Removing substitution rules based on user interactions

• US 8,965,875 B1
• Filed: 04/10/2012
• Issued: 02/24/2015
• Est. Priority Date: 01/03/2012
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Popovici, Dan
• Primary Examiner(s): Pyo, Monica
• Application Number:
  US13/443,405
• Time in Patent Office: 1,050 Days
• Field of Search: 707/708, 707/766, 707/767, 707/769
• US Class Current: 707/708
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for removing substitution rules. According to one implementation, a method includes identifying a plurality of revised search queries associated with a substitution rule, wherein the substitution rule is used to generate the revised search queries from each of a plurality of original search queries to include a substitute query term for an original query term included in the original search queries; for each of the revised search queries, accessing query log data that identifies (i) a set of search results generated by the respective revised search query, and (ii) a particular, user-selected search result in the set of search results; determining a quality metric for the substitution rule using the query log data; determining that the quality metric does not satisfy a threshold; and in response to determining that the quality metric does not satisfy the threshold, removing the substitution rule.

Claims

1. A computer-implemented method comprising:

identifying a plurality of revised search queries associated with a substitution rule, wherein the substitution rule is used to generate the revised search queries from a plurality of original search queries such that each of the revised search queries includes a substitute query term for an original query term that is included in each of the original search queries;

for each of the revised search queries associated with the substitution rule, accessing query log data that identifies (i) a set of search results that was presented to a user in response to the respective revised search query and that includes one or more search results that were not selected by the user, and (ii) a particular, user-selected search result that was selected by the user from among the set of search results;

determining a quality metric for the substitution rule based at least on the query log data that identifies, for each of the revised search queries associated with the substitution rule, (i) the set of search results that was presented to the user in response to the respective revised search query and that includes one or more search results that were not selected by the user, and (ii) the particular, user-selected search result that was selected by the user from among the set of search results;

determining that the quality metric does not satisfy a threshold; and

in response to determining that the quality metric does not satisfy the threshold, removing the substitution rule.

2. The method of claim 1, wherein removing the substitution rule comprises removing the substitution rule from a collection of substitution rules designated to be used to generate revised search queries, such that the removed substitution rule is not used to generate revised search queries.

3. The method of claim 1, wherein removing the substitution rule comprises adding the substitution rule to a collection of substitution rules that are designated to not be used to generate revised search queries, such that the added substitution rule is not used to generate revised search queries.

4. The method of claim 1, wherein identifying the plurality of revised search queries associated with the substitution rule further comprises:

identifying a context associated with the substitution rule, wherein the context associated with the substitution rule specifies one or more query terms other than the original query term that should be present in an original search query for the substitution rule to be used to revise the original search query.

5. The method of claim 4, wherein accessing query log data comprises accessing query log data for original search queries that include the original query term in the identified context.

6. The method of claim 5, wherein removing the substitution rule comprises:

removing the substitution rule from a collection of substitution rules in the identified context that are used to revise search queries, wherein the collection of substitution rules in the identified context are used to generate revised search queries from subsequent original search queries associated with the identified context.

7. The method of claim 1, wherein the quality metric is based on a click count, a skip count, a both click count, a both skip count, a critical click count, and a critical click count, and wherein the quality metric is determined by an equation:

8. The method of claim 1, comprising determining, based on the query log data, a measure of non-selected search results that each (i) include the substitute query term for the original query term, (ii) do not include the original query term, and (iii) were ranked above the particular, user-selected search result in the set of search results that includes the non-selected search result;

wherein determining the quality metric for the substitution rule comprises determining the quality metric for the substitution rule based at least on the measure of non-selected search results.

9. The method of claim 1, comprising determining, based on the query log data, a measure of selected search results that each include (i) the substitute term for the original query term, and (ii) the original query term;

wherein determining the quality metric for the substitution rule comprises determining the quality metric for the substitution rule based at least on the measure of selected search results.

10. The method of claim 1, comprising determining, based on the query log data, a measure of selected search results that each (i) include the substitute query term for the original query term, (ii) do not include the original query term, and (iii) do not include any other substitute term for the original query term;

wherein determining the quality metric for the substitution rule comprises determining the quality metric for the substitution rule based at least on the measure of selected search results.

