Determination of a desired repository

• US 7,584,177 B2
• Filed: 06/29/2005
• Issued: 09/01/2009
• Est. Priority Date: 06/29/2005
• Status: Expired due to Fees

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Angelo, Michael, Ginsberg, Jeremy, Tong, Simon, Braginsky, David
• Primary Examiner(s): Chace; Christian P.
• Assistant Examiner(s): BUCKINGHAM, KELLYE DEE
• Application Number:
  US11/169,285
• Publication Number:
  US 20070005568A1
• Time in Patent Office: 1,525 Days
• Field of Search: 707/3
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9538 Presentation of query results
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation

Abstract

A system receives a search query from a user and searches a group of repositories, based on the search query, to identify, for each of the repositories, a set of search results. The system also identifies one of the repositories based on a likelihood that the user desires information from the identified repository and presents the set of search results associated with the identifies repository.

Claims

1. A method performed by one or more devices, comprising:

storing, in one or more memory devices, log data associated with a plurality of searches of repositories based on search queries provided by a plurality of users;

generating rules for a model based on the log data, the model receiving information regarding a particular search query and a particular user as inputs and outputting a score reflecting a likelihood that a particular repository includes information responsive to the particular search query provided by the particular user;

receiving, by the one or more devices, a search query from a user via a client device;

searching, by the one or more devices, a plurality of repositories, based on the search query, to identify, for each of the plurality of repositories, a set of search results;

determining, for each of the plurality of repositories, a score reflecting a likelihood that the repository includes information that satisfies the search query by inputting information regarding the search query and information regarding the user into the model;

selecting, by the one or more devices, one of the plurality of repositories based on the determined score;

generating, by the one or more devices, a search result document based on the set of search results associated with the selected one of the plurality of repositories; and

providing, by the one or more devices, the search result document to the client device.

2. The method of claim 1, where generating the rules for the model further includes:

representing the log data as triples of data (u, q, r), where u refers to information regarding a user that provided a search query, q refers to information regarding the search query, and r refers to information regarding a repository from which search results were provided in response to the search query.

3. The method of claim 2, where the log data includes millions of the triples of data (u, q, r).

4. The method of claim 2, where generating the rules for the model further includes:

determining a label for each of the triples of data (u, q, r), where the label includes information regarding whether the repository r includes information that satisfies the search query q provided by the user u.

5. The method of claim 4, where generating the rules for the model further includes:

training the model based on the triples of data (u, q, r) and the associated labels.

6. The method of claim 1, where selecting one of the repositories includes:

selecting one of the plurality of repositories with a highest score among the scores of the plurality of repositories.

7. The method of claim 1, where generating the search result document includes:

selecting the sets of search results associated with two or more of the plurality of repositories, to include in the search result document, based on the scores for the two or more of the plurality of repositories.

8. The method of claim 7, where selecting the sets of search results associated with two or more of the plurality of repositories based on the scores includes:

arranging the sets of search results within the search result document based on the scores associated with the two or more of the plurality of repositories, and

presenting the search result document to the user.

9. The method of claim 8, where arranging the sets of search results within the search result document based on the scores associated with the two or more of the plurality of repositories includes:

placing the set of search results associated with a first one of the two or more of the plurality of repositories in a more prominent location within the search result document than the set of search results associated with a second one of the two or more of the plurality of repositories when the score associated with the first repository is higher than the score associated with the second repository.

10. The method of claim 8, where arranging the sets of search results within the search result document based on the scores associated with the two or more of the plurality of repositories includes:

providing only a link to the set of search results associated with at least one of the two or more of the plurality of repositories within the search result document.

11. The method of claim 1, further comprising:

selecting a group of repositories, of the plurality of repositories, to search based on the scores; and

where searching a plurality of repositories includes;

performing a search on only the group of repositories.

