Systems and methods for demoting personalized search results based on personal information

• US 7,827,170 B1
• Filed: 08/28/2007
• Issued: 11/02/2010
• Est. Priority Date: 03/13/2007
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Horling, Bryan C., Zamir, Oren E.
• Primary Examiner(s): LEROUX, ETIENNE PIERRE
• Application Number:
  US11/846,353
• Time in Patent Office: 1,162 Days
• Field of Search: 707/2, 707/5, 707/722, 709/238
• US Class Current: 707/722
• CPC Class Codes: G06F 16/24575 using context
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

An information server identifies a list of search results associated with a search query from a user. Each search result has an initial position in the list. The information server then identifies a user profile associated with the user. In some embodiments, the user profile includes a set of user-preferred search results that is determined at least in part from the user'"'"'s search history and one or more supplemental search results that are deemed similar to at least one of the user-preferred search results. Among the list of search results, the information server identifies one or more search results that are associated with the supplemental search results. Finally, the information server reorders the list of search results by moving each of the identified search results from its initial position by an offset and then provides the reordered list of search results to the user.

Claims

1. A computer-implemented method of producing personalized search results, performed on a server system having one or more processors and memory storing one or more programs for execution by the one or more processors to perform the method, comprising:

identifying a list of search results associated with a search query from a user, wherein each search result has an initial position in the list;

identifying a user profile of the user, wherein the user profile includes a set of user-disfavored search results that is determined from the user'"'"'s search history, wherein the user-disfavored search results include search results identified as user-disfavored search results in accordance with criteria for identifying a subset of search results that have been provided to the user, but not selected by the user;

identifying in the list of search results, one or more search results that are associated with at least one of the user-disfavored search results;

reordering the list of search results by moving each of the identified search results from its initial position by an offset; and

providing the reordered list of search results to the user.

2. The method of claim 1, wherein the set of user-disfavored search results is determined at real time.

3. The method of claim 1, wherein each of the identified search results is a member of the user-disfavored search results.

4. The method of claim 1, wherein each of the identified search results is associated with a source shared by at least one of the user-disfavored search results.

5. The method of claim 1, wherein the offset is a constant for the identified search results.

6. The method of claim 1, wherein the offset is a variable that is a function of an unpopularity metric associated with each of the identified search results.

7. A computer-implemented method of producing personalized search results, performed on a server system having one or more processors and memory storing one or more programs for execution by the one or more processors to perform the method, comprising:

identifying a list of search results associated with a search query from a user, wherein each search result has an initial position in the list;

identifying a user profile associated with the user, wherein the user profile includes a set of user-disfavored search results that is determined at least in part from the user'"'"'s search history, wherein the set of user-disfavored search results is determined by;

identifying a set of candidate search results, wherein each of the candidate search results has been provided to the user, but without being selected, for at least a predefined minimum number of times;

determining an unpopularity metric for each of the candidate search results; and

selecting a subset of the candidate search results whose associated unpopularity metrics exceed a predefined threshold as the set of user-disfavored search results;

identifying in the list of search results, one or more search results that are associated with at least one of the user-disfavored search results;

reordering the list of search results by moving each of the identified search results from its initial position by an offset; and

providing the reordered list of search results to the user.

8. The method of claim 7, wherein each of the candidate search results has been provided to the user, but without being selected, for at least the predefined minimum number of times during a time period having at least a predefined minimum duration.

9. The method of claim 7, wherein the unpopularity metric of a candidate search result is a function of one or more parameters including a number of times that the user unselects the candidate search result after being provided, a time span period from the user'"'"'s most remote unselecting of the candidate search result to the user'"'"'s most recent unselecting of the candidate search result, and a time decay period from the user'"'"'s most recent unselecting of the candidate search result to present.

10. The method of claim 7, wherein the unpopularity metric of a candidate search result is determined by:

determining a model predicting a likelihood of a user unselecting of a search result using multiple users'"'"' search histories; and

applying the model to the user'"'"'s search history to generate the unpopularity metric of the candidate search result.

