Personalized search result ranking

• US 8,762,373 B1
• Filed: 09/14/2012
• Issued: 06/24/2014
• Est. Priority Date: 09/29/2006
• Status: Expired due to Fees

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Zamir, Oren, Korn, Jeffrey
• Primary Examiner(s): BURKE, JEFF A
• Application Number:
  US13/617,209
• Time in Patent Office: 648 Days
• Field of Search: 707722-723, 707/732, 707/734
• US Class Current: 707/732
• CPC Class Codes: G06F 16/335 Filtering based on additional data, e.g. user or group profiles filtering in web context G06F16/9535, G06F16/9536
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9538 Presentation of query results

Abstract

A system determines a user'"'"'s past search result selection activity. The system adjusts the ranking of current search results provided to the user based on the user'"'"'s past search result selection activity.

Claims

1. A method comprising:

providing, by one or more computer devices, first search results based on a first query;

determining, by the one or more computer devices, a quantity of accesses, by a particular user, of a first document associated with a web site,

the first document being associated with a search result of the first search results;

comparing, by the one or more computer devices, the quantity of accesses to a threshold quantity of accesses;

determining, by the one or more computer devices and based on the comparison, that the quantity of accesses satisfies the threshold quantity of accesses;

identifying, by the one or more computer devices, second search results corresponding to documents that are relevant to a second query associated with the particular user,

the second search results including a particular search result corresponding to a second document associated with the web site, and

the second search results being ranked in a ranked list;

adjusting, by the one or more computer devices, a rank position of the particular search result in the ranked list to form adjusted search results, based on;

determining that the quantity of accesses satisfies the threshold quantity of accesses;

identifying another quantity of accesses, by the particular user, associated with the first search results,

the other quantity of accesses being unassociated with the first document; and

using at least one of;

an order in which the other quantity of accesses occurred relative to the quantity of accesses associated with the first document, or

a comparison of a first amount of time, spent by the particular user, when accessing the first document, and a second amount of time, spent by the particular user, when accessing documents associated with the other quantity of accesses; and

providing, by the one or more computer devices, the adjusted search results for display to the particular user.

2. The method of claim 1, where adjusting the rank position of the particular search result further includes:

identifying an amount of time spent accessing the first document by the particular user;

determining whether the amount of time satisfies a threshold time; and

adjusting the rank position of the particular search result, in the ranked list, further based on whether the amount of time satisfies the threshold time.

3. The method of claim 1, where the quantity of accesses of the first document are associated with selections, by the particular user and associated with the first document, from sets of search results.

4. The method of claim 3, where adjusting the rank position of the particular search result further includes:

determining respective similarities between the second query and a set of queries associated with the sets of search results; and

adjusting the rank position of the particular search result, in the ranked list, further based on the respective similarities between the second query and the set of queries.

5. The method of claim 3, where adjusting the rank position of the particular search result further includes:

determining respective particular rank positions, in the sets of search results, associated with the first document; and

adjusting the rank position of the particular search result, in the ranked list, further based on the respective particular rank positions.

6. The method of claim 1, where adjusting the rank position of the particular search result further includes:

identifying how often the first document is included in sets of search results, presented to the particular user, and not selected by the particular user; and

adjusting the rank position of the particular search result, in the ranked list, further based on how often the first document is included in the sets of search results and not selected by the particular user.

7. The method of claim 1, where adjusting the rank position of the particular search result further includes:

determining time information associated with the quantity of accesses of the first document, the time information including one or more of;

respective durations of time spent accessing the first document by the particular user,

an access pattern associated with the quantity of accesses of the first document,

a recency of a last access of the quantity of accesses of the first document, or

an amount of time between a first access, of the quantity of accesses of the first document, and the last access; and

adjusting the rank position of the particular search result further based on the time information.

8. The method of claim 1, where the first document is the second document.

9. The method of claim 1, where adjusting the rank position of the particular search result in the ranked list includes:

adjusting a rank position of the particular search result based on the quantity of accesses of the first document,

the first document being different than the second document.