11. A non-transitory computer-readable medium storing software having stored thereon instructions, which, when executed by one or more computers, cause the one or more computers to perform operations of:

identifying a plurality of revised search queries associated with a substitution rule, wherein the substitution rule is used to generate the revised search queries from a plurality of original search queries such that each of the revised search queries includes a substitute query term for an original query term that is included in each of the original search queries;

for each of the revised search queries associated with the substitution rule, accessing query log data that identifies (i) a set of search results that was presented to a user in response to the respective revised search query and that includes one or more search results that were not selected by the user, and (ii) a particular, user-selected search result that was selected by the user from among the set of search results;

determining a quality metric for the substitution rule based at least on the query log data that identifies, for each of the revised search queries associated with the substitution rule, (i) the set of search results that was presented to the user in response to the respective revised search query and that includes one or more search results that were not selected by the user, and (ii) the particular, user-selected search result that was selected by the user from among the set of search results;

determining that the quality metric does not satisfy a threshold; and

in response to determining that the quality metric does not satisfy the threshold, removing the substitution rule.

12. The non-transitory computer-readable medium of claim 11, wherein removing the substitution rule comprises removing the substitution rule from a collection of substitution rules designated to be used to generate revised search queries, such that the removed substitution rule is not used to generate revised search queries.

13. The non-transitory computer-readable medium of claim 11, wherein removing the substitution rule comprises adding the substitution rule to a collection of substitution rules that are designated to not be used to generate revised search queries, such that the added substitution rule is not used to generate revised search queries.

14. The non-transitory computer-readable medium of claim 11, wherein identifying the plurality of revised search queries associated with the substitution rule comprises:

identifying a context associated with the substitution rule, wherein the context associated with the substitution rule specifies one or more query terms other than the original query term that should be present in an original search query for the substitution rule to be used to revise the original search query.

15. The non-transitory computer-readable medium of claim 14, wherein accessing query log data comprises accessing query log data for original search queries that include the original query term in the identified context.

16. The non-transitory computer-readable medium of claim 15, wherein removing the substitution rule comprises:

removing the substitution rule from a collection of substitution rules in the identified context that are used to revise search queries, wherein the collection of substitution rules in the identified context are used to generate revised search queries from subsequent original search queries associated with the identified context.

17. The non-transitory computer-readable medium of claim 11, wherein the quality metric is based on a click count, a skip count, a both click count, a both skip count, a critical click count, and a critical click count, and wherein the quality metric is determined by an equation:

18. 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;

identifying a plurality of revised search queries associated with a substitution rule, wherein the substitution rule is used to generate the revised search queries from a plurality of original search queries such that each of the revised search queries includes a substitute query term for an original query term that is included in each of the original search queries;

for each of the revised search queries associated with the substitution rule, accessing query log data that identifies (i) a set of search results that was presented to a user in response to the respective revised search query and that includes one or more search results that were not selected by the user, and (ii) a particular, user-selected search result that was selected by the user from among the set of search results;

determining a quality metric for the substitution rule based at least on the query log data that identifies, for each of the revised search queries associated with the substitution rule, (i) the set of search results that was presented to the user in response to the respective revised search query and that includes one or more search results that were not selected by the user, and (ii) the particular, user-selected search result that was selected by the user from among the set of search results;

determining that the quality metric does not satisfy a threshold; and

in response to determining that the quality metric does not satisfy the threshold, removing the substitution rule.

19. The system of claim 18, wherein removing the substitution rule comprises removing the substitution rule from a collection of substitution rules designated to be used to generate revised search queries, such that the removed substitution rule is not used to generate revised search queries.

20. The system of claim 18, wherein removing the substitution rule comprises adding the substitution rule to a collection of substitution rules that are designated to not be used to generate revised search queries, such that the added substitution rule is not used to generate revised search queries.

21. The system of claim 18, wherein identifying the plurality of revised search queries associated with the substitution rule comprises:

identifying a context associated with the substitution rule, wherein the context associated with the substitution rule specifies one or more query terms other than the original query term that should be present in an original search query for the substitution rule to be used to revise the original search query.

22. The system of claim 21, wherein accessing query log data comprises accessing query log data for original search queries that include the original query term in the identified context.

23. The system of claim 22, wherein removing the substitution rule comprises:

removing the substitution rule from a collection of substitution rules in the identified context that are used to revise search queries, wherein the collection of substitution rules in the identified context are used to generate revised search queries from subsequent original search queries associated with the identified context.

24. The system of claim 18, wherein the quality metric is based on a click count, a skip count, a both click count, a both skip count, a critical click count, and a critical click count, and wherein the quality metric is determined by an equation: