Using related users data to enhance web search

• US 8,244,721 B2
• Filed: 02/13/2008
• Issued: 08/14/2012
• Est. Priority Date: 02/13/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Morris, Meredith June, Teevan, Jaime Brooks, Mickens, James W., Amershi, Saleema
• Primary Examiner(s): ALAM, SHAHID AL
• Application Number:
  US12/030,331
• Publication Number:
  US 20090204599A1
• Time in Patent Office: 1,644 Days
• Field of Search: 707/732, 707/721, 707/722, 707/723, 707/802, 715/753
• US Class Current: 707/723
• CPC Class Codes: G06F 16/9032 Query formulation
• CPC Class Codes: G06F 16/90335 Query processing
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

The claimed subject matter provides a system and/or a method that facilitates generating a personalized query result for a specific user. An interface can receive at least one of a portion of a text query to be searched or a portion of personalized content related to a user that submits the portion of the text query. A personalization component can combine the portion of personalized content related to the user with a portion of personalized content related to one or more disparate users to create group personalized content, wherein the group personalized content is compared with the portion of the text query to identify a relationship there between to generate a personalized query result in accordance with the relationship.

Claims

1. A system for generating a personalized query result for a specific user, the system comprising:

memory;

a processor;

an interface stored in the memory that receives at least one of a portion of a text query to be searched or a portion of personalized content related to a user that submits the portion of the text query;

a personalization component stored in the memory and executed by the processor that compares the portion of personalized content related to the user with a portion of personalized content related to one or more disparate users to create a group of users having group personalized content, the group personalized content is compared with the portion of the text query to identify a relationship there between to generate the personalized query result based in part on the relationship; and

a shared web task engine that provides a collaborative web search utilizing a smart splitting technique, the smart splitting technique receives the portion of the text query and distributes query results based on the text query to members of the group to enable parallel evaluation by the members of the group, the personalized query result being generated further based in part on the parallel evaluation.

2. The system of claim 1, the personalized content relates to at least one of a background, an interest, a portion of browsed content, a portion of Internet history, a portion of a user profile, a visited website, a questionnaire, a purchase, a shared task as established by joint participation in a search using a specialized collaborative web searching tool, a portion of data related to a social networking website, a specialized group search interface, a portion of desktop content related to a machine, or a shopping history.

3. The system of claim 1, the personalized content is user specific and is at least one of local data, remote data, data on a device, or data on a computing machine.

4. The system of claim 1, further comprising a search component that generates the query results based on the text query, the query results are filtered or ranked based on the relationship of the user and the group.

5. The system of claim 1, the personalization component employs a groupized ranking by at least one of the following:

calculating a personalization score for each member of the group and calculating a group score for each query result by combining the personalization score from each member, and re-ranking the query results based on the group score;

computing the group personalization score from a combination of the group member'"'"'s profiles and re-ranking the query results based on the group score;

or

utilizing an algorithm to combine personalization scores for members of the group into the group score.

6. The system of claim 5, the personalization score for each member of the group is based upon relevancy to the individual member and respective personalization content.

7. The system of claim 1, the query results are distributed to each member as non-overlapping subsets in which the distribution is based on a personalized score for each member.

8. The system of claim 1, the shared web task engine implements a group query expansion that expands the portion of the text query based on evaluating a text query from the group, the expansion of the text query is dependent upon having a relationship there between in which the relationship includes at least one of a semantic content, a time span of entry, or a manual inference.

9. The system of claim 8, the text query from the group is identified from a task session manually defined in at least one of the shared web task engine or the collaborative web search.

10. The system of claim 8, further comprising a highlighter component that employs an emphasis technique to a portion of the query result generated from the group query expansion search.

11. The system of claim 10, the emphasis technique is at least one of a text highlight, a magnification, a font, a bold, an underline, a pop-up, or a portion of a graphic.

12. The system of claim 10, the highlighter component utilizes a first emphasis technique for a first user and a second emphasis technique for a second user, the first user and the second user are members of the group.

13. The system of claim 1, further comprising a scope component that utilizes a confidence ranking for at least one member within the group, the confidence ranking defines a measurement of similarity between a first member and a second member within the group.

14. The system of claim 1, further comprising an edit component that enables definition of a security setting by the user for a first portion of personalized content, the security setting to allow or disallow exposure of the first portion of personalized content to be utilized to:

identify a group; and

/or

identify a personalized query result.

15. The system of claim 14, the edit component allows the user to manually edit the first or a second portion of personalized content with a level of granularity to expose or allow at least one of the first or the second portion of personalized content to a first group or a first user and disallow at least one of the first or the second portion of personalized content to a second group or a second user.

16. The system of claim 1, the group being created based upon an automatic evaluation of personalized content of the user and the one or more disparate users.

17. A computer-implemented method for enhancing identification of relevant search results, the method comprising:

receiving a search query from a user, the user having personalized content that includes data specific to the user;

receiving a selection specifying a subset of the data specific to the user, the subset being specified to define which of the data specific to the user will be considered for group assignment of the user and other users to a group;

evaluating the subset of the data specific to the user to identify at least one additional user with related data there between;

creating the group including the user and the at least one additional user based in part on the evaluating and the specifying of the subset;

identifying a correlation between the user, the at least one additional user, and the search query;

generating one or more search results for the search query;

distributing the one or more search results to the at least one additional user to enable parallel evaluation by the at least one additional user; and

filtering the one or more search results based at least upon the correlation and the parallel evaluation.

18. The method of claim 17, further comprising:

aggregating the at least one additional user into the group;

providing a personalized score for each of the one or more search results for each member in the group; and

utilizing each personalized score to re-rank the one or more search results, the utilizing comprising;

calculating a group score as a sum of each group member'"'"'s personalization score for each of the one or more search results; and

re-ranking the one or more search results based on each associated group score.

19. The method of claim 17, further comprising:

implementing a shared web search for the search query, wherein the shared web search comprises the user participating in a collaborative web search with the at least one additional user;

identifying relevant results for a particular user based on the shared web search; and

utilizing a group hit highlighting on the relevant results for the search query, the group hit highlighting comprising a unique emphasis technique for each user of the group.

20. A computer-implemented method for generating a query result relevant to a user, the method comprising:

receiving of a portion of a text query to be searched and a portion of personalized content items related to a user that submits the portion of the text query, the portion of the personalized content items being specified by the user to define which of the personalized content items related to the user will be considered for group assignment of the user and one or more other users to a group;

evaluating personalized content items related to at least one additional user and the user that submits the text query based in part on a relationship between the portion of personalized content items and the personalized content items related to the at least one additional user;

creating the group including the user that submits the text query and the at least one additional user based in part on the evaluating;

leveraging the group and the portion of the personalized content items to provide a search result associated with the portion of the text query by performing at least one of the following;

determining a groupized ranking by calculating a personalized ranking of the search result for each member of the group and combining the personalized ranking to provide a group score;

smart splitting by distributing the search results to the group for parallel evaluation;

or

group query expansion that broadens the text query by evaluating related queries generated by a member of the group.