11. The method of claim 10, wherein the model is determined by:

generating a profile that characterizes the multiple users'"'"' behaviors over multiple search results in the multiple users'"'"' search histories; and

determining as the model a set of coefficients from the profile, wherein the set of coefficients, when applied to a set of search results associated from a specific user'"'"'s search history, is configured to predict the likelihood of the specific user'"'"'s unselecting of each of the search results in the future.

12. A computer system for producing personalized search results, comprising:

memory;

one or more processors;

one or more programs stored in the memory and configured for execution by the one or more processors, the one or more programs including;

instructions for identifying a list of search results associated with a search query from a user, wherein each search result has an initial position in the list;

instructions for identifying a user profile of the user, further including instructions for determining a set of user-disfavored search results from the user'"'"'s search history, wherein the user-disfavored search results include search results identified as user-disfavored search results in accordance with criteria for identifying a subset of search results that have been provided to the user, but not selected by the user;

instructions for identifying in the list of search results, one or more search results that are associated with at least one of the user-disfavored search results;

instructions for reordering the list of search results by moving each of the identified search results from its initial position by an offset; and

instructions for providing the reordered list of search results to the user.

13. The computer system of claim 12, wherein the set of user-disfavored search results is determined at real time.

14. The computer system of claim 12, wherein each of the identified search results is a member of the user-disfavored search results.

15. The computer system of claim 12, wherein each of the identified search results is associated with a source shared by at least one of the user-disfavored search results.

16. The computer system of claim 12, wherein the offset is a constant for the identified search results.

17. The computer system of claim 12, wherein the offset is a variable that is a function of an unpopularity metric associated with each of the identified search results.

18. A computer system for producing personalized search results, comprising:

memory;

one or more processors;

one or more programs stored in the memory and configured for execution by the one or more processors, the one or more programs including;

instructions for identifying a list of search results associated with a search query from a user, wherein each search result has an initial position in the list;

instructions for identifying a user profile associated with the user, further including instructions for determining a set of user-disfavored search results at least in part from the user'"'"'s search history, wherein the instructions for determining the set of user-disfavored search results further include;

instructions for identifying a set of candidate search results, wherein each of the candidate search results has been provided to the user, but without being selected, for at least a predefined minimum number of times;

instructions for determining an unpopularity metric for each of the candidate search results; and

instructions for selecting a subset of the candidate search results whose associated unpopularity metrics exceed a predefined threshold as the set of user-disfavored search results;

instructions for identifying in the list of search results, one or more search results that are associated with at least one of the user-disfavored search results;

instructions for reordering the list of search results by moving each of the identified search results from its initial position by an offset; and

instructions for providing the reordered list of search results to the user.

19. The computer system of claim 18, wherein each of the candidate search results has been provided to the user, but without being selected, for at least the predefined minimum number of times during a time period having at least a predefined minimum duration.

20. The computer system of claim 18, wherein the unpopularity metric of a candidate search result is a function of one or more parameters including a number of times that the user unselects the candidate search result after being provided, a time span period from the user'"'"'s most remote unselecting of the candidate search result to the user'"'"'s most recent unselecting of the candidate search result, and a time decay period from the user'"'"'s most recent unselecting of the candidate search result to present.

21. The computer system of claim 18, wherein the instructions for determining the unpopularity metric of a candidate search result further include:

instructions for determining a model predicting a likelihood of a user unselecting of a search result using multiple users'"'"' search histories; and

instructions for applying the model to the user'"'"'s search history to generate the unpopularity metric of the candidate search result.

22. The computer system of claim 21, wherein the instructions for determining the model further include:

instructions for generating a profile that characterizes the multiple users'"'"' behaviors over multiple search results in the multiple users'"'"' search histories; and

instructions for determining as the model a set of coefficients from the profile, wherein the set of coefficients, when applied to a set of search results associated from a specific user'"'"'s search history, is configured to predict the likelihood of the specific user'"'"'s unselecting of each of the search results in the future.

