Federated implicit search

• US 8,359,311 B2
• Filed: 06/01/2010
• Issued: 01/22/2013
• Est. Priority Date: 06/01/2010
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Wang, Kuansan, Kiciman, Emre M., Hsu, Bo-June, Wang, Chun-Kai
• Primary Examiner(s): NGUYEN, CAM LINH T
• Application Number:
  US12/791,000
• Publication Number:
  US 20110295852A1
• Time in Patent Office: 966 Days
• Field of Search: 707/707, 707/709, 707/722, 707/723, 707/726, 707/728, 707/730, 707/732, 707/736, 707/737, 707/748, 707/758, 707/767
• US Class Current: 707/728
• CPC Class Codes: G06F 16/9038 Presentation of query results
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 10/00 Administration; Management

Abstract

A resource selection system is described for assisting a user in performing a task that includes multiple actions. At each stage of the task, the system presents a set resources from which the user may select to perform a subsequent action in the task. The system implicitly selects the set of resources based on context information that identifies the user'"'"'s current informational needs. For example, the context information may be derived from textual information that is being presented on a user device, which the user is presumed to be viewing at the current time. In one implementation, the system selects the set of resources by computing language models for respective domains and respective entities. The system uses the language models to determine the relevance of the context information to each of the domains. The system then selects resources associated with domains that have been assessed as relevant.

Claims

1. A method, performed using at least one computing device, for identifying a set of resources associated with respective domains, comprising:

providing context information that pertains to interaction, by a user, with a user device, the context information including textual information associated with content presented to the user by the user device;

providing, for each of a plurality of individual domains, individual-domain score information that indicates relevance of the context information to a corresponding individual domain, plural instances of the individual-domain score information comprising plural-domain score information, the providing including

(i) receiving data from an entity associated with corresponding individual domain, the data characterizing at least goods or services of the entity, and

(ii) generating a language model that identities characteristics of the entity, based on the received data; and

identifying a set of resources based on the plural-domain score information for presentation to the user, the set of resources assisting the user in performing an action within a task having plural parts.

2. The method of claim 1, wherein the textual information comprises a document that is presented to the user.

3. The method of claim 1, wherein said context information also includes supplemental information that is obtained from a source other than the content that is presented to the user.

4. The method of claim 1, wherein the individual-domain score information for the corresponding individual domain pertains to a probability of occurrence of the context information within the corresponding individual domain, divided by a probability of occurrence of the context information outside of the corresponding individual domain.

5. The method of claim 1, wherein said providing of individual-domain score information comprises:

generating the individual-domain score information by evaluating the context information with respect to the language model.

6. The method of claim 5, wherein said generating of the language model comprises using an n-gram technique to generate the language model.

7. The method of claim 6, wherein, in the n-gram technique, the language model approximates a probability of an occurrence (P) of a string (s) occurring within the textual information, based on the following:

8. The method of claim 5, wherein said generating of the language model comprises generating plural language models for the corresponding individual domain.

9. The method of claim 1, wherein said identifying of the set of resources comprises ranking resources based on the plural-domain score information.

10. The method of claim 9, wherein said ranking is also based on at least one factor other than the plural-domain score information, said at least one factor operating to modify the plural-domain score information.

11. The method of claim 1, further comprising repeating said identifying of the set of resources as the user progresses through the task, wherein, at each part of the task, the method presents the user with a current set of resources associated with current context information, the current set of resources assisting the user in completing a next part of the task.

12. The method of claim 1, wherein the set of resources comprises application modules that are executable by the user device.

13. A computer readable medium for storing computer readable instructions, the computer readable instructions providing a presentation module when executed by one or more processing devices, the computer readable instructions comprising:

logic configured to provide a current presentation to a user via a user device;

logic configured to provide a set of resources selected based on current context information associated with the current presentation, the set of resources associated with plural respective domains;

logic configured to receive a selection by a user of a resource from among the set of resources, to identify a selected resource;

logic configured to invoke an action provided by the selected resource, causing presentation of a new current presentation associated with new current context information; and

logic configured to provide a preview of the action provided by the selected resource, the action involving use of a text portion of the current context information,

wherein the logic configured to provide a set of resources is configured to;

make model-generating functionality and score-generating functionality available to an entity associated with the corresponding individual domain, together with the context information; and

receive a individual-domain score information from the entity, based on generation, by the entity, of the individual-domain score information using the model-generating functionally and the score-generating functionality.

14. The computer readable medium of claim 13, wherein said logic configured to provide a set of resources is operative to display the set of resources as a set of selectable icons.

15. The computer readable medium of claim 13, wherein said logic configured to provide a set of resources is operative to provide at least one resource that is independent of the context information.

16. A computer readable medium for storing computer readable instructions, the computer readable instructions providing a presentation module when executed by one or more processing devices, the computer readable instructions comprising:

logic configured to provide, for each of a plurality of domains, individual-domain score information that indicates relevance of context information to a corresponding individual domain, plural instances of the individual-domain score information comprising plural-domain score information, each instance of the individual-domain score information being generated based on a language model for the corresponding individual domain, the context information being extracted from displayable content before the displayable content is displayed to a user; and

logic configured to select a set of application modules based on the plural-domain score information for presentation to the user, the set of application modules facilitating completion, by the user, of an action within a task, and

wherein the logic configured to provide individual-domain score information is configured to;

make model-generating functionality available to an entity associated with the corresponding individual domain; and

receive a language model from the entity that is associate with the corresponding individual domain.

17. The resource selection system of claim 16, wherein at least a subset of the set of application modules correspond to application modules that are currently installed or referenced by the user device.