12. One or more computer devices, comprising:

means for storing log data associated with a plurality of searches of repositories based on search queries provided by a plurality of users;

means for generating rules for a model based on the log data, the model receiving information regarding a particular search query and a particular user as inputs and outputting a score reflecting a likelihood that a particular repository includes information responsive to the particular search query provided by the particular user;

means for receiving a search query from a user;

means for performing a search on a plurality of repositories, based on the search query, to identify, for each of the plurality of repositories, a set of search results;

means for determining a score for each of the plurality of repositories by inputting information regarding the search query and information regarding the user into the model, the score, for each of the repositories, reflecting a likelihood that the repository includes information that satisfies the search query from the user; and

means for providing one or more of the sets of search results, from one or more of the plurality of repositories, based on the scores.

13. The one or more computer devices of claim 12, further comprising:

means for selecting a group of repositories, of the plurality of repositories, to search based on the scores.

14. One or more computer devices, comprising:

one or more memory devices to store log data associated with a plurality of searches of repositories based on search queries provided by a plurality of users;

a model generating system to generate rules for a model based on the log data, the model receiving information regarding a particular search query and a particular user as inputs and outputting a score reflecting a likelihood that a particular repository includes information responsive to the particular search query provided by the particular user; and

a search engine system to;

receive a search query from a user,

determine a score, for each of a plurality of repositories, by inputting information regarding the search query and information regarding the user into the model, the score for one of the plurality of repositories reflecting a likelihood that the one repository includes information that satisfies the search query from the user,

perform a search on one or more of the plurality of repositories, based on the search query, to identify, for each of the one or more of the plurality of repositories, a set of search results, and

provide one or more of the sets of search results based on the scores.

15. The one or more computer devices of claim 14, where, when performing a search on one or more of the plurality of repositories, the search engine system is configured to:

identify a group of repositories, of the plurality of repositories, to search based on the scores, and

search only the group of repositories to identify, for each repository in the group of repositories, a set of search results.

16. The one or more computer devices of claim 14, where, when performing a search on one or more of the plurality of repositories, the search engine system is configured to:

search each of the plurality of repositories based on the search query.

17. The one or more computer devices of claim 14, where, when generating the rules for the model, the model generation system is further configured to:

represent the log data as triples of data (u, q, r), where u refers to information regarding a user that provides a search query, q refers to information regarding the search query, and r refers to information regarding a repository from which search results were provided in response to the search query.

18. The one or more computer devices of claim 17, where the log data includes millions of the triples of data (u, q, r).

19. The one or more computer devices of claim 17, where, when generating the rules for the model, the model generation system is configured to:

determine a label for each of the triples of data (u, q, r), where the label includes information regarding whether the repository r includes information that satisfies the search query q provided by the user u.

20. The one or more computer devices of claim 19, where when generating the rules for the model, the model generation system is configured to:

train the model based on the triples of data (u, q, r) and the associated labels.

21. The one or more computer devices of claim 14, where, when providing one or more of the sets of search results, the search engine system is configured to:

select one of the one or more of the plurality of repositories with a highest score among the scores for the one or more of the plurality of repositories, and

present only the set of search results associated with the selected one of the one or more of the plurality of repositories.

22. The one or more computer devices of claim 14, where, when providing one or more of the sets of search results, the search engine system is configured to:

arrange the one or more sets of search results within a search result document based on the scores associated with the one or more of the plurality of repositories, and

present the search result document to the user.

23. The one or more computer devices of claim 22, where, when arranging the one or more sets of search results within a search result document, the search engine system is configured to:

place the set of search results associated with a first one of the one or more of the plurality of repositories in a more prominent location within the search result document than the set of search results associated with a second one of the one or more of the plurality of repositories when the score associated with the first repository is higher than the score associated with the second repository.

24. The one or more computer devices of claim 22, where, when arranging the one or more sets of search results within a search result document, the search engine system is configured to:

provide only a link to the set of search results associated with at least one of the one or more of the plurality of repositories within the search result document.

