Adjusting stored query relevance data based on query term similarity

• US 9,400,843 B2
• Filed: 12/04/2007
• Issued: 07/26/2016
• Est. Priority Date: 12/04/2007
• Status: Active Grant

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Brady, Colin
• Primary Examiner(s): ADAMS, CHARLES D
• Application Number:
  US11/950,016
• Publication Number:
  US 20090144263A1
• Time in Patent Office: 3,157 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

Techniques are described to improve search results using a panel. A search engine deploys one or more network traffic monitors. Traffic monitors analyze network traffic and find HTTP requests made to search engines. When a search query is spotted, the traffic monitor records the sequence of user requests, including search engine, search terms, and sites visited. A sequence of queries where a user visits one search engine, enters a query, visits zero or more sites from the results listings, and visits a second search engine, enters a query and visits one or more websites and stops searching is used to determine whether to increase or decrease a relevance value between a search term and the sites visited.

Claims

1. A method comprising:

retrieving first search results for a first search performed by a user via a first search engine using a first set of one or more search terms, wherein the first search results include a first plurality of items that does not include one or more particular items and the user selects zero or more items from the first plurality of items in the first search results,

retrieving second search results for a second search performed by the user via a second search engine using a second set of one or more search terms, wherein the second search results include a second plurality of items that includes the one or more particular items and the user selects the one or more particular items from the second search results and wherein the second search engine is different than the first search engine;

determining a correlation value that indicates similarity between the first set of one or more search terms used for the first search performed by the user via the first search engine and the second set of one or more search terms used for the second search performed by the user via the second search engine;

determining whether the correlation value that indicates similarity between the first set of one or more search terms used for the first search performed by the user via the first search engine and the second set of one or more search terms used for the second search performed by the user via the second search engine satisfies a specified threshold;

in response to determining that the correlation value that indicates similarity between the first set of one or more search terms used for the first search performed by the user via the first search engine and the second set of one or more search terms used for the second search performed by the user via the second search engine satisfies the specified threshold and to the user selection of the one or more particular items from the second search results, increasing a likelihood that the one or more particular items selected by the user from the second search results appear in search results for subsequent searches performed via the first search engine using the first set of one or more search terms by decreasing, at the first search engine, a relevance value between one or more other items from the first plurality of items in the first search results and the first set of one or more search terms, wherein the one or more other items from the first plurality of items in the first search results are different than the one or more particular items;

determining whether the correlation value that indicates similarity between the first set of one or more search terms used for the first search performed by the user via the first search engine and the second set of one or more search terms used for the second search performed by the user via the second search engine is below the specified threshold and above another threshold; and

upon determining that the correlation value that indicates similarity between the first set of one or more search terms used for the first search performed by the user via the first search engine and the second set of one or more search terms used for the second search performed by the user via the second search engine is below the specified threshold and above the other threshold;

causing a third search to be performed via the first search engine using the second set of one or more search terms that were used to perform the second search on the second search engine to obtain third search results,

determining a similarity of the third search results to the first search results by comparing the third search results to the first search results, and

in response to the determining that the third search results are similar to the first search results, then increasing a relevance value between the second set of one or more search terms and the one or more particular items selected by the user from the second search results;

wherein the method is performed by one or more computing devices.

2. The method of claim 1, further comprising:

upon determining that the correlation value that indicates similarity between the first set of one or more search terms used for the first search performed by the user via the first search engine and the second set of one or more search terms used for the second search performed by the user via the second search engine is below the specified threshold and above the other threshold, marking for review by a human operator one or more of the determined correlation value that indicates similarity between the first set of one or more search terms used for the first search performed by the user via the first search engine and the second set of one or more search terms used for the second search performed by the user via the second search engine or the relevance value between the one or more other items from the first plurality of items in the first search results and the first set of one or more search terms.

3. The method of claim 1, wherein the first search results and the second search results are derived from one or more of network logs from a traffic monitor or web traffic recorded by a widget in a toolbar.

4. The method of claim 1, wherein if the user selection of zero or more items from the first plurality of items in the first search results is a selection of the one or more other items from the first plurality of items in the first search results, then decreasing, at the first search engine, a relevance value between one or more other items from the first plurality of items in the first search results and the first set of one or more search terms includes decreasing, at the first search engine, a relevance value between the one or more other items from the first plurality of items in the first search results indicated by the selection of the one or more other items and the first set of one or more search terms.

5. The method of claim 1,

further comprising in response to determining that the correlation value that indicates similarity between the first set of one or more search terms used for the first search performed by the user via the first search engine and the second set of one or more search terms used for the second search performed by the user via the second search engine is above the specified threshold, increasing the relevance value between the one or more particular items selected by the user and the second set of one or more search terms.

6. The method of claim 1, wherein the specified threshold is a value such that the correlation value that indicates similarity between the first set of one or more search terms used for the first search performed by the user via the first search engine and the second set of one or more search terms used for the second search performed by the user via the second search engine indicates that the first and second search terms are identical.

7. The method of claim 1, wherein the relevance value between one or more other items from the first plurality of items in the first search results and the first set of one or more search terms is maintained by the first search engine.

8. The method of claim 1, wherein the determining a correlation value that indicates similarity between the first set of one or more search terms used for the first search performed by the user via the first search engine and the second set of one or more search terms used for the second search performed by the user via the second search engine includes:

basing the correlation value on values within a database of correlation values between search terms; and

comparing the first set of one or more search terms to the second set of one or more search terms.

