search results based upon the modified search query; and

means for rendering a subset of the plurality of search results to a user.

20. The computer-implemented system of claim 19, further comprising means for training the model based at least in part upon a role of at least one of the plurality of individuals associated with the entity.