Optimization-based data content determination

• US 7,809,720 B2
• Filed: 06/14/2007
• Issued: 10/05/2010
• Est. Priority Date: 10/20/2004
• Status: Expired due to Fees

• Current Assignee: LinkedIn Corporation (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: Aggarwal, Vikram, Zhou, Michelle X.
• Primary Examiner(s): Wassum; Luke S.
• Assistant Examiner(s): Reyes; Mariela D
• Application Number:
  US11/763,132
• Publication Number:
  US 20070239670A1
• Time in Patent Office: 1,209 Days
• Field of Search: 707/723, 707/724, 707/725, 707/946, 707/999.102, 707/999.104, 707/999.107
• US Class Current: 707/724
• CPC Class Codes: G06F 16/00 Information retrieval; Database structures therefor; File system structures therefor
• CPC Class Codes: Y10S 707/946 Real estate

Abstract

Improved data content determination techniques are disclosed for use in accordance with information-seeking systems. For example, in one illustrative aspect of the invention, a technique for determining data content for a response to a query comprises obtaining a user query, and dynamically determining data content suitable for generating a response to the query, wherein data content determination is modeled as an optimization operation which attempts to balance context-based selection constraints. Further, the step of dynamically determining data content may further comprise modeling the context-based selection constraints as feature-based metrics. The feature-based metrics may be formulated using contextual information. Still further, the step of dynamically determining data content may further comprise performing the optimization operation such that one or more desirability metrics are maximized and one or more cost metrics are minimized, thus balancing the various constraints.

Claims

1. A method of determining data content for a response to a query, comprising the steps of:

obtaining a user query;

selecting one or more of a plurality of modalities for outputting a response to the query;

dynamically determining data content suitable for generating a response to the query, wherein data content determination is modeled as an optimization operation which attempts to balance context-based selection constraints; and

outputting the response to the query using the selected one or more of the plurality of modalities;

wherein the data content determination further comprises modeling the context-based selection constraints as feature-based metrics, the feature-based metrics measuring a presentation desirability value and one or more cost values associated with the selected one or more of the plurality of modalities; and

wherein the selection comprises;

determining a cost value associated with a visual presentation of at least a portion of the response;

determining a cost value associated with a spoken presentation of the at least a portion of the response;

comparing the cost value associated with the visual presentation and the cost value associated with the spoken presentation; and

selecting at least one of the visual presentation and the spoken presentation based at least in part on the comparison of the cost value associated with the visual presentation and the cost value associated with the spoken presentation.

2. The method of claim 1, wherein the step of the feature-based metrics measuring a presentation desirability value further comprises the step of using one or more data content properties.

3. The method of claim 2, wherein the step of using one or more data content properties further comprises using at least one of data volume, data availability or data informativeness.

4. The method of claim 1, wherein the step of the feature-based metrics measuring a presentation desirability value further comprises the step of using one or more data relationships.

5. The method of claim 4, wherein the step of using one or more data relationships further comprises using a data dependency.

6. The method of claim 1, wherein the step of the feature-based metrics measuring a presentation desirability value further comprises the step of using user infomation.

7. The method of claim 6, wherein the step of using user information further comprises using user interests.

8. The method of claim 1, wherein the step of the feature-based metrics measuring a presentation desirability value further comprises the step of using environment information.

9. The method of claim 8, wherein the step of using environment information further comprises using data associated with at least one of device features and availability.

10. The method of claim 1, wherein the step of the feature-based metrics measuring a presentation desirability value further comprises the step of using a conversation history.

11. The method of claim 10, wherein the step of using a conversation history further comprises using a sequence of previous user queries and system responses.

12. The method of claim 1, wherein the cost value associated with the visual presentation comprises a space cost associated with outputting the at least a portion of the response to the query using the visual presentation, wherein the space cost is based at least in part on a number of pixels required to output the at least a portion of the response to the query using the visual presentation.

13. The method of claim 1, wherein the cost value associated with the spoken presentation comprises a time cost associated with outputting the at least a portion of the response to the query using the spoken presentation.

14. The method of claim 1, wherein the step of dynamically determining data content further comprises the step of formulating the feature-based metrics using contextual information.

15. The method of claim 14, wherein the step of the feature-based metrics using contextual information further comprises the contextual information comprising at least one of query information, a conversation history, and a user model.

16. The method of claim 1, wherein the step of dynamically determining data content further comprises the step of performing the optimization operation such that one or more desirability metrics are maximized and one or more cost metrics are minimized.

20. The method of claim 13, wherein the time cost is based at least in part on a number of words in the at least a portion of the response.

17. Apparatus for determining data content for a response to a query, comprising:

a memory; and

at least one processor coupled to the memory and operative to obtain a user query;

to select one or more of a plurality of modalities for outputting a response to the query;

to dynamically determine data content suitable for generating a response to the query, wherein data content determination is modeled as an optimization operation which attempts to balance context-based selection constraints; and

to output the response to the query using the selected one or more of the plurality of modalities;

wherein the data content determination further comprises modeling the context-based selection constraints as feature-based metrics, the feature-based metrics measuring a presentation desirability value and one or more cost values associated with the selected one or more of the plurality of modalities; and

wherein the selection comprises;

determining a cost value associated with a visual presentation of at least a portion of the response;

determining a cost value associated with a spoken presentation of the at least a portion of the response;

comparing the cost value associated with the visual presentation and the cost value associated with the spoken presentation; and

selecting at least one of the visual presentation and the spoken presentation based at least in part on the comparison of the cost value associated with the visual presentation and the cost value associated with the spoken presentation.

18. An article of manufacture for determining data content for a response to a query, comprising a machine readable medium containing one or more programs which when executed implement the steps of:

obtaining a user query;

selecting one or more of a plurality of modalities for outputting a response to the query;

dynamically determining data content suitable for generating a response to the query, wherein data content determination is modeled as an optimization operation which attempts to balance context-based selection constraints; and

outputting the response to the query using the selected one or more of the plurality of modalities;

wherein the data content determination further comprises modeling the context-based selection constraints as feature-based metrics, the feature-based metrics measuring a presentation desirability value and one or more cost values associated with the selected one or more of the plurality of modalities; and

wherein the selection comprises;

determining a cost value associated with a visual presentation of at least a portion of the response;

determining a cost value associated with a spoken presentation of the at least a portion of the response;

comparing the cost value associated with the visual presentation and the cost value associated with the spoken presentation; and

selecting at least one of the visual presentation and the spoken presentation based at least in part on the comparison of the cost value associated with the visual presentation and the cost value associated with the spoken presentation.

19. A method of providing a service for determining data content for a response to a query, comprising the step of:

a service provider, in response to an obtained query, enabling the steps of;

selecting one or more of a plurality of modalities for outputting a response to the query;

dynamically determining data content suitable for generating a response to the query, wherein data content determination is modeled as an optimization operation which attempts to balance context-based selection constraints; and

outputting the response to the query using the selected one or more of the plurality of modalities;

wherein the data content determination further comprises modeling the context-based selection constraints as feature-based metrics, the feature-based metrics measuring a presentation desirability value and one or more cost values associated with the selected one or more of the plurality of modalities; and

wherein the selection comprises;

determining a cost associated with a visual presentation of at least a portion of the response;

determining a cost associated with a spoken presentation of the at least a portion of the response;

comparing the cost associated with the visual presentation and the cost associated with the spoken presentation; and

selecting at least one of the visual presentation and the spoken presentation based at least in part on the comparison of the cost associated with the visual presentation and the cost associated with the spoken presentation.