9. The method of claim 1, wherein decreasing, at the first search engine, the relevance value between the one or more other items from the first plurality of items in the first search results and the first set of one or more search terms includes decreasing, at the first search engine, the relevance value between the one or more other items from the first plurality of items in the first search results and the first set of one or more search terms by an extent that is based at least in part on a quantity of service failures that occurred relative to searches that the first search engine performed based on the first set of one or more search terms.

10. One or more non-transitory computer-readable storage media storing instructions which, when processed by one or more processors, causes:

retrieving first search results for a first search performed by a user via a first search engine using a first set of one or more search terms, wherein the first search results include a first plurality of items that does not include one or more particular items and the user selects zero or more items from the first plurality of items in the first search results,

retrieving second search results for a second search performed by the user via a second search engine using a second set of one or more search terms, wherein the second search results include a second plurality of items that includes the one or more particular items and the user selects the one or more particular items from the second search results and wherein the second search engine is different than the first search engine;

determining a correlation value that indicates similarity between the first set of one or more search terms used for the first search performed by the user via the first search engine and the second set of one or more search terms used for the second search performed by the user via the second search engine;

determining whether the correlation value that indicates similarity between the first set of one or more search terms used for the first search performed by the user via the first search engine and the second set of one or more search terms used for the second search performed by the user via the second search engine satisfies a specified threshold; and

in response to determining that the correlation value that indicates similarity between the first set of one or more search terms used for the first search performed by the user via the first search engine and the second set of one or more search terms used for the second search performed by the user via the second search engine satisfies the specified threshold and to the user selection of the one or more particular items from the second search results, increasing a likelihood that the one or more particular items selected by the user from the second search results appear in search results for subsequent searches performed via the first search engine using the first set of one or more search terms by decreasing, at the first search engine, a relevance value between one or more other items from the first plurality of items in the first search results and the first set of one or more search terms, wherein the one or more other items from the first plurality of items in the first search results are different than the one or more particular items;

determining whether the correlation value that indicates similarity between the first set of one or more search terms used for the first search performed by the user via the first search engine and the second set of one or more search terms used for the second search performed by the user via the second search engine is below the specified threshold and above another threshold; and

upon determining that the correlation value that indicates similarity between the first set of one or more search terms used for the first search performed by the user via the first search engine and the second set of one or more search terms used for the second search performed by the user via the second search engine is below the specified threshold and above the other threshold;

causing a third search to be performed via the first search engine using the second set of one or more search terms that were used to perform the second search on the second search engine to obtain third search results,

determining a similarity of the third search results to the first search results by comparing the third search results to the first search results, and

in response to the determining that the third search results are similar to the first search results, then increasing a relevance value between the second set of one or more search terms and the one or more particular items selected by the user from the second search results.

11. The one or more non-transitory computer-readable media of claim 10, further comprising additional instructions which, when processed by the one or more processors, cause:

upon determining that the correlation value that indicates similarity between the first set of one or more search terms used for the first search performed by the user via the first search engine and the second set of one or more search terms used for the second search performed by the user via the second search engine is below the specified threshold and above the other threshold, marking for review by a human operator one or more of the determined correlation value that indicates similarity between the first set of one or more search terms used for the first search performed by the user via the first search engine and the second set of one or more search terms used for the second search performed by the user via the second search engine or the relevance value between the one or more other items from the first plurality of items in the first search results and the first set of one or more search terms.

12. The one or more non-transitory computer-readable media of claim 10, wherein the first search results and the second search results are derived from one or more of network logs from a traffic monitor or web traffic recorded by a widget in a toolbar.

13. The one or more non-transitory computer-readable media of claim 10, wherein if the user selection of zero or more items from the first plurality of items in the first search results is a selection of the one or more other items from the first plurality of items in the first search results, then decreasing, at the first search engine, a relevance value between one or more other items from the first plurality of items in the first search results and the first set of one or more search terms includes decreasing, at the first search engine, a relevance value between the one or more other items from the first plurality of items in the first search results indicated by the selection of the one or more other items and the first set of one or more search terms.

14. The one or more non-transitory computer-readable media of claim 10,

the one or more non-transitory computer-readable media further comprise additional instructions which, when processed by the one or more processors, cause in response to determining that the correlation value that indicates similarity between the first set of one or more search terms used for the first search performed by the user via the first search engine and the second set of one or more search terms used for the second search performed by the user via the second search engine is above the specified threshold, increasing the relevance value between the one or more particular items selected by the user and the second set of one or more search terms.

15. The one or more non-transitory computer-readable media of claim 10, wherein the specified threshold is a value such that the correlation value that indicates similarity between the first set of one or more search terms used for the first search performed by the user via the first search engine and the second set of one or more search terms used for the second search performed by the user via the second search engine indicates that the first and second search terms are identical.

16. The one or more non-transitory computer-readable media of claim 10, wherein the relevance value between one or more other items from the first plurality of items in the first search results and the first set of one or more search terms is maintained by the first search engine.

17. The one or more non-transitory computer-readable media of claim 10, wherein the determining a correlation value that indicates similarity between the first set of one or more search terms used for the first search performed by the user via the first search engine and the second set of one or more search terms used for the second search performed by the user via the second search engine includes:

basing the correlation value on values within a database of correlation values between search terms; and

comparing the first set of one or more search terms to the second set of one or more search terms.

18. The one or more non-transitory computer-readable storage media of claim 10, decreasing, at the first search engine, the relevance value between the one or more other items from the first plurality of items in the first search results and the first set of one or more search terms includes decreasing, at the first search engine, the relevance value between the one or more other items from the first plurality of items the first search results and the first set of one or more search terms by an extent that is based at least in part on a quantity of service failures that occurred relative to searches that the first search engine performed based on the first set of one or more search terms.