Click or skip evaluation of proximity rules

• US 9,146,966 B1
• Filed: 01/07/2013
• Issued: 09/29/2015
• Est. Priority Date: 10/04/2012
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Popovici, Dan
• Primary Examiner(s): Perveen, Rehana
• Assistant Examiner(s): Khong, Alexander
• Application Number:
  US13/735,541
• Time in Patent Office: 995 Days
• Field of Search: 707/723
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/3325 Reformulation based on results of preceding query
• CPC Class Codes: G06F 16/3331 Query processing

Abstract

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for click or skip evaluation of proximity rules. In one aspect, a method includes accessing query data that identifies, for a search query, a particular query term and other query terms included in the search query, search results, and a particular search result selected by a user. The method further includes determining, using the query data, that, (i) in text associated with the particular search result, the particular query term is separated from the other query terms by a minimum number of terms, and (ii) in text associated with a search result that was ranked higher than the particular search result, the particular query term is not separated from the other query terms by the minimum number of terms, then incrementing a click count for a query term proximity rule corresponding to the particular query term.

Claims

1. A computer-implemented method comprising:

accessing query log data that identifies, for a search query;

a particular query term and one or more other query terms that are included in the search query,

search results that were generated using the search query,

a ranking of the search results that were generated using the search query, and

a particular search result selected by a user from among the search results that were generated using the search query;

determining, using the query kg data, that, (i) in text associated with the particular search result selected by the user, the particular query term is separated from all of the one or more other query terms by at least a predetermined minimum number of terms, and (ii) in text associated with a search result that was ranked higher than the particular search result selected by the user, the particular query term is not separated from all of the one or more other query terms by at least the predetermined minimum number of terms; and

based on the determination, incrementing a click count for a query term proximity rule corresponding to the particular query term.

2. The method of claim 1, comprising assigning a score to the query term proximity rule based on the click count, a skip count, and a fake skip count.

3. The method of claim 2, wherein the score for the query term proximity rule is based on a ratio of (i) the click count to (ii) the click count, the skip count, and the fake skip count.

4. The method of claim 2, further comprising:

determining that the score for the query term proximity rule does not meet a threshold; and

removing the query term proximity rule from a set of query term proximity rules used to generate revised search queries based on determining that the score for the query term proximity rule does not meet the threshold.

5. The method of claim 2, associating a first weight with the click count, a second weight with the skip count, and a third weight associated with the fake skip count.

6. The method of claim 5, wherein the score for the query term proximity rule satisfies

7. The method of claim 1, wherein the query term proximity rule comprises a query term revision rule for revising search queries by separating the particular query term and the one or more other query terms by a particular number of words.

8. The method of claim 1, wherein the query term proximity rule comprises a query term scoring rule for scoring separation of the particular query term and the one or more other query terms that occur in resources that are identified in response to search queries.

33. The method of claim 1, comprising:

after incrementing the click count for the query term proximity rule, determining, based at least on the click count for the query term proximity rule, to revise a subsequently-received search query that includes the particular query term and one or more other query terms using the query term proximity rule; and

in response to determining to revise the subsequently-received search query, revising the subsequently-received search query using the query term proximity rule.

34. The method of claim 1, comprising:

after incrementing the click count for the query term proximity rule and based at least on the click count for the query term proximity rule, adjusting the rank of a search result included among search results that are generated using a subsequently-received search query that includes the particular query term and one or more other query terms.

9. A system comprising:

one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising;

accessing query log data that identifies, for a search query;

a particular query term and one or more other query terms that are included in the search query,

search results that were generated using the search query,

a ranking of the search results that were generated using the search query, and

a particular search result selected by a user from among the search results that were generated using the search query;

determining, using the query log data, that, (i) in text associated with the particular search result selected by the user, the particular query term is separated from all of the one or more other query terms by at least a predetermined minimum number of terms, and (ii) in text associated with a search result that was ranked higher than the particular search result selected by the user, the particular query term is not separated from all of the one or more other query terms by at least the predetermined minimum number of terms; and

based on the determination, incrementing a click count for a query term proximity rule corresponding to the particular query term.

10. The system of claim 9, the operations further comprising assigning a score to the query term proximity rule based on the click count, a skip count, and a fake skip count.

11. The system of claim 10, wherein the score for the query term proximity rule is based on a ratio of (i) the click count to (ii) the click count, the skip count, and the fake skip count.