23. A computer readable storage medium storing one or more programs to be executed by a computer system, the one or more programs comprising:

instructions for identifying a list of search results associated with a search query from a user, wherein each search result has an initial position in the list;

instructions for identifying a user profile of the user, further including instructions for determining a set of user-disfavored search results from the user'"'"'s search history, wherein the user-disfavored search results include search results identified as user-disfavored search results in accordance with criteria for identifying a subset of search results that have been provided to the user, but not selected by the user;

instructions for identifying in the list of search results, one or more search results that are associated with at least one of the user-disfavored search results;

instructions for reordering the list of search results by moving each of the identified search results from its initial position by an offset; and

instructions for providing the reordered list of search results to the user.

24. A computer readable storage medium storing one or more programs to be executed by a computer system, the one or more programs comprising:

instructions for identifying a list of search results associated with a search query from a user, wherein each search result has an initial position in the list;

instructions for identifying a user profile associated with the user, further including instructions for determining a set of user-disfavored search results at least in part from the user'"'"'s search history, wherein the instructions for determining the set of user-disfavored search results further include;

instructions for identifying a set of candidate search results, wherein each of the candidate search results has been provided to the user, but without being selected, for at least a predefined minimum number of times;

instructions for determining an unpopularity metric for each of the candidate search results; and

instructions for selecting a subset of the candidate search results whose associated unpopularity metrics exceed a predefined threshold as the set of user-disfavored search results;

instructions for identifying in the list of search results, one or more search results that are associated with at least one of the user-disfavored search results;

instructions for reordering the list of search results by moving each of the identified search results from its initial position by an offset; and

instructions for providing the reordered list of search results to the user.

25. The computer readable storage medium of claim 24, wherein each of the candidate search results has been provided to the user, but without being selected, for at least the predefined minimum number of times during a time period having at least a predefined minimum duration.

26. The computer readable storage medium of claim 24, wherein the unpopularity metric of a candidate search result is a function of one or more parameters including a number of times that the user unselects the candidate search result after being provided, a time span period from the user'"'"'s most remote unselecting of the candidate search result to the user'"'"'s most recent unselecting of the candidate search result, and a time decay period from the user'"'"'s most recent unselecting of the candidate search result to present.

27. The computer readable storage medium of claim 24, wherein the instructions for determining the unpopularity metric of a candidate search result further include:

instructions for determining a model predicting a likelihood of a user unselecting of a search result using multiple users'"'"' search histories; and

instructions for applying the model to the user'"'"'s search history to generate the unpopularity metric of the candidate search result.

28. The computer readable storage medium of claim 24, wherein the instructions for determining the model further include:

instructions for generating a profile that characterizes the multiple users'"'"' behaviors over multiple search results in the multiple users'"'"' search histories; and

instructions for determining as the model a set of coefficients from the profile, wherein the set of coefficients, when applied to a set of search results associated from a specific user'"'"'s search history, is configured to predict the likelihood of the specific user'"'"'s unselecting of each of the search results in the future.

29. A computer system for producing personalized search results, comprising:

memory;

one or more processors;

one or more programs stored in the memory and configured for execution by the one or more processors, the one or more programs including;

means for identifying a list of search results associated with a search query from a user, wherein each search result has an initial position in the list;

means for identifying a user profile of the user, further including means for determining a set of user-disfavored search results from the user'"'"'s search history, wherein the user-disfavored search results include search results identified as user-disfavored search results in accordance with criteria for identifying a subset of search results that have been provided to the user, but not selected by the user;

means for identifying in the list of search results, one or more search results that are associated with at least one of the user-disfavored search results;

means for reordering the list of search results by moving each of the identified search results from its initial position by an offset; and

means for providing the reordered list of search results to the user.