PERFORMING A COLLABORATIVE SEARCH IN A COMPUTING NETWORK

• US 20100005089A1
• Filed: 11/18/2008
• Published: 01/07/2010
• Est. Priority Date: 07/03/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Yacobi, Yacov, Kajiya, Jim
• Granted Patent: US 8,015,177 B2
• Time in Patent Office: Days
• US Class Current: 707/5
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 10/06398 Performance of employee with respect to a job function
• CPC Class Codes: G06Q 10/10 Office automation; Time management

Abstract

Method for performing a collaborative search in a computing network. In one implementation, the method may include receiving a search criteria from a user, identifying one or more agents who have performed a search using the search criteria, the agents and the user belonging to the computing network, identifying one or more search results that the agents have previously selected as being relevant to the search criteria, ranking the search results, and displaying the search results according to the ranking.

Claims

1. A method for performing a collaborative search in a computing network, comprising:

receiving a search criteria from a user;

identifying one or more agents who have performed a search using the search criteria, the agents and the user belonging to the computing network;

identifying one or more search results that the agents have previously selected as being relevant to the search criteria;

ranking the search results; and

displaying the search results according to the ranking.

2. The method of claim 1, wherein ranking the search results comprises:

calculating one or more confidence values for the one or more agents;

calculating one or more weight values of the one or more agents; and

determining the ranking of the one or more search results based on the confidence values and the weight values.

3. The method of claim 2, wherein calculating the confidence values comprises:

assigning a low discrepancy value to each agent; and

subtracting the low discrepancy value from one.

4. The method of claim 3, wherein the discrepancy value is a derivative of conditional entropies.

5. The method of claim 2, wherein calculating the confidence values comprises:

calculating one or more probabilities that the one or more agents and the user have identical search results;

assigning a low discrepancy value to each agent if the probabilities are high;

assigning a high discrepancy value to each agent if the probabilities are low; and

subtracting the low discrepancy value from one.

6. The method of claim 2, wherein calculating the weight values comprises receiving a weight value for each agent from the user.

7. The method of claim 2, wherein calculating the weight values comprises:

receiving one or more search results of the user;

comparing the search results of the user with the search results of the agents; and

assigning higher weight values to one or more agents whose search results are substantially similar to the search results of the user.

8. The method of claim 2, wherein determining the ranking comprises computing a weighted average of the confidence values.

9. The method of claim 8, wherein the weighted average is based on the weight values.

10. The method of claim 1 further comprising:

receiving a selection of one of the search results from the user; and

reranking the search results in view of the selection.

11. A method for performing a collaborative search in a computing network, comprising:

receiving a first search criteria from a user;

identifying one or more agents who have performed a search using the first search criteria, the agents and the user belonging to the computing network;

identifying a first set of search results that the agents have previously selected as being relevant to the first search criteria;

ranking the first set of search results;

displaying the first set of search results according to the ranking;

receiving a second search criteria from the user;

identifying a portion of the agents who have performed a search using the second search criteria;

identifying a second set of search results that the portion of the agents have previously selected as being relevant to the second search criteria; and

ranking the second set of search results.

12. The method of claim 11, wherein ranking the first search results comprises:

calculating one or more confidence values for the one or more agents based on the first search criteria;

calculating one or more weight values of the one or more agents based on the first search criteria; and

determining a ranking of the first set of search results based on the confidence values and the weight values.

13. The method of claim 12, wherein ranking the second set of search results comprises:

recalculating the confidence values for the portion of the agents based on the second search criteria;

recalculating the weight values of the portion of the agents based on the second search criteria; and

determining a ranking of the second set of search results based on the confidence values and the weight values.

14. The method of claim 11, wherein the second search criteria comprises a history of internet links accessed by the user.

15. A computer system, comprising:

a processor; and

a memory comprising program instructions executable by the processor to;

receive a search criteria from a user;

identify one or more agents who have performed a search using the search criteria, the agents and the user belonging to the computing network;

identify one or more search results that the agents have previously selected as being relevant to the search criteria;

calculate one or more confidence values for the one or more agents;

calculate one or more weight values of the one or more agents; and

determine a ranking of the search results.

16. The computer system of claim 15, wherein the memory further comprises program instructions executable by the processor to display the search results according to the ranking.

17. The computer system of claim 15, wherein the ranking of the search results is based on the confidence values and weight values of the agents.

18. The computer system of claim 15, wherein the program instructions executable the processor to calculate the confidence values comprise program instructions executable by the processor to:

assign a low discrepancy value to each agent; and

subtract the low discrepancy value from one

19. The computer system of claim 15, wherein the discrepancy value is a derivative of conditional entropies.

20. The computer system of claim 15, wherein the ranking is determined by computing a weighted average of the confidence values, wherein the weighted average is based on the weight values.