31. The one or more computer devices of claim 14, where the information regarding the user includes one or more of:

an Internet Protocol address associated with the user, cookie information associated with the user, a language associated with the user, geographical information associated with the user, prior queries provided by the user, or a period of time that the user provided a query; and

where the information regarding the search query includes one or more query terms included in the search query.

25. One or more computer devices, comprising:

a model generation system to generate rules for a model that determines a score reflecting a likelihood that a repository includes information responsive to a particular search query from a particular user, the rules for the model being generating based on information regarding a plurality of prior searches of a plurality of repositories by a plurality of users using a plurality of search queries, the model receiving information regarding the particular search query and the particular user as inputs and outputting the score reflecting the likelihood that a particular repository includes information responsive to the particular search query from the particular user; and

a search engine system to;

receive a search query from a user,

determine a score, for each of the plurality of repositories, by inputting information regarding the user and information associated with the search query into the model,

perform a search of one or more of the plurality of repositories based on the scores, and

present search results from at least one of the one or more of the plurality of repositories based on the scores.

32. The one or more computer devices of claim 25, where the information regarding the user includes one or more of:

an Internet Protocol address associated with the user, cookie information associated with the user, a language associated with the user, geographical information associated with the user, prior queries provided by the user, or a period of time that the user provided a query; and

where the information regarding the search query includes one or more query terms included in the search query.

26. A method performed by one or more devices, comprising:

storing, in one or more memory devices, log data associated with a plurality of searches of repositories based on search queries provided by a plurality of users;

generating rules for a model based on the log data, the model receiving information regarding a particular search query and a particular user as inputs and outputting a score reflecting a likelihood that a particular repository includes information responsive to the particular search query provided by the particular user;

receiving a search query from a user via a client device;

determining, for each of a plurality of repositories, a score reflecting a likelihood that the repository includes information that satisfies the search query from the user by inputting information regarding the search query and information regarding the user into the model;

performing a search on at least one of the plurality of repositories, based on the search query and the determine scores, to identify, for each of the at least one of the plurality of repositories, a set of search results;

generating a search result document based on one or more of the sets of search results; and

providing the search result document to the client device.

27. The method of claim 26, where generating the search result document includes determining an order in which to present the one or more of the sets of search results based on the determined scores.

28. One or more computer devices, comprising:

one or more memory devices to store log data associated with a plurality of searches of repositories based on search queries provided by a plurality of users;

a model generation system to generate a model that determines a score reflecting a likelihood that a particular repository includes information responsive to a particular search query provided by a particular user, the model generation system being configured to;

represent the log data as triples of data (u, q, r), where u refers to information regarding a user that provided a search query, q refers to information regarding the search query, and r refers to information regarding a repository from which search results were provided in response to the search query,

determine a label for each of the triples of data (u, q, r), where the label includes information regarding whether the repository r includes information responsive to the search query q provided by the user u, and

train the model based on the triples of data (u, q, r), and the associated labels, where the model predicts whether a particular repository includes information responsive to a particular search query provided by a particular user; and

a search engine system to;

receive a search query from a user,

search a plurality of the repositories, based on the search query, to identify, for each of the plurality of repositories, a set of search results,

input information regarding the user and information regarding the search query into the model to obtain a score for each of the plurality of repositories, the score reflecting, for each of the plurality of repositories, a likelihood that the repository includes information responsive to the search query provided by the user,

identify one of the plurality of repositories based on the score, and

provide the set of search results associated with the identified repository without providing the set of search results associated with at least one other one of the plurality of repositories.

29. The one or more computer devices of claim 28, where, when providing the set of search results, the search engine system is configured to:

identify one or more of the plurality of repositories that has a highest score among the scores for the plurality of repositories, and

provide the set of search results associated with only the identified one or more of the plurality of repositories.

30. The one or more computer devices of claim 28, where, when providing the set of search results, the search engine system is configured to:

identify one or more of the plurality of repositories that has a score greater than a threshold, and

provide the set of search results associated with only the identified one or more of the plurality of repositories.