12. The system of claim 10, the operations further comprising:

determining that the score for the query term proximity rule does not meet a threshold; and

removing the query term proximity rule from a set of query term proximity rules used to generate revised search queries based on determining that the score for the query term proximity rule does not meet the threshold.

13. The system of claim 10, associating a first weight with the click count, a second weight with the skip count, and a third weight with the fake skip count.

14. The system of claim 13, wherein the score for the query term proximity rule satisfies

15. The system of claim 9, wherein the query term proximity rule comprises a query term revision rule for revising search queries by separating the particular query term and the one or more other query terms by a particular number of words.

16. The system of claim 9, wherein the query term proximity rule comprises a query term scoring rule for scoring separation of the particular query term and the one or more other query terms that occur in resources that are identified in response to search queries.

17. A non-transitory computer-readable medium storing software comprising instructions executable by one or more computers which, upon such execution, cause the one or more computers to perform operations comprising:

accessing query log data that identifies, for a search query;

a particular query term and one or more other query terms that are included in the search query,

search results that were generated using the search query,

a ranking of the search result that were generated using the search query, and

a particular search result selected by a user from among the search results that were generated using the search query;

determining, using the query kg data, that, (i) in text associated with the particular search result selected by the user, the particular query term is separated from all of the one or more other query terms by at least a predetermined minimum number of terms, and (ii) in text associated with a search result that was ranked higher than the particular search result, the particular query term is not separated from all of the one or more other query terms by at least the predetermined minimum number of terms; and

based on the determination, incrementing a click count for a query term proximity rule corresponding to the particular query term.

18. The medium of claim 17, the operations further comprising assigning a score to the query term proximity rule based on the click count, a skip count, and a fake skip count.

19. The medium of claim 18, wherein the score for the query term proximity rule is based on a ratio of (i) the click count to (ii) the click count, the skip count, and the fake skip count.

20. The medium of claim 18, the operations further comprising:

determining that the score for the query term proximity rule does not meet a threshold; and

removing the query term proximity rule from a set of query term proximity rules used to generate revised search queries based on determining that the score for the query term proximity rule does not meet the threshold.

21. The medium of claim 18, associating a first weight with the click count, a second weight with the skip count, and a third weight with the fake skip count.

22. The medium of claim 21, wherein the score for the query term proximity rule satisfies

23. The medium of claim 17, wherein the query term proximity rule comprises a query term revision rule for revising search queries by separating the particular query term and the one or more other query terms by a particular number of words.

24. The medium of claim 17, wherein the query term proximity rule comprises a query term scoring rule for scoring separation of the particular query term and the one or more other query terms that occur in resources that are identified in response to search queries.

25. A computer-implemented method comprising:

accessing query data that identifies, for a search query;

a particular query term and one or more other query terms that are included in the search query,

search results that were generated using the search query,

a ranking of the search results that were generated using the search query, and

a particular search result selected by a user from among the search results that were generated using the search query;

determining, using the query log data, that, (i) in text associated with the particular search result selected by the user, the particular query term is not separated from all of the one or more other query terms by at least a predetermined minimum number of terms, and (ii) in text associated with a search result that is ranked higher than the particular search result selected by the user, the particular query term is separated from all of the one or more other query terms by at least the predetermined minimum number of terms; and

based on the determination, incrementing a skip count for a query term proximity rule corresponding to the particular query term.

26. The method of claim 25, comprising assigning a score to the query term proximity rule based on a click count, the skip count, and a fake skip count.

27. The method of claim 26, wherein the score for the query term proximity rule is based on a ratio of (i) the click count to (ii) the click count, the skip count, and the fake skip count.

28. The method of claim 26, further comprising:

determining that the score for the query term proximity rule does not meet a threshold; and

removing the query term proximity rule from a set of query term proximity rules used to generate revised search queries based on determining that the score for the query term proximity rule does not meet the threshold.

29. The method of claim 26, associating a first weight with the click count, a second weight with the skip count, and a third weight with the fake skip count.

30. The method of claim 29, wherein the score for the query term proximity rule satisfies

31. The method of claim 25, wherein the query term proximity rule comprises a query term revision rule for revising search queries by separating the particular query term and the one or more other query terms by a particular number of words.

32. The method of claim 25, wherein the query term proximity rule comprises a query term scoring rule for scoring separation of the particular query term and the one or more other query terms that occur in resources that are identified in response to search queries.