10. A system comprising:

one or more processors associated with one or more devices, the one or more processors being configured to;

provide first search results based on a first query;

determine a quantity of accesses, by a particular user, of a first document associated with a web site,

the first document being associated with a search result of the first search results;

compare the quantity of accesses to a threshold quantity of accesses;

determine, based on the comparison, that the quantity of accesses satisfies the threshold quantity of accesses;

identify second search results corresponding to documents that are relevant to a second query associated with the particular user,

the second search results including a particular search result corresponding to a second document associated with the web site, and

the second search results being ranked in a ranked list;

adjust a rank position of the particular search result in the ranked list ranked list to form adjusted search results, based on;

determining that the quantity of accesses satisfies the threshold quantity of accesses, to form adjusted search results;

identifying another quantity of accesses, by the particular user, associated with the first search results,

the other quantity of accesses being unassociated with the first document; and

using at least one of;

an order in which the other quantity of accesses occurred relative to the quantity of accesses associated with the first document, or

a comparison of a first amount of time, spent by the particular user, when accessing the first document, and a second amount of time, spent by the particular user, when accessing documents associated with the other quantity of accesses; and

provide the adjusted search results for display to the particular user.

11. The system of claim 10, where the one or more processors, when adjusting the rank position of the particular search result, are further configured to:

identify an amount of time spent accessing the first document by the particular user;

determine whether the amount of time satisfies a threshold time; and

adjust the rank position of the particular search result, in the ranked list, further based on whether the amount of time satisfies the threshold time.

12. The system of claim 10, where the quantity of accesses of the first document are associated with selections, by the particular user and associated with the first document, from sets of search results.

13. The system of claim 12, where the one or more processors, when adjusting the rank position of the particular search result, are further configured to:

determine respective similarities between the second query and a set of queries associated with the sets of search results; and

adjust the rank position of the particular search result, in the ranked list, further based on the respective similarities between the second query and the set of queries.

14. The system of claim 12, where the one or more processors, when adjusting the rank position of the particular search result, are further configured to:

determine respective particular rank positions, in the sets of search results, associated with the first document; and

adjust the rank position of the particular search result, in the ranked list, further based on the respective particular rank positions.

15. The system of claim 10, where the one or more processors, when adjusting the rank position of the particular search result, are further configured to:

identify how often the first document is included in sets of search results, presented to the particular user, and not selected by the particular user; and

adjust the rank position of the particular search result, in the ranked list, further based on how often the first document is included in the sets of search results and not selected by the particular user.

16. The system of claim 10, where the one or more processors, when adjusting the rank position of the particular search result, are further configured to:

determine time information associated with the quantity of accesses of the first document, the time information including one or more of;

respective durations of time spent accessing the first document by the particular user,

an access pattern associated with the quantity of accesses of the first document,

a recency of a last access of the quantity of accesses of the first document, or

an amount of time between a first access, of the quantity of accesses of the first document, and the last access; and

adjust the rank position of the particular search result further based on the time information.

17. The system of claim 10, where the one or more processors, when adjusting the rank position of the particular search result, are further configured to:

adjust a rank position of the particular search result based on the quantity of accesses of the first document,

the first document being different than the second document.

18. A non-transitory memory storage device to store instructions, the instructions comprising:

one or more instructions that, when executed by one or more processors associated with one or more devices, cause the one or more processors to;

provide first search results based on a first query;

determine a quantity of accesses, by a particular user, of a first document associated with a web site,

the first document being associated with a search result of the first search results;

compare the quantity of accesses to a threshold quantity of accesses;

determine, based on the comparison, that the quantity of accesses satisfies the threshold quantity of accesses;

identify second search results corresponding to documents that are relevant to a second query associated with the particular user,

the second search results including a particular search result corresponding to a second document associated with the web site, and

the second search results being ranked in a ranked list;

adjust a rank position of the particular search result in the ranked list to form adjusted search results, based on;

determining that the quantity of accesses satisfies the threshold quantity of accesses;

identifying another quantity of accesses, by the particular user, associated with the first search results,

the other quantity of accesses being unassociated with the first document; and

using at least one of;

an order in which the other quantity of accesses occurred relative to the quantity of accesses associated with the first document, or

a comparison of a first amount of time, spent by the particular user, when accessing the first document, and a second amount of time, spent by the particular user, when accessing documents associated with the other quantity of accesses; and

provide the adjusted search results for display to the particular user.

19. The non-transitory memory storage device of claim 18, where the one or more instructions to adjust the rank position of the particular search result include:

one or more instructions to identify an amount of time spent accessing the first document by the particular user;

one or more instructions to determine whether the amount of time satisfies a threshold time; and

one or more instructions to adjust the rank position of the particular search result, in the ranked list, further based on whether the amount of time satisfies the threshold time.

20. The non-transitory memory storage device of claim 18, where the one or more instructions to adjust the rank position of the particular search result include:

one or more instructions to determine respective similarities between the second query and a set of queries associated with sets of search results; and

one or more instructions to adjust the rank position of the particular search result, in the ranked list, further based on the respective similarities between the second query and the set of queries.